2022/09/06 17:19:30 - mmengine - INFO - ------------------------------------------------------------ System environment: sys.platform: linux Python: 3.7.13 (default, Mar 29 2022, 02:18:16) [GCC 7.5.0] CUDA available: True numpy_random_seed: 429443971 GPU 0,1,2,3,4,5,6,7: NVIDIA A100-SXM4-80GB CUDA_HOME: /mnt/cache/share/cuda-11.1 NVCC: Cuda compilation tools, release 11.1, V11.1.74 GCC: gcc (GCC) 5.4.0 PyTorch: 1.11.0 PyTorch compiling details: PyTorch built with: - GCC 7.3 - C++ Version: 201402 - Intel(R) Math Kernel Library Version 2020.0.2 Product Build 20200624 for Intel(R) 64 architecture applications - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) - OpenMP 201511 (a.k.a. OpenMP 4.5) - LAPACK is enabled (usually provided by MKL) - NNPACK is enabled - CPU capability usage: AVX2 - CUDA Runtime 11.3 - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 - CuDNN 8.2 - Magma 2.5.2 - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, TorchVision: 0.12.0 OpenCV: 4.5.5 MMEngine: 0.1.0 Runtime environment: cudnn_benchmark: False mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} dist_cfg: {'backend': 'nccl'} seed: None Distributed launcher: slurm Distributed training: True GPU number: 8 ------------------------------------------------------------ 2022/09/06 17:19:30 - mmengine - INFO - Config: model = dict( type='Recognizer2D', data_preprocessor=dict( type='ActionDataPreprocessor', mean=[123.675, 116.28, 103.5], std=[58.395, 57.12, 57.375], format_shape='NCHW'), backbone=dict( type='TANet', pretrained='torchvision://resnet50', depth=50, num_segments=8, tam_cfg=None), cls_head=dict( type='TSMHead', num_classes=400, in_channels=2048, spatial_type='avg', consensus=dict(type='AvgConsensus', dim=1), dropout_ratio=0.5, init_std=0.001, average_clips='prob')) train_cfg = dict( type='EpochBasedTrainLoop', max_epochs=100, val_begin=1, val_interval=1) val_cfg = dict(type='ValLoop') test_cfg = dict(type='TestLoop') param_scheduler = [ dict( type='MultiStepLR', begin=0, end=100, by_epoch=True, milestones=[50, 75, 90], gamma=0.1) ] optim_wrapper = dict( constructor='TSMOptimWrapperConstructor', paramwise_cfg=dict(fc_lr5=True), optimizer=dict(type='SGD', lr=0.01, momentum=0.9, weight_decay=0.0001), clip_grad=dict(max_norm=20, norm_type=2)) default_scope = 'mmaction' default_hooks = dict( runtime_info=dict(type='RuntimeInfoHook'), timer=dict(type='IterTimerHook'), logger=dict(type='LoggerHook', interval=100, ignore_last=False), param_scheduler=dict(type='ParamSchedulerHook'), checkpoint=dict( type='CheckpointHook', interval=1, save_best='auto', max_keep_ckpts=5), sampler_seed=dict(type='DistSamplerSeedHook')) env_cfg = dict( cudnn_benchmark=False, mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), dist_cfg=dict(backend='nccl')) log_processor = dict(type='LogProcessor', window_size=20, by_epoch=True) vis_backends = [dict(type='LocalVisBackend')] visualizer = dict( type='ActionVisualizer', vis_backends=[dict(type='LocalVisBackend')]) log_level = 'INFO' load_from = None resume = False file_client_args = dict( io_backend='petrel', path_mapping=dict({ 'data/kinetics400': 's254:s3://openmmlab/datasets/action/Kinetics400' })) dataset_type = 'VideoDataset' data_root = 'data/kinetics400/videos_train' data_root_val = 'data/kinetics400/videos_val' ann_file_train = 'data/kinetics400/kinetics400_train_list_videos.txt' ann_file_val = 'data/kinetics400/kinetics400_val_list_videos.txt' ann_file_test = 'data/kinetics400/kinetics400_val_list_videos.txt' train_pipeline = [ dict( type='DecordInit', io_backend='petrel', path_mapping=dict({ 'data/kinetics400': 's254:s3://openmmlab/datasets/action/Kinetics400' })), dict(type='DenseSampleFrames', clip_len=1, frame_interval=1, num_clips=8), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict( type='MultiScaleCrop', input_size=224, scales=(1, 0.875, 0.75, 0.66), random_crop=False, max_wh_scale_gap=1, num_fixed_crops=13), dict(type='Resize', scale=(224, 224), keep_ratio=False), dict(type='Flip', flip_ratio=0.5), dict(type='FormatShape', input_format='NCHW'), dict(type='PackActionInputs') ] val_pipeline = [ dict( type='DecordInit', io_backend='petrel', path_mapping=dict({ 'data/kinetics400': 's254:s3://openmmlab/datasets/action/Kinetics400' })), dict( type='DenseSampleFrames', clip_len=1, frame_interval=1, num_clips=8, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict(type='CenterCrop', crop_size=224), dict(type='FormatShape', input_format='NCHW'), dict(type='PackActionInputs') ] test_pipeline = [ dict( type='DecordInit', io_backend='petrel', path_mapping=dict({ 'data/kinetics400': 's254:s3://openmmlab/datasets/action/Kinetics400' })), dict( type='DenseSampleFrames', clip_len=1, frame_interval=1, num_clips=8, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict(type='CenterCrop', crop_size=224), dict(type='FormatShape', input_format='NCHW'), dict(type='PackActionInputs') ] train_dataloader = dict( batch_size=8, num_workers=8, persistent_workers=True, sampler=dict(type='DefaultSampler', shuffle=True), dataset=dict( type='VideoDataset', ann_file='data/kinetics400/kinetics400_train_list_videos.txt', data_prefix=dict(video='data/kinetics400/videos_train'), pipeline=[ dict( type='DecordInit', io_backend='petrel', path_mapping=dict({ 'data/kinetics400': 's254:s3://openmmlab/datasets/action/Kinetics400' })), dict( type='DenseSampleFrames', clip_len=1, frame_interval=1, num_clips=8), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict( type='MultiScaleCrop', input_size=224, scales=(1, 0.875, 0.75, 0.66), random_crop=False, max_wh_scale_gap=1, num_fixed_crops=13), dict(type='Resize', scale=(224, 224), keep_ratio=False), dict(type='Flip', flip_ratio=0.5), dict(type='FormatShape', input_format='NCHW'), dict(type='PackActionInputs') ])) val_dataloader = dict( batch_size=8, num_workers=8, persistent_workers=True, sampler=dict(type='DefaultSampler', shuffle=False), dataset=dict( type='VideoDataset', ann_file='data/kinetics400/kinetics400_val_list_videos.txt', data_prefix=dict(video='data/kinetics400/videos_val'), pipeline=[ dict( type='DecordInit', io_backend='petrel', path_mapping=dict({ 'data/kinetics400': 's254:s3://openmmlab/datasets/action/Kinetics400' })), dict( type='DenseSampleFrames', clip_len=1, frame_interval=1, num_clips=8, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict(type='CenterCrop', crop_size=224), dict(type='FormatShape', input_format='NCHW'), dict(type='PackActionInputs') ], test_mode=True)) test_dataloader = dict( batch_size=8, num_workers=8, persistent_workers=True, sampler=dict(type='DefaultSampler', shuffle=False), dataset=dict( type='VideoDataset', ann_file='data/kinetics400/kinetics400_val_list_videos.txt', data_prefix=dict(video='data/kinetics400/videos_val'), pipeline=[ dict( type='DecordInit', io_backend='petrel', path_mapping=dict({ 'data/kinetics400': 's254:s3://openmmlab/datasets/action/Kinetics400' })), dict( type='DenseSampleFrames', clip_len=1, frame_interval=1, num_clips=8, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict(type='CenterCrop', crop_size=224), dict(type='FormatShape', input_format='NCHW'), dict(type='PackActionInputs') ], test_mode=True)) val_evaluator = dict(type='AccMetric') test_evaluator = dict(type='AccMetric') launcher = 'slurm' work_dir = './work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb' 2022/09/06 17:19:33 - mmengine - INFO - These parameters in pretrained checkpoint are not loaded: {'fc.bias', 'fc.weight'} 2022/09/06 17:19:35 - mmengine - INFO - Checkpoints will be saved to /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb by HardDiskBackend. 2022/09/06 17:21:13 - mmengine - INFO - Epoch(train) [1][100/3757] lr: 1.0000e-02 eta: 4 days, 6:27:59 time: 0.3529 data_time: 0.0056 memory: 7124 grad_norm: 5.2387 top1_acc: 0.1250 top5_acc: 0.2500 loss_cls: 5.1425 loss: 5.1425 2022/09/06 17:21:36 - mmengine - INFO - Epoch(train) [1][200/3757] lr: 1.0000e-02 eta: 2 days, 14:59:52 time: 0.2733 data_time: 0.0207 memory: 7124 grad_norm: 6.0496 top1_acc: 0.0000 top5_acc: 0.1250 loss_cls: 4.9492 loss: 4.9492 2022/09/06 17:21:54 - mmengine - INFO - Epoch(train) [1][300/3757] lr: 1.0000e-02 eta: 2 days, 0:25:22 time: 0.1842 data_time: 0.0066 memory: 7124 grad_norm: 5.9973 top1_acc: 0.0000 top5_acc: 0.5000 loss_cls: 4.3250 loss: 4.3250 2022/09/06 17:22:14 - mmengine - INFO - Epoch(train) [1][400/3757] lr: 1.0000e-02 eta: 1 day, 17:25:42 time: 0.1445 data_time: 0.0070 memory: 7124 grad_norm: 5.9382 top1_acc: 0.1250 top5_acc: 0.2500 loss_cls: 4.5387 loss: 4.5387 2022/09/06 17:22:33 - mmengine - INFO - Epoch(train) [1][500/3757] lr: 1.0000e-02 eta: 1 day, 13:07:07 time: 0.1787 data_time: 0.0078 memory: 7124 grad_norm: 5.7333 top1_acc: 0.1250 top5_acc: 0.2500 loss_cls: 4.3781 loss: 4.3781 2022/09/06 17:22:52 - mmengine - INFO - Epoch(train) [1][600/3757] lr: 1.0000e-02 eta: 1 day, 10:08:43 time: 0.1728 data_time: 0.0075 memory: 7124 grad_norm: 5.8642 top1_acc: 0.0000 top5_acc: 0.2500 loss_cls: 4.2557 loss: 4.2557 2022/09/06 17:23:11 - mmengine - INFO - Epoch(train) [1][700/3757] lr: 1.0000e-02 eta: 1 day, 8:04:22 time: 0.1911 data_time: 0.0055 memory: 7124 grad_norm: 5.9283 top1_acc: 0.0000 top5_acc: 0.0000 loss_cls: 4.2971 loss: 4.2971 2022/09/06 17:23:28 - mmengine - INFO - Epoch(train) [1][800/3757] lr: 1.0000e-02 eta: 1 day, 6:20:28 time: 0.1599 data_time: 0.0072 memory: 7124 grad_norm: 5.8336 top1_acc: 0.0000 top5_acc: 0.3750 loss_cls: 3.7983 loss: 3.7983 2022/09/06 17:23:46 - mmengine - INFO - Epoch(train) [1][900/3757] lr: 1.0000e-02 eta: 1 day, 5:03:20 time: 0.1781 data_time: 0.0063 memory: 7124 grad_norm: 5.6827 top1_acc: 0.0000 top5_acc: 0.2500 loss_cls: 3.9086 loss: 3.9086 2022/09/06 17:24:04 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 17:24:04 - mmengine - INFO - Epoch(train) [1][1000/3757] lr: 1.0000e-02 eta: 1 day, 4:02:10 time: 0.1662 data_time: 0.0085 memory: 7124 grad_norm: 5.8581 top1_acc: 0.1250 top5_acc: 0.5000 loss_cls: 3.7443 loss: 3.7443 2022/09/06 17:24:22 - mmengine - INFO - Epoch(train) [1][1100/3757] lr: 1.0000e-02 eta: 1 day, 3:08:06 time: 0.2115 data_time: 0.0078 memory: 7124 grad_norm: 6.0831 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 3.7057 loss: 3.7057 2022/09/06 17:24:39 - mmengine - INFO - Epoch(train) [1][1200/3757] lr: 1.0000e-02 eta: 1 day, 2:21:57 time: 0.1697 data_time: 0.0077 memory: 7124 grad_norm: 6.1362 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 3.7292 loss: 3.7292 2022/09/06 17:24:59 - mmengine - INFO - Epoch(train) [1][1300/3757] lr: 1.0000e-02 eta: 1 day, 1:53:04 time: 0.2381 data_time: 0.0088 memory: 7124 grad_norm: 5.7870 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 3.6373 loss: 3.6373 2022/09/06 17:25:16 - mmengine - INFO - Epoch(train) [1][1400/3757] lr: 1.0000e-02 eta: 1 day, 1:17:37 time: 0.1730 data_time: 0.0164 memory: 7124 grad_norm: 6.0505 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 3.4137 loss: 3.4137 2022/09/06 17:25:34 - mmengine - INFO - Epoch(train) [1][1500/3757] lr: 1.0000e-02 eta: 1 day, 0:51:13 time: 0.1686 data_time: 0.0084 memory: 7124 grad_norm: 5.8871 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 3.6559 loss: 3.6559 2022/09/06 17:25:51 - mmengine - INFO - Epoch(train) [1][1600/3757] lr: 1.0000e-02 eta: 1 day, 0:25:14 time: 0.1602 data_time: 0.0095 memory: 7124 grad_norm: 5.9936 top1_acc: 0.2500 top5_acc: 0.2500 loss_cls: 3.8073 loss: 3.8073 2022/09/06 17:26:08 - mmengine - INFO - Epoch(train) [1][1700/3757] lr: 1.0000e-02 eta: 1 day, 0:00:31 time: 0.1845 data_time: 0.0251 memory: 7124 grad_norm: 5.7612 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 3.4711 loss: 3.4711 2022/09/06 17:26:30 - mmengine - INFO - Epoch(train) [1][1800/3757] lr: 1.0000e-02 eta: 23:58:26 time: 0.2415 data_time: 0.0069 memory: 7124 grad_norm: 5.7839 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 3.5728 loss: 3.5728 2022/09/06 17:26:50 - mmengine - INFO - Epoch(train) [1][1900/3757] lr: 1.0000e-02 eta: 23:46:47 time: 0.1976 data_time: 0.0086 memory: 7124 grad_norm: 5.6250 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 3.3350 loss: 3.3350 2022/09/06 17:27:08 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 17:27:08 - mmengine - INFO - Epoch(train) [1][2000/3757] lr: 1.0000e-02 eta: 23:30:15 time: 0.1705 data_time: 0.0065 memory: 7124 grad_norm: 6.1278 top1_acc: 0.0000 top5_acc: 0.5000 loss_cls: 3.7192 loss: 3.7192 2022/09/06 17:27:26 - mmengine - INFO - Epoch(train) [1][2100/3757] lr: 1.0000e-02 eta: 23:17:40 time: 0.1869 data_time: 0.0276 memory: 7124 grad_norm: 5.8450 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 3.2502 loss: 3.2502 2022/09/06 17:27:45 - mmengine - INFO - Epoch(train) [1][2200/3757] lr: 1.0000e-02 eta: 23:07:28 time: 0.1561 data_time: 0.0075 memory: 7124 grad_norm: 5.7542 top1_acc: 0.1250 top5_acc: 0.6250 loss_cls: 3.3400 loss: 3.3400 2022/09/06 17:28:03 - mmengine - INFO - Epoch(train) [1][2300/3757] lr: 1.0000e-02 eta: 22:55:30 time: 0.1781 data_time: 0.0068 memory: 7124 grad_norm: 5.7549 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 3.3065 loss: 3.3065 2022/09/06 17:28:20 - mmengine - INFO - Epoch(train) [1][2400/3757] lr: 1.0000e-02 eta: 22:41:43 time: 0.1523 data_time: 0.0091 memory: 7124 grad_norm: 5.6001 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 3.1018 loss: 3.1018 2022/09/06 17:28:37 - mmengine - INFO - Epoch(train) [1][2500/3757] lr: 1.0000e-02 eta: 22:29:35 time: 0.1661 data_time: 0.0062 memory: 7124 grad_norm: 5.5341 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 3.2500 loss: 3.2500 2022/09/06 17:28:55 - mmengine - INFO - Epoch(train) [1][2600/3757] lr: 1.0000e-02 eta: 22:20:05 time: 0.1598 data_time: 0.0095 memory: 7124 grad_norm: 5.8209 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 3.3687 loss: 3.3687 2022/09/06 17:29:12 - mmengine - INFO - Epoch(train) [1][2700/3757] lr: 1.0000e-02 eta: 22:09:39 time: 0.1540 data_time: 0.0081 memory: 7124 grad_norm: 5.9002 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 3.2625 loss: 3.2625 2022/09/06 17:29:31 - mmengine - INFO - Epoch(train) [1][2800/3757] lr: 1.0000e-02 eta: 22:01:54 time: 0.1878 data_time: 0.0405 memory: 7124 grad_norm: 5.8861 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 3.4185 loss: 3.4185 2022/09/06 17:29:48 - mmengine - INFO - Epoch(train) [1][2900/3757] lr: 1.0000e-02 eta: 21:53:57 time: 0.1491 data_time: 0.0088 memory: 7124 grad_norm: 5.8230 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 3.2991 loss: 3.2991 2022/09/06 17:30:05 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 17:30:05 - mmengine - INFO - Epoch(train) [1][3000/3757] lr: 1.0000e-02 eta: 21:45:06 time: 0.1494 data_time: 0.0072 memory: 7124 grad_norm: 5.7757 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.9513 loss: 2.9513 2022/09/06 17:30:22 - mmengine - INFO - Epoch(train) [1][3100/3757] lr: 1.0000e-02 eta: 21:36:48 time: 0.1693 data_time: 0.0091 memory: 7124 grad_norm: 5.8216 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 3.1508 loss: 3.1508 2022/09/06 17:30:40 - mmengine - INFO - Epoch(train) [1][3200/3757] lr: 1.0000e-02 eta: 21:30:57 time: 0.2009 data_time: 0.0074 memory: 7124 grad_norm: 5.8201 top1_acc: 0.1250 top5_acc: 0.5000 loss_cls: 3.0221 loss: 3.0221 2022/09/06 17:31:01 - mmengine - INFO - Epoch(train) [1][3300/3757] lr: 1.0000e-02 eta: 21:29:57 time: 0.2687 data_time: 0.0434 memory: 7124 grad_norm: 5.8954 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 3.1642 loss: 3.1642 2022/09/06 17:31:18 - mmengine - INFO - Epoch(train) [1][3400/3757] lr: 1.0000e-02 eta: 21:22:54 time: 0.1621 data_time: 0.0168 memory: 7124 grad_norm: 5.7868 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 3.3268 loss: 3.3268 2022/09/06 17:31:37 - mmengine - INFO - Epoch(train) [1][3500/3757] lr: 1.0000e-02 eta: 21:19:40 time: 0.2430 data_time: 0.0071 memory: 7124 grad_norm: 5.5443 top1_acc: 0.1250 top5_acc: 0.3750 loss_cls: 3.3416 loss: 3.3416 2022/09/06 17:31:54 - mmengine - INFO - Epoch(train) [1][3600/3757] lr: 1.0000e-02 eta: 21:12:12 time: 0.1549 data_time: 0.0089 memory: 7124 grad_norm: 5.3093 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 3.3941 loss: 3.3941 2022/09/06 17:32:12 - mmengine - INFO - Epoch(train) [1][3700/3757] lr: 1.0000e-02 eta: 21:08:55 time: 0.1537 data_time: 0.0071 memory: 7124 grad_norm: 5.7047 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.8351 loss: 2.8351 2022/09/06 17:32:23 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 17:32:23 - mmengine - INFO - Epoch(train) [1][3757/3757] lr: 1.0000e-02 eta: 21:08:48 time: 0.1523 data_time: 0.0050 memory: 7124 grad_norm: 5.7736 top1_acc: 0.4286 top5_acc: 0.5714 loss_cls: 3.2342 loss: 3.2342 2022/09/06 17:32:23 - mmengine - INFO - Saving checkpoint at 1 epochs 2022/09/06 17:35:35 - mmengine - INFO - Epoch(val) [1][100/310] eta: 0:04:30 time: 1.2899 data_time: 0.9890 memory: 7627 2022/09/06 17:37:52 - mmengine - INFO - Epoch(val) [1][200/310] eta: 0:02:23 time: 1.3077 data_time: 1.0041 memory: 7627 2022/09/06 17:39:58 - mmengine - INFO - Epoch(val) [1][300/310] eta: 0:00:12 time: 1.2218 data_time: 0.9235 memory: 7627 2022/09/06 17:40:20 - mmengine - INFO - Epoch(val) [1][310/310] acc/top1: 0.3659 acc/top5: 0.6423 acc/mean1: 0.3656 2022/09/06 17:40:23 - mmengine - INFO - The best checkpoint with 0.3659 acc/top1 at 1 epoch is saved to best_acc/top1_epoch_1.pth. 2022/09/06 17:40:39 - mmengine - INFO - Epoch(train) [2][100/3757] lr: 1.0000e-02 eta: 20:55:53 time: 0.1542 data_time: 0.0085 memory: 7627 grad_norm: 5.3745 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.6742 loss: 2.6742 2022/09/06 17:40:55 - mmengine - INFO - Epoch(train) [2][200/3757] lr: 1.0000e-02 eta: 20:48:48 time: 0.1531 data_time: 0.0094 memory: 7124 grad_norm: 5.8004 top1_acc: 0.1250 top5_acc: 0.5000 loss_cls: 3.0155 loss: 3.0155 2022/09/06 17:41:02 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 17:41:11 - mmengine - INFO - Epoch(train) [2][300/3757] lr: 1.0000e-02 eta: 20:41:31 time: 0.1555 data_time: 0.0106 memory: 7124 grad_norm: 5.7435 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.9906 loss: 2.9906 2022/09/06 17:41:26 - mmengine - INFO - Epoch(train) [2][400/3757] lr: 1.0000e-02 eta: 20:34:40 time: 0.1557 data_time: 0.0084 memory: 7124 grad_norm: 5.6743 top1_acc: 0.1250 top5_acc: 0.5000 loss_cls: 2.9692 loss: 2.9692 2022/09/06 17:41:42 - mmengine - INFO - Epoch(train) [2][500/3757] lr: 1.0000e-02 eta: 20:28:02 time: 0.1585 data_time: 0.0105 memory: 7124 grad_norm: 5.7289 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 3.0834 loss: 3.0834 2022/09/06 17:41:57 - mmengine - INFO - Epoch(train) [2][600/3757] lr: 1.0000e-02 eta: 20:21:40 time: 0.1553 data_time: 0.0097 memory: 7124 grad_norm: 5.4135 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.9577 loss: 2.9577 2022/09/06 17:42:13 - mmengine - INFO - Epoch(train) [2][700/3757] lr: 1.0000e-02 eta: 20:15:49 time: 0.1586 data_time: 0.0090 memory: 7124 grad_norm: 5.7255 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.9532 loss: 2.9532 2022/09/06 17:42:29 - mmengine - INFO - Epoch(train) [2][800/3757] lr: 1.0000e-02 eta: 20:10:11 time: 0.1596 data_time: 0.0093 memory: 7124 grad_norm: 5.6214 top1_acc: 0.1250 top5_acc: 0.8750 loss_cls: 2.6562 loss: 2.6562 2022/09/06 17:42:45 - mmengine - INFO - Epoch(train) [2][900/3757] lr: 1.0000e-02 eta: 20:04:48 time: 0.1550 data_time: 0.0086 memory: 7124 grad_norm: 5.3142 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 3.0428 loss: 3.0428 2022/09/06 17:43:00 - mmengine - INFO - Epoch(train) [2][1000/3757] lr: 1.0000e-02 eta: 19:59:28 time: 0.1610 data_time: 0.0088 memory: 7124 grad_norm: 5.4999 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 3.0660 loss: 3.0660 2022/09/06 17:43:16 - mmengine - INFO - Epoch(train) [2][1100/3757] lr: 1.0000e-02 eta: 19:54:03 time: 0.1537 data_time: 0.0087 memory: 7124 grad_norm: 5.4952 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 3.0954 loss: 3.0954 2022/09/06 17:43:31 - mmengine - INFO - Epoch(train) [2][1200/3757] lr: 1.0000e-02 eta: 19:49:10 time: 0.1562 data_time: 0.0089 memory: 7124 grad_norm: 5.6711 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.9200 loss: 2.9200 2022/09/06 17:43:38 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 17:43:47 - mmengine - INFO - Epoch(train) [2][1300/3757] lr: 1.0000e-02 eta: 19:45:01 time: 0.1597 data_time: 0.0083 memory: 7124 grad_norm: 5.7218 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.9037 loss: 2.9037 2022/09/06 17:44:03 - mmengine - INFO - Epoch(train) [2][1400/3757] lr: 1.0000e-02 eta: 19:40:48 time: 0.1561 data_time: 0.0088 memory: 7124 grad_norm: 5.8131 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.8496 loss: 2.8496 2022/09/06 17:44:19 - mmengine - INFO - Epoch(train) [2][1500/3757] lr: 1.0000e-02 eta: 19:36:21 time: 0.1585 data_time: 0.0097 memory: 7124 grad_norm: 6.0332 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 3.1531 loss: 3.1531 2022/09/06 17:44:35 - mmengine - INFO - Epoch(train) [2][1600/3757] lr: 1.0000e-02 eta: 19:31:57 time: 0.1532 data_time: 0.0091 memory: 7124 grad_norm: 5.7141 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.8448 loss: 2.8448 2022/09/06 17:44:51 - mmengine - INFO - Epoch(train) [2][1700/3757] lr: 1.0000e-02 eta: 19:28:14 time: 0.1554 data_time: 0.0091 memory: 7124 grad_norm: 5.4576 top1_acc: 0.1250 top5_acc: 0.7500 loss_cls: 2.9912 loss: 2.9912 2022/09/06 17:45:06 - mmengine - INFO - Epoch(train) [2][1800/3757] lr: 1.0000e-02 eta: 19:24:13 time: 0.1573 data_time: 0.0087 memory: 7124 grad_norm: 5.6789 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.6239 loss: 2.6239 2022/09/06 17:45:22 - mmengine - INFO - Epoch(train) [2][1900/3757] lr: 1.0000e-02 eta: 19:20:20 time: 0.1576 data_time: 0.0094 memory: 7124 grad_norm: 5.6034 top1_acc: 0.1250 top5_acc: 0.2500 loss_cls: 3.1995 loss: 3.1995 2022/09/06 17:45:37 - mmengine - INFO - Epoch(train) [2][2000/3757] lr: 1.0000e-02 eta: 19:16:43 time: 0.1538 data_time: 0.0094 memory: 7124 grad_norm: 5.7548 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.7053 loss: 2.7053 2022/09/06 17:45:53 - mmengine - INFO - Epoch(train) [2][2100/3757] lr: 1.0000e-02 eta: 19:13:04 time: 0.1571 data_time: 0.0100 memory: 7124 grad_norm: 5.7693 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.5469 loss: 2.5469 2022/09/06 17:46:08 - mmengine - INFO - Epoch(train) [2][2200/3757] lr: 1.0000e-02 eta: 19:09:21 time: 0.1553 data_time: 0.0098 memory: 7124 grad_norm: 5.5325 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.7057 loss: 2.7057 2022/09/06 17:46:15 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 17:46:24 - mmengine - INFO - Epoch(train) [2][2300/3757] lr: 1.0000e-02 eta: 19:06:00 time: 0.1549 data_time: 0.0083 memory: 7124 grad_norm: 6.3919 top1_acc: 0.1250 top5_acc: 0.3750 loss_cls: 3.1410 loss: 3.1410 2022/09/06 17:46:40 - mmengine - INFO - Epoch(train) [2][2400/3757] lr: 1.0000e-02 eta: 19:02:30 time: 0.1555 data_time: 0.0089 memory: 7124 grad_norm: 5.7225 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 2.8808 loss: 2.8808 2022/09/06 17:46:55 - mmengine - INFO - Epoch(train) [2][2500/3757] lr: 1.0000e-02 eta: 18:59:14 time: 0.1528 data_time: 0.0089 memory: 7124 grad_norm: 5.7664 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.7376 loss: 2.7376 2022/09/06 17:47:11 - mmengine - INFO - Epoch(train) [2][2600/3757] lr: 1.0000e-02 eta: 18:56:11 time: 0.1555 data_time: 0.0094 memory: 7124 grad_norm: 5.5777 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.6210 loss: 2.6210 2022/09/06 17:47:26 - mmengine - INFO - Epoch(train) [2][2700/3757] lr: 1.0000e-02 eta: 18:53:02 time: 0.1523 data_time: 0.0085 memory: 7124 grad_norm: 5.3626 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.9768 loss: 2.9768 2022/09/06 17:47:42 - mmengine - INFO - Epoch(train) [2][2800/3757] lr: 1.0000e-02 eta: 18:50:09 time: 0.1628 data_time: 0.0128 memory: 7124 grad_norm: 5.6722 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.6685 loss: 2.6685 2022/09/06 17:47:59 - mmengine - INFO - Epoch(train) [2][2900/3757] lr: 1.0000e-02 eta: 18:48:21 time: 0.1535 data_time: 0.0101 memory: 7124 grad_norm: 5.3203 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.5726 loss: 2.5726 2022/09/06 17:48:14 - mmengine - INFO - Epoch(train) [2][3000/3757] lr: 1.0000e-02 eta: 18:45:35 time: 0.1552 data_time: 0.0095 memory: 7124 grad_norm: 5.3468 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.7832 loss: 2.7832 2022/09/06 17:48:30 - mmengine - INFO - Epoch(train) [2][3100/3757] lr: 1.0000e-02 eta: 18:42:52 time: 0.1528 data_time: 0.0078 memory: 7124 grad_norm: 5.4520 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.9280 loss: 2.9280 2022/09/06 17:48:46 - mmengine - INFO - Epoch(train) [2][3200/3757] lr: 1.0000e-02 eta: 18:40:18 time: 0.1590 data_time: 0.0092 memory: 7124 grad_norm: 5.4694 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.6447 loss: 2.6447 2022/09/06 17:48:52 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 17:49:01 - mmengine - INFO - Epoch(train) [2][3300/3757] lr: 1.0000e-02 eta: 18:37:35 time: 0.1563 data_time: 0.0098 memory: 7124 grad_norm: 5.5207 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.6230 loss: 2.6230 2022/09/06 17:49:17 - mmengine - INFO - Epoch(train) [2][3400/3757] lr: 1.0000e-02 eta: 18:35:05 time: 0.1533 data_time: 0.0094 memory: 7124 grad_norm: 5.5818 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.9307 loss: 2.9307 2022/09/06 17:49:33 - mmengine - INFO - Epoch(train) [2][3500/3757] lr: 1.0000e-02 eta: 18:32:51 time: 0.1556 data_time: 0.0079 memory: 7124 grad_norm: 5.3038 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 3.0199 loss: 3.0199 2022/09/06 17:49:48 - mmengine - INFO - Epoch(train) [2][3600/3757] lr: 1.0000e-02 eta: 18:30:37 time: 0.1667 data_time: 0.0101 memory: 7124 grad_norm: 5.4861 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.7670 loss: 2.7670 2022/09/06 17:50:04 - mmengine - INFO - Epoch(train) [2][3700/3757] lr: 1.0000e-02 eta: 18:28:10 time: 0.1540 data_time: 0.0089 memory: 7124 grad_norm: 5.4551 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 3.0803 loss: 3.0803 2022/09/06 17:50:12 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 17:50:12 - mmengine - INFO - Epoch(train) [2][3757/3757] lr: 1.0000e-02 eta: 18:27:12 time: 0.1340 data_time: 0.0066 memory: 7124 grad_norm: 5.2923 top1_acc: 0.8571 top5_acc: 0.8571 loss_cls: 2.6851 loss: 2.6851 2022/09/06 17:50:12 - mmengine - INFO - Saving checkpoint at 2 epochs 2022/09/06 17:52:31 - mmengine - INFO - Epoch(val) [2][100/310] eta: 0:04:25 time: 1.2655 data_time: 0.9621 memory: 7627 2022/09/06 17:54:47 - mmengine - INFO - Epoch(val) [2][200/310] eta: 0:02:14 time: 1.2249 data_time: 0.9216 memory: 7627 2022/09/06 17:56:52 - mmengine - INFO - Epoch(val) [2][300/310] eta: 0:00:12 time: 1.2230 data_time: 0.9113 memory: 7627 2022/09/06 17:57:12 - mmengine - INFO - Epoch(val) [2][310/310] acc/top1: 0.4550 acc/top5: 0.7320 acc/mean1: 0.4548 2022/09/06 17:57:12 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_1.pth is removed 2022/09/06 17:57:14 - mmengine - INFO - The best checkpoint with 0.4550 acc/top1 at 2 epoch is saved to best_acc/top1_epoch_2.pth. 2022/09/06 17:57:30 - mmengine - INFO - Epoch(train) [3][100/3757] lr: 1.0000e-02 eta: 18:23:19 time: 0.1568 data_time: 0.0111 memory: 7627 grad_norm: 5.3899 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 2.8959 loss: 2.8959 2022/09/06 17:57:46 - mmengine - INFO - Epoch(train) [3][200/3757] lr: 1.0000e-02 eta: 18:21:19 time: 0.1579 data_time: 0.0097 memory: 7124 grad_norm: 5.4108 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 2.9549 loss: 2.9549 2022/09/06 17:58:02 - mmengine - INFO - Epoch(train) [3][300/3757] lr: 1.0000e-02 eta: 18:19:09 time: 0.1553 data_time: 0.0089 memory: 7124 grad_norm: 5.5527 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.6732 loss: 2.6732 2022/09/06 17:58:17 - mmengine - INFO - Epoch(train) [3][400/3757] lr: 1.0000e-02 eta: 18:17:09 time: 0.1545 data_time: 0.0092 memory: 7124 grad_norm: 5.5588 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.6751 loss: 2.6751 2022/09/06 17:58:31 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 17:58:33 - mmengine - INFO - Epoch(train) [3][500/3757] lr: 1.0000e-02 eta: 18:15:15 time: 0.1593 data_time: 0.0121 memory: 7124 grad_norm: 5.3393 top1_acc: 0.1250 top5_acc: 0.5000 loss_cls: 2.5062 loss: 2.5062 2022/09/06 17:58:49 - mmengine - INFO - Epoch(train) [3][600/3757] lr: 1.0000e-02 eta: 18:13:11 time: 0.1511 data_time: 0.0089 memory: 7124 grad_norm: 5.5898 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.7264 loss: 2.7264 2022/09/06 17:59:04 - mmengine - INFO - Epoch(train) [3][700/3757] lr: 1.0000e-02 eta: 18:11:13 time: 0.1573 data_time: 0.0095 memory: 7124 grad_norm: 5.7791 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.8891 loss: 2.8891 2022/09/06 17:59:20 - mmengine - INFO - Epoch(train) [3][800/3757] lr: 1.0000e-02 eta: 18:09:17 time: 0.1616 data_time: 0.0097 memory: 7124 grad_norm: 5.4619 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 2.6983 loss: 2.6983 2022/09/06 17:59:39 - mmengine - INFO - Epoch(train) [3][900/3757] lr: 1.0000e-02 eta: 18:09:56 time: 0.1543 data_time: 0.0099 memory: 7124 grad_norm: 5.4463 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.6877 loss: 2.6877 2022/09/06 17:59:55 - mmengine - INFO - Epoch(train) [3][1000/3757] lr: 1.0000e-02 eta: 18:08:05 time: 0.1493 data_time: 0.0089 memory: 7124 grad_norm: 5.4990 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.4025 loss: 2.4025 2022/09/06 18:00:15 - mmengine - INFO - Epoch(train) [3][1100/3757] lr: 1.0000e-02 eta: 18:09:44 time: 0.1558 data_time: 0.0093 memory: 7124 grad_norm: 5.3905 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.7598 loss: 2.7598 2022/09/06 18:00:31 - mmengine - INFO - Epoch(train) [3][1200/3757] lr: 1.0000e-02 eta: 18:08:01 time: 0.1520 data_time: 0.0081 memory: 7124 grad_norm: 5.5248 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.7312 loss: 2.7312 2022/09/06 18:00:47 - mmengine - INFO - Epoch(train) [3][1300/3757] lr: 1.0000e-02 eta: 18:06:20 time: 0.1556 data_time: 0.0091 memory: 7124 grad_norm: 5.5296 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.7547 loss: 2.7547 2022/09/06 18:01:03 - mmengine - INFO - Epoch(train) [3][1400/3757] lr: 1.0000e-02 eta: 18:04:34 time: 0.1629 data_time: 0.0094 memory: 7124 grad_norm: 5.2278 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.9843 loss: 2.9843 2022/09/06 18:01:16 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:01:18 - mmengine - INFO - Epoch(train) [3][1500/3757] lr: 1.0000e-02 eta: 18:02:48 time: 0.1569 data_time: 0.0087 memory: 7124 grad_norm: 5.2946 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.5676 loss: 2.5676 2022/09/06 18:01:34 - mmengine - INFO - Epoch(train) [3][1600/3757] lr: 1.0000e-02 eta: 18:01:07 time: 0.1592 data_time: 0.0111 memory: 7124 grad_norm: 5.4756 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.4484 loss: 2.4484 2022/09/06 18:01:49 - mmengine - INFO - Epoch(train) [3][1700/3757] lr: 1.0000e-02 eta: 17:59:21 time: 0.1555 data_time: 0.0097 memory: 7124 grad_norm: 5.3712 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.7398 loss: 2.7398 2022/09/06 18:02:05 - mmengine - INFO - Epoch(train) [3][1800/3757] lr: 1.0000e-02 eta: 17:57:58 time: 0.1769 data_time: 0.0188 memory: 7124 grad_norm: 5.5815 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.7489 loss: 2.7489 2022/09/06 18:02:21 - mmengine - INFO - Epoch(train) [3][1900/3757] lr: 1.0000e-02 eta: 17:56:20 time: 0.1577 data_time: 0.0106 memory: 7124 grad_norm: 5.3264 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.3605 loss: 2.3605 2022/09/06 18:02:36 - mmengine - INFO - Epoch(train) [3][2000/3757] lr: 1.0000e-02 eta: 17:54:44 time: 0.1506 data_time: 0.0097 memory: 7124 grad_norm: 5.5289 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 2.6919 loss: 2.6919 2022/09/06 18:02:52 - mmengine - INFO - Epoch(train) [3][2100/3757] lr: 1.0000e-02 eta: 17:53:14 time: 0.1550 data_time: 0.0105 memory: 7124 grad_norm: 5.3953 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.9380 loss: 2.9380 2022/09/06 18:03:08 - mmengine - INFO - Epoch(train) [3][2200/3757] lr: 1.0000e-02 eta: 17:51:42 time: 0.1550 data_time: 0.0093 memory: 7124 grad_norm: 5.4570 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 2.3141 loss: 2.3141 2022/09/06 18:03:23 - mmengine - INFO - Epoch(train) [3][2300/3757] lr: 1.0000e-02 eta: 17:50:15 time: 0.1530 data_time: 0.0089 memory: 7124 grad_norm: 5.5824 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.6849 loss: 2.6849 2022/09/06 18:03:39 - mmengine - INFO - Epoch(train) [3][2400/3757] lr: 1.0000e-02 eta: 17:48:47 time: 0.1516 data_time: 0.0084 memory: 7124 grad_norm: 5.5140 top1_acc: 0.1250 top5_acc: 0.3750 loss_cls: 2.6753 loss: 2.6753 2022/09/06 18:03:53 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:03:55 - mmengine - INFO - Epoch(train) [3][2500/3757] lr: 1.0000e-02 eta: 17:47:26 time: 0.1532 data_time: 0.0090 memory: 7124 grad_norm: 5.5178 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.6948 loss: 2.6948 2022/09/06 18:04:11 - mmengine - INFO - Epoch(train) [3][2600/3757] lr: 1.0000e-02 eta: 17:46:07 time: 0.1541 data_time: 0.0100 memory: 7124 grad_norm: 5.4060 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.6963 loss: 2.6963 2022/09/06 18:04:26 - mmengine - INFO - Epoch(train) [3][2700/3757] lr: 1.0000e-02 eta: 17:44:43 time: 0.1563 data_time: 0.0095 memory: 7124 grad_norm: 5.4159 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.6314 loss: 2.6314 2022/09/06 18:04:42 - mmengine - INFO - Epoch(train) [3][2800/3757] lr: 1.0000e-02 eta: 17:43:25 time: 0.1602 data_time: 0.0085 memory: 7124 grad_norm: 5.4084 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.7946 loss: 2.7946 2022/09/06 18:04:58 - mmengine - INFO - Epoch(train) [3][2900/3757] lr: 1.0000e-02 eta: 17:41:59 time: 0.1602 data_time: 0.0099 memory: 7124 grad_norm: 5.1483 top1_acc: 0.3750 top5_acc: 0.3750 loss_cls: 2.6612 loss: 2.6612 2022/09/06 18:05:13 - mmengine - INFO - Epoch(train) [3][3000/3757] lr: 1.0000e-02 eta: 17:40:39 time: 0.1629 data_time: 0.0089 memory: 7124 grad_norm: 5.1461 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.2634 loss: 2.2634 2022/09/06 18:05:29 - mmengine - INFO - Epoch(train) [3][3100/3757] lr: 1.0000e-02 eta: 17:39:17 time: 0.1530 data_time: 0.0095 memory: 7124 grad_norm: 5.4306 top1_acc: 0.1250 top5_acc: 0.3750 loss_cls: 3.1599 loss: 3.1599 2022/09/06 18:05:45 - mmengine - INFO - Epoch(train) [3][3200/3757] lr: 1.0000e-02 eta: 17:38:04 time: 0.1582 data_time: 0.0100 memory: 7124 grad_norm: 5.5099 top1_acc: 0.1250 top5_acc: 0.5000 loss_cls: 2.3733 loss: 2.3733 2022/09/06 18:06:00 - mmengine - INFO - Epoch(train) [3][3300/3757] lr: 1.0000e-02 eta: 17:36:46 time: 0.1597 data_time: 0.0086 memory: 7124 grad_norm: 5.2359 top1_acc: 0.1250 top5_acc: 0.7500 loss_cls: 2.4968 loss: 2.4968 2022/09/06 18:06:16 - mmengine - INFO - Epoch(train) [3][3400/3757] lr: 1.0000e-02 eta: 17:35:30 time: 0.1554 data_time: 0.0107 memory: 7124 grad_norm: 5.4481 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.5348 loss: 2.5348 2022/09/06 18:06:29 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:06:31 - mmengine - INFO - Epoch(train) [3][3500/3757] lr: 1.0000e-02 eta: 17:34:11 time: 0.1574 data_time: 0.0102 memory: 7124 grad_norm: 5.2650 top1_acc: 0.0000 top5_acc: 0.2500 loss_cls: 2.7605 loss: 2.7605 2022/09/06 18:06:47 - mmengine - INFO - Epoch(train) [3][3600/3757] lr: 1.0000e-02 eta: 17:33:06 time: 0.1594 data_time: 0.0090 memory: 7124 grad_norm: 5.3752 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.5131 loss: 2.5131 2022/09/06 18:07:03 - mmengine - INFO - Epoch(train) [3][3700/3757] lr: 1.0000e-02 eta: 17:31:49 time: 0.1563 data_time: 0.0086 memory: 7124 grad_norm: 5.2308 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 2.6561 loss: 2.6561 2022/09/06 18:07:11 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:07:11 - mmengine - INFO - Epoch(train) [3][3757/3757] lr: 1.0000e-02 eta: 17:31:20 time: 0.1340 data_time: 0.0071 memory: 7124 grad_norm: 5.3359 top1_acc: 0.0000 top5_acc: 0.1429 loss_cls: 2.7521 loss: 2.7521 2022/09/06 18:07:11 - mmengine - INFO - Saving checkpoint at 3 epochs 2022/09/06 18:09:30 - mmengine - INFO - Epoch(val) [3][100/310] eta: 0:04:03 time: 1.1575 data_time: 0.8569 memory: 7627 2022/09/06 18:11:47 - mmengine - INFO - Epoch(val) [3][200/310] eta: 0:02:16 time: 1.2437 data_time: 0.9428 memory: 7627 2022/09/06 18:13:52 - mmengine - INFO - Epoch(val) [3][300/310] eta: 0:00:12 time: 1.2395 data_time: 0.9413 memory: 7627 2022/09/06 18:14:10 - mmengine - INFO - Epoch(val) [3][310/310] acc/top1: 0.4929 acc/top5: 0.7613 acc/mean1: 0.4927 2022/09/06 18:14:10 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_2.pth is removed 2022/09/06 18:14:12 - mmengine - INFO - The best checkpoint with 0.4929 acc/top1 at 3 epoch is saved to best_acc/top1_epoch_3.pth. 2022/09/06 18:14:28 - mmengine - INFO - Epoch(train) [4][100/3757] lr: 1.0000e-02 eta: 17:29:00 time: 0.1599 data_time: 0.0165 memory: 7627 grad_norm: 5.4975 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.4330 loss: 2.4330 2022/09/06 18:14:44 - mmengine - INFO - Epoch(train) [4][200/3757] lr: 1.0000e-02 eta: 17:27:49 time: 0.1580 data_time: 0.0098 memory: 7124 grad_norm: 5.6126 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.6753 loss: 2.6753 2022/09/06 18:15:00 - mmengine - INFO - Epoch(train) [4][300/3757] lr: 1.0000e-02 eta: 17:26:45 time: 0.1536 data_time: 0.0086 memory: 7124 grad_norm: 5.3813 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.8996 loss: 2.8996 2022/09/06 18:15:15 - mmengine - INFO - Epoch(train) [4][400/3757] lr: 1.0000e-02 eta: 17:25:39 time: 0.1533 data_time: 0.0092 memory: 7124 grad_norm: 5.3299 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.4664 loss: 2.4664 2022/09/06 18:15:31 - mmengine - INFO - Epoch(train) [4][500/3757] lr: 1.0000e-02 eta: 17:24:37 time: 0.1565 data_time: 0.0091 memory: 7124 grad_norm: 5.2529 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.3674 loss: 2.3674 2022/09/06 18:15:47 - mmengine - INFO - Epoch(train) [4][600/3757] lr: 1.0000e-02 eta: 17:23:33 time: 0.1521 data_time: 0.0085 memory: 7124 grad_norm: 5.4493 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.3579 loss: 2.3579 2022/09/06 18:16:02 - mmengine - INFO - Epoch(train) [4][700/3757] lr: 1.0000e-02 eta: 17:22:32 time: 0.1517 data_time: 0.0084 memory: 7124 grad_norm: 5.4971 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 2.7268 loss: 2.7268 2022/09/06 18:16:07 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:16:18 - mmengine - INFO - Epoch(train) [4][800/3757] lr: 1.0000e-02 eta: 17:21:32 time: 0.1569 data_time: 0.0091 memory: 7124 grad_norm: 5.5029 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 2.7474 loss: 2.7474 2022/09/06 18:16:34 - mmengine - INFO - Epoch(train) [4][900/3757] lr: 1.0000e-02 eta: 17:20:31 time: 0.1649 data_time: 0.0089 memory: 7124 grad_norm: 5.3334 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.3889 loss: 2.3889 2022/09/06 18:16:50 - mmengine - INFO - Epoch(train) [4][1000/3757] lr: 1.0000e-02 eta: 17:19:28 time: 0.1559 data_time: 0.0097 memory: 7124 grad_norm: 5.6077 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.4779 loss: 2.4779 2022/09/06 18:17:05 - mmengine - INFO - Epoch(train) [4][1100/3757] lr: 1.0000e-02 eta: 17:18:28 time: 0.1588 data_time: 0.0100 memory: 7124 grad_norm: 5.4439 top1_acc: 0.1250 top5_acc: 0.6250 loss_cls: 2.6242 loss: 2.6242 2022/09/06 18:17:21 - mmengine - INFO - Epoch(train) [4][1200/3757] lr: 1.0000e-02 eta: 17:17:31 time: 0.1576 data_time: 0.0082 memory: 7124 grad_norm: 5.1617 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.7383 loss: 2.7383 2022/09/06 18:17:37 - mmengine - INFO - Epoch(train) [4][1300/3757] lr: 1.0000e-02 eta: 17:16:42 time: 0.1566 data_time: 0.0094 memory: 7124 grad_norm: 5.5249 top1_acc: 0.3750 top5_acc: 0.3750 loss_cls: 2.4209 loss: 2.4209 2022/09/06 18:17:53 - mmengine - INFO - Epoch(train) [4][1400/3757] lr: 1.0000e-02 eta: 17:15:42 time: 0.1545 data_time: 0.0105 memory: 7124 grad_norm: 5.0945 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.5501 loss: 2.5501 2022/09/06 18:18:09 - mmengine - INFO - Epoch(train) [4][1500/3757] lr: 1.0000e-02 eta: 17:14:45 time: 0.1582 data_time: 0.0097 memory: 7124 grad_norm: 5.1844 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.4835 loss: 2.4835 2022/09/06 18:18:24 - mmengine - INFO - Epoch(train) [4][1600/3757] lr: 1.0000e-02 eta: 17:13:47 time: 0.1570 data_time: 0.0101 memory: 7124 grad_norm: 5.3256 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.4459 loss: 2.4459 2022/09/06 18:18:40 - mmengine - INFO - Epoch(train) [4][1700/3757] lr: 1.0000e-02 eta: 17:12:48 time: 0.1542 data_time: 0.0104 memory: 7124 grad_norm: 5.3243 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.7173 loss: 2.7173 2022/09/06 18:18:44 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:18:56 - mmengine - INFO - Epoch(train) [4][1800/3757] lr: 1.0000e-02 eta: 17:11:53 time: 0.1609 data_time: 0.0087 memory: 7124 grad_norm: 5.1482 top1_acc: 0.1250 top5_acc: 0.3750 loss_cls: 2.7390 loss: 2.7390 2022/09/06 18:19:11 - mmengine - INFO - Epoch(train) [4][1900/3757] lr: 1.0000e-02 eta: 17:11:01 time: 0.1614 data_time: 0.0089 memory: 7124 grad_norm: 5.3629 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 2.7927 loss: 2.7927 2022/09/06 18:19:27 - mmengine - INFO - Epoch(train) [4][2000/3757] lr: 1.0000e-02 eta: 17:10:06 time: 0.1518 data_time: 0.0092 memory: 7124 grad_norm: 5.4083 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.4403 loss: 2.4403 2022/09/06 18:19:43 - mmengine - INFO - Epoch(train) [4][2100/3757] lr: 1.0000e-02 eta: 17:09:22 time: 0.1567 data_time: 0.0089 memory: 7124 grad_norm: 5.2935 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.3870 loss: 2.3870 2022/09/06 18:19:59 - mmengine - INFO - Epoch(train) [4][2200/3757] lr: 1.0000e-02 eta: 17:08:30 time: 0.1554 data_time: 0.0093 memory: 7124 grad_norm: 5.5389 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.3555 loss: 2.3555 2022/09/06 18:20:14 - mmengine - INFO - Epoch(train) [4][2300/3757] lr: 1.0000e-02 eta: 17:07:35 time: 0.1596 data_time: 0.0095 memory: 7124 grad_norm: 5.2137 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.4222 loss: 2.4222 2022/09/06 18:20:30 - mmengine - INFO - Epoch(train) [4][2400/3757] lr: 1.0000e-02 eta: 17:06:43 time: 0.1610 data_time: 0.0084 memory: 7124 grad_norm: 5.4065 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.6518 loss: 2.6518 2022/09/06 18:20:46 - mmengine - INFO - Epoch(train) [4][2500/3757] lr: 1.0000e-02 eta: 17:05:51 time: 0.1559 data_time: 0.0101 memory: 7124 grad_norm: 5.4575 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.5166 loss: 2.5166 2022/09/06 18:21:02 - mmengine - INFO - Epoch(train) [4][2600/3757] lr: 1.0000e-02 eta: 17:05:00 time: 0.1545 data_time: 0.0101 memory: 7124 grad_norm: 5.3356 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.2822 loss: 2.2822 2022/09/06 18:21:18 - mmengine - INFO - Epoch(train) [4][2700/3757] lr: 1.0000e-02 eta: 17:04:15 time: 0.1517 data_time: 0.0086 memory: 7124 grad_norm: 5.4559 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.4487 loss: 2.4487 2022/09/06 18:21:22 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:21:33 - mmengine - INFO - Epoch(train) [4][2800/3757] lr: 1.0000e-02 eta: 17:03:27 time: 0.1619 data_time: 0.0102 memory: 7124 grad_norm: 5.2725 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.6479 loss: 2.6479 2022/09/06 18:21:49 - mmengine - INFO - Epoch(train) [4][2900/3757] lr: 1.0000e-02 eta: 17:02:43 time: 0.1572 data_time: 0.0095 memory: 7124 grad_norm: 5.2575 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.4039 loss: 2.4039 2022/09/06 18:22:05 - mmengine - INFO - Epoch(train) [4][3000/3757] lr: 1.0000e-02 eta: 17:01:56 time: 0.1615 data_time: 0.0094 memory: 7124 grad_norm: 5.2311 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.3524 loss: 2.3524 2022/09/06 18:22:21 - mmengine - INFO - Epoch(train) [4][3100/3757] lr: 1.0000e-02 eta: 17:01:03 time: 0.1547 data_time: 0.0091 memory: 7124 grad_norm: 5.3503 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 2.4046 loss: 2.4046 2022/09/06 18:22:36 - mmengine - INFO - Epoch(train) [4][3200/3757] lr: 1.0000e-02 eta: 17:00:14 time: 0.1596 data_time: 0.0094 memory: 7124 grad_norm: 5.3035 top1_acc: 0.3750 top5_acc: 0.3750 loss_cls: 2.7370 loss: 2.7370 2022/09/06 18:22:52 - mmengine - INFO - Epoch(train) [4][3300/3757] lr: 1.0000e-02 eta: 16:59:26 time: 0.1549 data_time: 0.0085 memory: 7124 grad_norm: 5.2927 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.4640 loss: 2.4640 2022/09/06 18:23:08 - mmengine - INFO - Epoch(train) [4][3400/3757] lr: 1.0000e-02 eta: 16:58:40 time: 0.1535 data_time: 0.0099 memory: 7124 grad_norm: 5.3212 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.3218 loss: 2.3218 2022/09/06 18:23:23 - mmengine - INFO - Epoch(train) [4][3500/3757] lr: 1.0000e-02 eta: 16:57:53 time: 0.1553 data_time: 0.0095 memory: 7124 grad_norm: 5.1275 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.7624 loss: 2.7624 2022/09/06 18:23:39 - mmengine - INFO - Epoch(train) [4][3600/3757] lr: 1.0000e-02 eta: 16:57:08 time: 0.1574 data_time: 0.0099 memory: 7124 grad_norm: 5.3080 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.4094 loss: 2.4094 2022/09/06 18:23:55 - mmengine - INFO - Epoch(train) [4][3700/3757] lr: 1.0000e-02 eta: 16:56:23 time: 0.1629 data_time: 0.0089 memory: 7124 grad_norm: 5.3050 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.7335 loss: 2.7335 2022/09/06 18:24:00 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:24:03 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:24:03 - mmengine - INFO - Epoch(train) [4][3757/3757] lr: 1.0000e-02 eta: 16:56:05 time: 0.1347 data_time: 0.0067 memory: 7124 grad_norm: 5.2924 top1_acc: 0.7143 top5_acc: 0.8571 loss_cls: 2.2362 loss: 2.2362 2022/09/06 18:24:03 - mmengine - INFO - Saving checkpoint at 4 epochs 2022/09/06 18:26:21 - mmengine - INFO - Epoch(val) [4][100/310] eta: 0:03:56 time: 1.1268 data_time: 0.8180 memory: 7627 2022/09/06 18:28:40 - mmengine - INFO - Epoch(val) [4][200/310] eta: 0:02:30 time: 1.3663 data_time: 1.0616 memory: 7627 2022/09/06 18:30:44 - mmengine - INFO - Epoch(val) [4][300/310] eta: 0:00:11 time: 1.1423 data_time: 0.8315 memory: 7627 2022/09/06 18:31:02 - mmengine - INFO - Epoch(val) [4][310/310] acc/top1: 0.5355 acc/top5: 0.7853 acc/mean1: 0.5351 2022/09/06 18:31:02 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_3.pth is removed 2022/09/06 18:31:03 - mmengine - INFO - The best checkpoint with 0.5355 acc/top1 at 4 epoch is saved to best_acc/top1_epoch_4.pth. 2022/09/06 18:31:20 - mmengine - INFO - Epoch(train) [5][100/3757] lr: 1.0000e-02 eta: 16:54:28 time: 0.1564 data_time: 0.0108 memory: 7627 grad_norm: 5.3763 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.3719 loss: 2.3719 2022/09/06 18:31:36 - mmengine - INFO - Epoch(train) [5][200/3757] lr: 1.0000e-02 eta: 16:53:58 time: 0.1878 data_time: 0.0099 memory: 7124 grad_norm: 5.2820 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.5960 loss: 2.5960 2022/09/06 18:31:52 - mmengine - INFO - Epoch(train) [5][300/3757] lr: 1.0000e-02 eta: 16:53:10 time: 0.1523 data_time: 0.0103 memory: 7124 grad_norm: 5.2474 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 2.1915 loss: 2.1915 2022/09/06 18:32:08 - mmengine - INFO - Epoch(train) [5][400/3757] lr: 1.0000e-02 eta: 16:52:28 time: 0.1618 data_time: 0.0094 memory: 7124 grad_norm: 5.2847 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.3439 loss: 2.3439 2022/09/06 18:32:23 - mmengine - INFO - Epoch(train) [5][500/3757] lr: 1.0000e-02 eta: 16:51:44 time: 0.1577 data_time: 0.0105 memory: 7124 grad_norm: 5.1780 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.4652 loss: 2.4652 2022/09/06 18:32:39 - mmengine - INFO - Epoch(train) [5][600/3757] lr: 1.0000e-02 eta: 16:51:00 time: 0.1556 data_time: 0.0125 memory: 7124 grad_norm: 5.2700 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.3912 loss: 2.3912 2022/09/06 18:32:55 - mmengine - INFO - Epoch(train) [5][700/3757] lr: 1.0000e-02 eta: 16:50:16 time: 0.1555 data_time: 0.0110 memory: 7124 grad_norm: 5.1999 top1_acc: 0.1250 top5_acc: 0.3750 loss_cls: 2.6280 loss: 2.6280 2022/09/06 18:33:10 - mmengine - INFO - Epoch(train) [5][800/3757] lr: 1.0000e-02 eta: 16:49:35 time: 0.1608 data_time: 0.0112 memory: 7124 grad_norm: 5.3266 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.4137 loss: 2.4137 2022/09/06 18:33:26 - mmengine - INFO - Epoch(train) [5][900/3757] lr: 1.0000e-02 eta: 16:48:50 time: 0.1577 data_time: 0.0095 memory: 7124 grad_norm: 5.1557 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.7473 loss: 2.7473 2022/09/06 18:33:37 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:33:42 - mmengine - INFO - Epoch(train) [5][1000/3757] lr: 1.0000e-02 eta: 16:48:07 time: 0.1549 data_time: 0.0105 memory: 7124 grad_norm: 5.3931 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.4785 loss: 2.4785 2022/09/06 18:33:57 - mmengine - INFO - Epoch(train) [5][1100/3757] lr: 1.0000e-02 eta: 16:47:23 time: 0.1556 data_time: 0.0099 memory: 7124 grad_norm: 5.3396 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.2837 loss: 2.2837 2022/09/06 18:34:13 - mmengine - INFO - Epoch(train) [5][1200/3757] lr: 1.0000e-02 eta: 16:46:41 time: 0.1537 data_time: 0.0109 memory: 7124 grad_norm: 5.2910 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.5098 loss: 2.5098 2022/09/06 18:34:29 - mmengine - INFO - Epoch(train) [5][1300/3757] lr: 1.0000e-02 eta: 16:46:03 time: 0.1572 data_time: 0.0094 memory: 7124 grad_norm: 5.9663 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 2.7589 loss: 2.7589 2022/09/06 18:34:45 - mmengine - INFO - Epoch(train) [5][1400/3757] lr: 1.0000e-02 eta: 16:45:23 time: 0.1580 data_time: 0.0119 memory: 7124 grad_norm: 5.4677 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 2.2734 loss: 2.2734 2022/09/06 18:35:00 - mmengine - INFO - Epoch(train) [5][1500/3757] lr: 1.0000e-02 eta: 16:44:44 time: 0.1558 data_time: 0.0114 memory: 7124 grad_norm: 5.4274 top1_acc: 0.2500 top5_acc: 0.2500 loss_cls: 2.6040 loss: 2.6040 2022/09/06 18:35:16 - mmengine - INFO - Epoch(train) [5][1600/3757] lr: 1.0000e-02 eta: 16:44:03 time: 0.1557 data_time: 0.0119 memory: 7124 grad_norm: 5.5303 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.4691 loss: 2.4691 2022/09/06 18:35:32 - mmengine - INFO - Epoch(train) [5][1700/3757] lr: 1.0000e-02 eta: 16:43:21 time: 0.1537 data_time: 0.0094 memory: 7124 grad_norm: 5.1969 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.3023 loss: 2.3023 2022/09/06 18:35:47 - mmengine - INFO - Epoch(train) [5][1800/3757] lr: 1.0000e-02 eta: 16:42:42 time: 0.1563 data_time: 0.0101 memory: 7124 grad_norm: 5.2991 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.3948 loss: 2.3948 2022/09/06 18:36:03 - mmengine - INFO - Epoch(train) [5][1900/3757] lr: 1.0000e-02 eta: 16:42:02 time: 0.1571 data_time: 0.0105 memory: 7124 grad_norm: 5.3163 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 2.3510 loss: 2.3510 2022/09/06 18:36:14 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:36:19 - mmengine - INFO - Epoch(train) [5][2000/3757] lr: 1.0000e-02 eta: 16:41:24 time: 0.1577 data_time: 0.0108 memory: 7124 grad_norm: 5.3335 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.3946 loss: 2.3946 2022/09/06 18:36:34 - mmengine - INFO - Epoch(train) [5][2100/3757] lr: 1.0000e-02 eta: 16:40:43 time: 0.1576 data_time: 0.0111 memory: 7124 grad_norm: 5.0621 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.6583 loss: 2.6583 2022/09/06 18:36:50 - mmengine - INFO - Epoch(train) [5][2200/3757] lr: 1.0000e-02 eta: 16:40:11 time: 0.1560 data_time: 0.0104 memory: 7124 grad_norm: 5.3444 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 2.2148 loss: 2.2148 2022/09/06 18:37:06 - mmengine - INFO - Epoch(train) [5][2300/3757] lr: 1.0000e-02 eta: 16:39:34 time: 0.1600 data_time: 0.0101 memory: 7124 grad_norm: 5.0686 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.5019 loss: 2.5019 2022/09/06 18:37:22 - mmengine - INFO - Epoch(train) [5][2400/3757] lr: 1.0000e-02 eta: 16:38:54 time: 0.1558 data_time: 0.0108 memory: 7124 grad_norm: 5.2173 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.6396 loss: 2.6396 2022/09/06 18:37:38 - mmengine - INFO - Epoch(train) [5][2500/3757] lr: 1.0000e-02 eta: 16:38:25 time: 0.1625 data_time: 0.0107 memory: 7124 grad_norm: 5.2351 top1_acc: 0.3750 top5_acc: 0.3750 loss_cls: 2.7359 loss: 2.7359 2022/09/06 18:37:54 - mmengine - INFO - Epoch(train) [5][2600/3757] lr: 1.0000e-02 eta: 16:37:47 time: 0.1555 data_time: 0.0099 memory: 7124 grad_norm: 5.1402 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.1502 loss: 2.1502 2022/09/06 18:38:10 - mmengine - INFO - Epoch(train) [5][2700/3757] lr: 1.0000e-02 eta: 16:37:20 time: 0.1506 data_time: 0.0100 memory: 7124 grad_norm: 5.1585 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.3850 loss: 2.3850 2022/09/06 18:38:26 - mmengine - INFO - Epoch(train) [5][2800/3757] lr: 1.0000e-02 eta: 16:36:46 time: 0.1603 data_time: 0.0117 memory: 7124 grad_norm: 5.3311 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.1785 loss: 2.1785 2022/09/06 18:38:41 - mmengine - INFO - Epoch(train) [5][2900/3757] lr: 1.0000e-02 eta: 16:36:10 time: 0.1557 data_time: 0.0093 memory: 7124 grad_norm: 5.2196 top1_acc: 0.1250 top5_acc: 0.2500 loss_cls: 2.6321 loss: 2.6321 2022/09/06 18:38:53 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:38:57 - mmengine - INFO - Epoch(train) [5][3000/3757] lr: 1.0000e-02 eta: 16:35:33 time: 0.1591 data_time: 0.0102 memory: 7124 grad_norm: 5.3707 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.4955 loss: 2.4955 2022/09/06 18:39:13 - mmengine - INFO - Epoch(train) [5][3100/3757] lr: 1.0000e-02 eta: 16:34:59 time: 0.1561 data_time: 0.0102 memory: 7124 grad_norm: 5.1918 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.3642 loss: 2.3642 2022/09/06 18:39:29 - mmengine - INFO - Epoch(train) [5][3200/3757] lr: 1.0000e-02 eta: 16:34:23 time: 0.1577 data_time: 0.0112 memory: 7124 grad_norm: 5.3130 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.1116 loss: 2.1116 2022/09/06 18:39:45 - mmengine - INFO - Epoch(train) [5][3300/3757] lr: 1.0000e-02 eta: 16:33:49 time: 0.1561 data_time: 0.0109 memory: 7124 grad_norm: 5.3881 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.5105 loss: 2.5105 2022/09/06 18:40:00 - mmengine - INFO - Epoch(train) [5][3400/3757] lr: 1.0000e-02 eta: 16:33:17 time: 0.1554 data_time: 0.0104 memory: 7124 grad_norm: 5.2984 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.4024 loss: 2.4024 2022/09/06 18:40:16 - mmengine - INFO - Epoch(train) [5][3500/3757] lr: 1.0000e-02 eta: 16:32:45 time: 0.1615 data_time: 0.0105 memory: 7124 grad_norm: 5.2741 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.4690 loss: 2.4690 2022/09/06 18:40:32 - mmengine - INFO - Epoch(train) [5][3600/3757] lr: 1.0000e-02 eta: 16:32:07 time: 0.1512 data_time: 0.0105 memory: 7124 grad_norm: 5.3950 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 2.5804 loss: 2.5804 2022/09/06 18:40:48 - mmengine - INFO - Epoch(train) [5][3700/3757] lr: 1.0000e-02 eta: 16:31:33 time: 0.1570 data_time: 0.0114 memory: 7124 grad_norm: 5.4021 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.2542 loss: 2.2542 2022/09/06 18:40:56 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:40:56 - mmengine - INFO - Epoch(train) [5][3757/3757] lr: 1.0000e-02 eta: 16:31:17 time: 0.1343 data_time: 0.0079 memory: 7124 grad_norm: 5.3495 top1_acc: 0.4286 top5_acc: 0.7143 loss_cls: 2.3576 loss: 2.3576 2022/09/06 18:40:56 - mmengine - INFO - Saving checkpoint at 5 epochs 2022/09/06 18:43:15 - mmengine - INFO - Epoch(val) [5][100/310] eta: 0:04:06 time: 1.1728 data_time: 0.8667 memory: 7627 2022/09/06 18:45:33 - mmengine - INFO - Epoch(val) [5][200/310] eta: 0:02:28 time: 1.3466 data_time: 1.0312 memory: 7627 2022/09/06 18:47:36 - mmengine - INFO - Epoch(val) [5][300/310] eta: 0:00:11 time: 1.1122 data_time: 0.8077 memory: 7627 2022/09/06 18:47:52 - mmengine - INFO - Epoch(val) [5][310/310] acc/top1: 0.5421 acc/top5: 0.7964 acc/mean1: 0.5418 2022/09/06 18:47:52 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_4.pth is removed 2022/09/06 18:47:54 - mmengine - INFO - The best checkpoint with 0.5421 acc/top1 at 5 epoch is saved to best_acc/top1_epoch_5.pth. 2022/09/06 18:48:10 - mmengine - INFO - Epoch(train) [6][100/3757] lr: 1.0000e-02 eta: 16:30:00 time: 0.1570 data_time: 0.0110 memory: 7627 grad_norm: 5.1937 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.3685 loss: 2.3685 2022/09/06 18:48:26 - mmengine - INFO - Epoch(train) [6][200/3757] lr: 1.0000e-02 eta: 16:29:32 time: 0.1580 data_time: 0.0105 memory: 7124 grad_norm: 5.1722 top1_acc: 0.1250 top5_acc: 0.7500 loss_cls: 2.3568 loss: 2.3568 2022/09/06 18:48:29 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:48:42 - mmengine - INFO - Epoch(train) [6][300/3757] lr: 1.0000e-02 eta: 16:28:56 time: 0.1562 data_time: 0.0093 memory: 7124 grad_norm: 5.3365 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.3874 loss: 2.3874 2022/09/06 18:48:58 - mmengine - INFO - Epoch(train) [6][400/3757] lr: 1.0000e-02 eta: 16:28:23 time: 0.1579 data_time: 0.0120 memory: 7124 grad_norm: 5.2825 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.2665 loss: 2.2665 2022/09/06 18:49:13 - mmengine - INFO - Epoch(train) [6][500/3757] lr: 1.0000e-02 eta: 16:27:49 time: 0.1540 data_time: 0.0100 memory: 7124 grad_norm: 5.1427 top1_acc: 0.3750 top5_acc: 1.0000 loss_cls: 2.1940 loss: 2.1940 2022/09/06 18:49:29 - mmengine - INFO - Epoch(train) [6][600/3757] lr: 1.0000e-02 eta: 16:27:16 time: 0.1596 data_time: 0.0087 memory: 7124 grad_norm: 5.1927 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.3212 loss: 2.3212 2022/09/06 18:49:45 - mmengine - INFO - Epoch(train) [6][700/3757] lr: 1.0000e-02 eta: 16:26:40 time: 0.1580 data_time: 0.0113 memory: 7124 grad_norm: 5.4204 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.1992 loss: 2.1992 2022/09/06 18:50:00 - mmengine - INFO - Epoch(train) [6][800/3757] lr: 1.0000e-02 eta: 16:26:08 time: 0.1606 data_time: 0.0091 memory: 7124 grad_norm: 5.1789 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.6170 loss: 2.6170 2022/09/06 18:50:16 - mmengine - INFO - Epoch(train) [6][900/3757] lr: 1.0000e-02 eta: 16:25:35 time: 0.1542 data_time: 0.0086 memory: 7124 grad_norm: 5.1687 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.1177 loss: 2.1177 2022/09/06 18:50:32 - mmengine - INFO - Epoch(train) [6][1000/3757] lr: 1.0000e-02 eta: 16:25:01 time: 0.1544 data_time: 0.0088 memory: 7124 grad_norm: 5.3092 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.4713 loss: 2.4713 2022/09/06 18:50:48 - mmengine - INFO - Epoch(train) [6][1100/3757] lr: 1.0000e-02 eta: 16:24:30 time: 0.1614 data_time: 0.0094 memory: 7124 grad_norm: 5.1322 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.4104 loss: 2.4104 2022/09/06 18:51:03 - mmengine - INFO - Epoch(train) [6][1200/3757] lr: 1.0000e-02 eta: 16:23:56 time: 0.1561 data_time: 0.0101 memory: 7124 grad_norm: 5.4709 top1_acc: 0.1250 top5_acc: 0.5000 loss_cls: 2.4576 loss: 2.4576 2022/09/06 18:51:06 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:51:19 - mmengine - INFO - Epoch(train) [6][1300/3757] lr: 1.0000e-02 eta: 16:23:22 time: 0.1554 data_time: 0.0098 memory: 7124 grad_norm: 5.3050 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.5323 loss: 2.5323 2022/09/06 18:51:35 - mmengine - INFO - Epoch(train) [6][1400/3757] lr: 1.0000e-02 eta: 16:22:49 time: 0.1570 data_time: 0.0108 memory: 7124 grad_norm: 5.1215 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 2.4349 loss: 2.4349 2022/09/06 18:51:50 - mmengine - INFO - Epoch(train) [6][1500/3757] lr: 1.0000e-02 eta: 16:22:18 time: 0.1573 data_time: 0.0094 memory: 7124 grad_norm: 5.3948 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.0933 loss: 2.0933 2022/09/06 18:52:06 - mmengine - INFO - Epoch(train) [6][1600/3757] lr: 1.0000e-02 eta: 16:21:43 time: 0.1590 data_time: 0.0086 memory: 7124 grad_norm: 5.0779 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.2933 loss: 2.2933 2022/09/06 18:52:22 - mmengine - INFO - Epoch(train) [6][1700/3757] lr: 1.0000e-02 eta: 16:21:11 time: 0.1548 data_time: 0.0074 memory: 7124 grad_norm: 5.1030 top1_acc: 0.2500 top5_acc: 0.8750 loss_cls: 2.4182 loss: 2.4182 2022/09/06 18:52:37 - mmengine - INFO - Epoch(train) [6][1800/3757] lr: 1.0000e-02 eta: 16:20:40 time: 0.1572 data_time: 0.0102 memory: 7124 grad_norm: 5.3234 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.1666 loss: 2.1666 2022/09/06 18:52:53 - mmengine - INFO - Epoch(train) [6][1900/3757] lr: 1.0000e-02 eta: 16:20:08 time: 0.1534 data_time: 0.0093 memory: 7124 grad_norm: 5.3254 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.2245 loss: 2.2245 2022/09/06 18:53:09 - mmengine - INFO - Epoch(train) [6][2000/3757] lr: 1.0000e-02 eta: 16:19:37 time: 0.1642 data_time: 0.0126 memory: 7124 grad_norm: 5.1469 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.3221 loss: 2.3221 2022/09/06 18:53:24 - mmengine - INFO - Epoch(train) [6][2100/3757] lr: 1.0000e-02 eta: 16:19:06 time: 0.1557 data_time: 0.0097 memory: 7124 grad_norm: 5.2542 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.4914 loss: 2.4914 2022/09/06 18:53:40 - mmengine - INFO - Epoch(train) [6][2200/3757] lr: 1.0000e-02 eta: 16:18:34 time: 0.1549 data_time: 0.0094 memory: 7124 grad_norm: 5.0062 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.5020 loss: 2.5020 2022/09/06 18:53:43 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:53:56 - mmengine - INFO - Epoch(train) [6][2300/3757] lr: 1.0000e-02 eta: 16:18:00 time: 0.1550 data_time: 0.0092 memory: 7124 grad_norm: 5.1287 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.4746 loss: 2.4746 2022/09/06 18:54:11 - mmengine - INFO - Epoch(train) [6][2400/3757] lr: 1.0000e-02 eta: 16:17:30 time: 0.1575 data_time: 0.0097 memory: 7124 grad_norm: 5.3286 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.3816 loss: 2.3816 2022/09/06 18:54:27 - mmengine - INFO - Epoch(train) [6][2500/3757] lr: 1.0000e-02 eta: 16:16:59 time: 0.1622 data_time: 0.0105 memory: 7124 grad_norm: 5.3345 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.2167 loss: 2.2167 2022/09/06 18:54:43 - mmengine - INFO - Epoch(train) [6][2600/3757] lr: 1.0000e-02 eta: 16:16:26 time: 0.1542 data_time: 0.0092 memory: 7124 grad_norm: 5.5636 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.3659 loss: 2.3659 2022/09/06 18:54:59 - mmengine - INFO - Epoch(train) [6][2700/3757] lr: 1.0000e-02 eta: 16:15:59 time: 0.1579 data_time: 0.0091 memory: 7124 grad_norm: 5.1602 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.5592 loss: 2.5592 2022/09/06 18:55:14 - mmengine - INFO - Epoch(train) [6][2800/3757] lr: 1.0000e-02 eta: 16:15:30 time: 0.1542 data_time: 0.0078 memory: 7124 grad_norm: 5.1080 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.4102 loss: 2.4102 2022/09/06 18:55:30 - mmengine - INFO - Epoch(train) [6][2900/3757] lr: 1.0000e-02 eta: 16:14:59 time: 0.1550 data_time: 0.0095 memory: 7124 grad_norm: 5.2492 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.5401 loss: 2.5401 2022/09/06 18:55:46 - mmengine - INFO - Epoch(train) [6][3000/3757] lr: 1.0000e-02 eta: 16:14:29 time: 0.1526 data_time: 0.0082 memory: 7124 grad_norm: 5.2176 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.9641 loss: 1.9641 2022/09/06 18:56:01 - mmengine - INFO - Epoch(train) [6][3100/3757] lr: 1.0000e-02 eta: 16:14:00 time: 0.1561 data_time: 0.0090 memory: 7124 grad_norm: 5.2640 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.5143 loss: 2.5143 2022/09/06 18:56:17 - mmengine - INFO - Epoch(train) [6][3200/3757] lr: 1.0000e-02 eta: 16:13:29 time: 0.1561 data_time: 0.0106 memory: 7124 grad_norm: 4.9702 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.3697 loss: 2.3697 2022/09/06 18:56:20 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:56:33 - mmengine - INFO - Epoch(train) [6][3300/3757] lr: 1.0000e-02 eta: 16:13:00 time: 0.1551 data_time: 0.0101 memory: 7124 grad_norm: 5.4148 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 2.4570 loss: 2.4570 2022/09/06 18:56:49 - mmengine - INFO - Epoch(train) [6][3400/3757] lr: 1.0000e-02 eta: 16:12:30 time: 0.1656 data_time: 0.0141 memory: 7124 grad_norm: 5.2057 top1_acc: 0.1250 top5_acc: 0.6250 loss_cls: 2.3249 loss: 2.3249 2022/09/06 18:57:04 - mmengine - INFO - Epoch(train) [6][3500/3757] lr: 1.0000e-02 eta: 16:12:00 time: 0.1608 data_time: 0.0107 memory: 7124 grad_norm: 5.1084 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.1245 loss: 2.1245 2022/09/06 18:57:20 - mmengine - INFO - Epoch(train) [6][3600/3757] lr: 1.0000e-02 eta: 16:11:30 time: 0.1569 data_time: 0.0083 memory: 7124 grad_norm: 5.1309 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.4007 loss: 2.4007 2022/09/06 18:57:36 - mmengine - INFO - Epoch(train) [6][3700/3757] lr: 1.0000e-02 eta: 16:11:00 time: 0.1564 data_time: 0.0098 memory: 7124 grad_norm: 5.1383 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.2175 loss: 2.2175 2022/09/06 18:57:44 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 18:57:44 - mmengine - INFO - Epoch(train) [6][3757/3757] lr: 1.0000e-02 eta: 16:10:50 time: 0.1343 data_time: 0.0068 memory: 7124 grad_norm: 5.1824 top1_acc: 0.7143 top5_acc: 0.8571 loss_cls: 2.0755 loss: 2.0755 2022/09/06 18:57:44 - mmengine - INFO - Saving checkpoint at 6 epochs 2022/09/06 19:00:03 - mmengine - INFO - Epoch(val) [6][100/310] eta: 0:04:17 time: 1.2255 data_time: 0.9233 memory: 7627 2022/09/06 19:02:19 - mmengine - INFO - Epoch(val) [6][200/310] eta: 0:02:11 time: 1.1974 data_time: 0.8976 memory: 7627 2022/09/06 19:04:25 - mmengine - INFO - Epoch(val) [6][300/310] eta: 0:00:12 time: 1.2632 data_time: 0.9636 memory: 7627 2022/09/06 19:04:42 - mmengine - INFO - Epoch(val) [6][310/310] acc/top1: 0.5554 acc/top5: 0.7987 acc/mean1: 0.5552 2022/09/06 19:04:42 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_5.pth is removed 2022/09/06 19:04:43 - mmengine - INFO - The best checkpoint with 0.5554 acc/top1 at 6 epoch is saved to best_acc/top1_epoch_6.pth. 2022/09/06 19:05:00 - mmengine - INFO - Epoch(train) [7][100/3757] lr: 1.0000e-02 eta: 16:09:44 time: 0.1595 data_time: 0.0096 memory: 7627 grad_norm: 4.8383 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.3516 loss: 2.3516 2022/09/06 19:05:15 - mmengine - INFO - Epoch(train) [7][200/3757] lr: 1.0000e-02 eta: 16:09:15 time: 0.1545 data_time: 0.0087 memory: 7124 grad_norm: 5.1414 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.4290 loss: 2.4290 2022/09/06 19:05:31 - mmengine - INFO - Epoch(train) [7][300/3757] lr: 1.0000e-02 eta: 16:08:48 time: 0.1574 data_time: 0.0103 memory: 7124 grad_norm: 5.1741 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.2199 loss: 2.2199 2022/09/06 19:05:47 - mmengine - INFO - Epoch(train) [7][400/3757] lr: 1.0000e-02 eta: 16:08:21 time: 0.1564 data_time: 0.0093 memory: 7124 grad_norm: 5.3008 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.3966 loss: 2.3966 2022/09/06 19:05:56 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:06:03 - mmengine - INFO - Epoch(train) [7][500/3757] lr: 1.0000e-02 eta: 16:07:54 time: 0.1589 data_time: 0.0112 memory: 7124 grad_norm: 5.2969 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.3721 loss: 2.3721 2022/09/06 19:06:19 - mmengine - INFO - Epoch(train) [7][600/3757] lr: 1.0000e-02 eta: 16:07:28 time: 0.1652 data_time: 0.0124 memory: 7124 grad_norm: 5.0674 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.2283 loss: 2.2283 2022/09/06 19:06:34 - mmengine - INFO - Epoch(train) [7][700/3757] lr: 1.0000e-02 eta: 16:07:00 time: 0.1551 data_time: 0.0102 memory: 7124 grad_norm: 5.5209 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.2256 loss: 2.2256 2022/09/06 19:06:50 - mmengine - INFO - Epoch(train) [7][800/3757] lr: 1.0000e-02 eta: 16:06:36 time: 0.1689 data_time: 0.0100 memory: 7124 grad_norm: 5.3115 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 2.3537 loss: 2.3537 2022/09/06 19:07:06 - mmengine - INFO - Epoch(train) [7][900/3757] lr: 1.0000e-02 eta: 16:06:06 time: 0.1543 data_time: 0.0090 memory: 7124 grad_norm: 5.1450 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.2346 loss: 2.2346 2022/09/06 19:07:22 - mmengine - INFO - Epoch(train) [7][1000/3757] lr: 1.0000e-02 eta: 16:05:40 time: 0.1628 data_time: 0.0083 memory: 7124 grad_norm: 5.4623 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9973 loss: 1.9973 2022/09/06 19:07:38 - mmengine - INFO - Epoch(train) [7][1100/3757] lr: 1.0000e-02 eta: 16:05:13 time: 0.1571 data_time: 0.0106 memory: 7124 grad_norm: 5.2436 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.1225 loss: 2.1225 2022/09/06 19:07:53 - mmengine - INFO - Epoch(train) [7][1200/3757] lr: 1.0000e-02 eta: 16:04:45 time: 0.1564 data_time: 0.0093 memory: 7124 grad_norm: 5.0736 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.4230 loss: 2.4230 2022/09/06 19:08:09 - mmengine - INFO - Epoch(train) [7][1300/3757] lr: 1.0000e-02 eta: 16:04:20 time: 0.1620 data_time: 0.0106 memory: 7124 grad_norm: 4.9539 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 2.4419 loss: 2.4419 2022/09/06 19:08:25 - mmengine - INFO - Epoch(train) [7][1400/3757] lr: 1.0000e-02 eta: 16:03:54 time: 0.1553 data_time: 0.0088 memory: 7124 grad_norm: 5.0438 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.5618 loss: 2.5618 2022/09/06 19:08:34 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:08:41 - mmengine - INFO - Epoch(train) [7][1500/3757] lr: 1.0000e-02 eta: 16:03:27 time: 0.1596 data_time: 0.0090 memory: 7124 grad_norm: 5.2976 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.2686 loss: 2.2686 2022/09/06 19:08:56 - mmengine - INFO - Epoch(train) [7][1600/3757] lr: 1.0000e-02 eta: 16:02:59 time: 0.1546 data_time: 0.0101 memory: 7124 grad_norm: 5.2048 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.4051 loss: 2.4051 2022/09/06 19:09:12 - mmengine - INFO - Epoch(train) [7][1700/3757] lr: 1.0000e-02 eta: 16:02:32 time: 0.1546 data_time: 0.0091 memory: 7124 grad_norm: 5.1836 top1_acc: 0.3750 top5_acc: 0.3750 loss_cls: 2.4066 loss: 2.4066 2022/09/06 19:09:28 - mmengine - INFO - Epoch(train) [7][1800/3757] lr: 1.0000e-02 eta: 16:02:06 time: 0.1554 data_time: 0.0087 memory: 7124 grad_norm: 5.3367 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 2.3627 loss: 2.3627 2022/09/06 19:09:44 - mmengine - INFO - Epoch(train) [7][1900/3757] lr: 1.0000e-02 eta: 16:01:41 time: 0.1547 data_time: 0.0095 memory: 7124 grad_norm: 5.2860 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.5117 loss: 2.5117 2022/09/06 19:10:00 - mmengine - INFO - Epoch(train) [7][2000/3757] lr: 1.0000e-02 eta: 16:01:17 time: 0.1687 data_time: 0.0099 memory: 7124 grad_norm: 5.1846 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.3624 loss: 2.3624 2022/09/06 19:10:15 - mmengine - INFO - Epoch(train) [7][2100/3757] lr: 1.0000e-02 eta: 16:00:53 time: 0.1562 data_time: 0.0101 memory: 7124 grad_norm: 5.1868 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.5098 loss: 2.5098 2022/09/06 19:10:31 - mmengine - INFO - Epoch(train) [7][2200/3757] lr: 1.0000e-02 eta: 16:00:27 time: 0.1578 data_time: 0.0097 memory: 7124 grad_norm: 5.2287 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 2.2134 loss: 2.2134 2022/09/06 19:10:47 - mmengine - INFO - Epoch(train) [7][2300/3757] lr: 1.0000e-02 eta: 16:00:02 time: 0.1571 data_time: 0.0092 memory: 7124 grad_norm: 5.2219 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.5060 loss: 2.5060 2022/09/06 19:11:03 - mmengine - INFO - Epoch(train) [7][2400/3757] lr: 1.0000e-02 eta: 15:59:39 time: 0.1565 data_time: 0.0099 memory: 7124 grad_norm: 5.2652 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.2972 loss: 2.2972 2022/09/06 19:11:12 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:11:19 - mmengine - INFO - Epoch(train) [7][2500/3757] lr: 1.0000e-02 eta: 15:59:14 time: 0.1580 data_time: 0.0108 memory: 7124 grad_norm: 5.6656 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.2567 loss: 2.2567 2022/09/06 19:11:35 - mmengine - INFO - Epoch(train) [7][2600/3757] lr: 1.0000e-02 eta: 15:58:48 time: 0.1576 data_time: 0.0101 memory: 7124 grad_norm: 5.4879 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 2.5320 loss: 2.5320 2022/09/06 19:11:51 - mmengine - INFO - Epoch(train) [7][2700/3757] lr: 1.0000e-02 eta: 15:58:24 time: 0.1600 data_time: 0.0095 memory: 7124 grad_norm: 5.7083 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.3113 loss: 2.3113 2022/09/06 19:12:06 - mmengine - INFO - Epoch(train) [7][2800/3757] lr: 1.0000e-02 eta: 15:57:57 time: 0.1542 data_time: 0.0100 memory: 7124 grad_norm: 5.4580 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.4328 loss: 2.4328 2022/09/06 19:12:22 - mmengine - INFO - Epoch(train) [7][2900/3757] lr: 1.0000e-02 eta: 15:57:29 time: 0.1541 data_time: 0.0099 memory: 7124 grad_norm: 5.6862 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.1284 loss: 2.1284 2022/09/06 19:12:37 - mmengine - INFO - Epoch(train) [7][3000/3757] lr: 1.0000e-02 eta: 15:57:03 time: 0.1536 data_time: 0.0097 memory: 7124 grad_norm: 5.3602 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.3196 loss: 2.3196 2022/09/06 19:12:53 - mmengine - INFO - Epoch(train) [7][3100/3757] lr: 1.0000e-02 eta: 15:56:37 time: 0.1579 data_time: 0.0091 memory: 7124 grad_norm: 5.5620 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 2.5089 loss: 2.5089 2022/09/06 19:13:09 - mmengine - INFO - Epoch(train) [7][3200/3757] lr: 1.0000e-02 eta: 15:56:12 time: 0.1621 data_time: 0.0101 memory: 7124 grad_norm: 5.2586 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.4042 loss: 2.4042 2022/09/06 19:13:25 - mmengine - INFO - Epoch(train) [7][3300/3757] lr: 1.0000e-02 eta: 15:55:45 time: 0.1553 data_time: 0.0086 memory: 7124 grad_norm: 5.1246 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9722 loss: 1.9722 2022/09/06 19:13:40 - mmengine - INFO - Epoch(train) [7][3400/3757] lr: 1.0000e-02 eta: 15:55:19 time: 0.1578 data_time: 0.0083 memory: 7124 grad_norm: 5.4548 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.4865 loss: 2.4865 2022/09/06 19:13:49 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:13:56 - mmengine - INFO - Epoch(train) [7][3500/3757] lr: 1.0000e-02 eta: 15:54:53 time: 0.1584 data_time: 0.0104 memory: 7124 grad_norm: 5.1607 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.1383 loss: 2.1383 2022/09/06 19:14:12 - mmengine - INFO - Epoch(train) [7][3600/3757] lr: 1.0000e-02 eta: 15:54:28 time: 0.1578 data_time: 0.0105 memory: 7124 grad_norm: 5.1304 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.6609 loss: 2.6609 2022/09/06 19:14:27 - mmengine - INFO - Epoch(train) [7][3700/3757] lr: 1.0000e-02 eta: 15:54:02 time: 0.1536 data_time: 0.0094 memory: 7124 grad_norm: 5.2762 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.1393 loss: 2.1393 2022/09/06 19:14:36 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:14:36 - mmengine - INFO - Epoch(train) [7][3757/3757] lr: 1.0000e-02 eta: 15:53:52 time: 0.1347 data_time: 0.0066 memory: 7124 grad_norm: 5.1802 top1_acc: 0.7143 top5_acc: 1.0000 loss_cls: 2.1514 loss: 2.1514 2022/09/06 19:14:36 - mmengine - INFO - Saving checkpoint at 7 epochs 2022/09/06 19:16:54 - mmengine - INFO - Epoch(val) [7][100/310] eta: 0:03:52 time: 1.1061 data_time: 0.8050 memory: 7627 2022/09/06 19:19:11 - mmengine - INFO - Epoch(val) [7][200/310] eta: 0:02:31 time: 1.3781 data_time: 1.0739 memory: 7627 2022/09/06 19:21:15 - mmengine - INFO - Epoch(val) [7][300/310] eta: 0:00:11 time: 1.1422 data_time: 0.8441 memory: 7627 2022/09/06 19:21:34 - mmengine - INFO - Epoch(val) [7][310/310] acc/top1: 0.5690 acc/top5: 0.8100 acc/mean1: 0.5688 2022/09/06 19:21:34 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_6.pth is removed 2022/09/06 19:21:36 - mmengine - INFO - The best checkpoint with 0.5690 acc/top1 at 7 epoch is saved to best_acc/top1_epoch_7.pth. 2022/09/06 19:21:52 - mmengine - INFO - Epoch(train) [8][100/3757] lr: 1.0000e-02 eta: 15:52:54 time: 0.1584 data_time: 0.0101 memory: 7627 grad_norm: 5.4089 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.2866 loss: 2.2866 2022/09/06 19:22:08 - mmengine - INFO - Epoch(train) [8][200/3757] lr: 1.0000e-02 eta: 15:52:25 time: 0.1564 data_time: 0.0088 memory: 7124 grad_norm: 5.1325 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.3303 loss: 2.3303 2022/09/06 19:22:23 - mmengine - INFO - Epoch(train) [8][300/3757] lr: 1.0000e-02 eta: 15:51:57 time: 0.1526 data_time: 0.0088 memory: 7124 grad_norm: 5.2972 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.2396 loss: 2.2396 2022/09/06 19:22:39 - mmengine - INFO - Epoch(train) [8][400/3757] lr: 1.0000e-02 eta: 15:51:34 time: 0.1623 data_time: 0.0096 memory: 7124 grad_norm: 5.5778 top1_acc: 0.1250 top5_acc: 0.7500 loss_cls: 2.3656 loss: 2.3656 2022/09/06 19:22:55 - mmengine - INFO - Epoch(train) [8][500/3757] lr: 1.0000e-02 eta: 15:51:10 time: 0.1565 data_time: 0.0093 memory: 7124 grad_norm: 5.3289 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.9833 loss: 1.9833 2022/09/06 19:23:10 - mmengine - INFO - Epoch(train) [8][600/3757] lr: 1.0000e-02 eta: 15:50:44 time: 0.1568 data_time: 0.0109 memory: 7124 grad_norm: 5.3508 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.1091 loss: 2.1091 2022/09/06 19:23:26 - mmengine - INFO - Epoch(train) [8][700/3757] lr: 1.0000e-02 eta: 15:50:17 time: 0.1560 data_time: 0.0110 memory: 7124 grad_norm: 5.2903 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.2554 loss: 2.2554 2022/09/06 19:23:26 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:23:42 - mmengine - INFO - Epoch(train) [8][800/3757] lr: 1.0000e-02 eta: 15:49:51 time: 0.1533 data_time: 0.0099 memory: 7124 grad_norm: 5.1004 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.1890 loss: 2.1890 2022/09/06 19:23:57 - mmengine - INFO - Epoch(train) [8][900/3757] lr: 1.0000e-02 eta: 15:49:25 time: 0.1548 data_time: 0.0104 memory: 7124 grad_norm: 5.2420 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9021 loss: 1.9021 2022/09/06 19:24:13 - mmengine - INFO - Epoch(train) [8][1000/3757] lr: 1.0000e-02 eta: 15:48:59 time: 0.1542 data_time: 0.0101 memory: 7124 grad_norm: 5.3384 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.2259 loss: 2.2259 2022/09/06 19:24:28 - mmengine - INFO - Epoch(train) [8][1100/3757] lr: 1.0000e-02 eta: 15:48:33 time: 0.1558 data_time: 0.0102 memory: 7124 grad_norm: 5.2910 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.7972 loss: 1.7972 2022/09/06 19:24:44 - mmengine - INFO - Epoch(train) [8][1200/3757] lr: 1.0000e-02 eta: 15:48:08 time: 0.1678 data_time: 0.0098 memory: 7124 grad_norm: 5.2729 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.4603 loss: 2.4603 2022/09/06 19:24:59 - mmengine - INFO - Epoch(train) [8][1300/3757] lr: 1.0000e-02 eta: 15:47:40 time: 0.1547 data_time: 0.0101 memory: 7124 grad_norm: 5.1489 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 2.2963 loss: 2.2963 2022/09/06 19:25:15 - mmengine - INFO - Epoch(train) [8][1400/3757] lr: 1.0000e-02 eta: 15:47:15 time: 0.1506 data_time: 0.0089 memory: 7124 grad_norm: 5.3975 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.1185 loss: 2.1185 2022/09/06 19:25:31 - mmengine - INFO - Epoch(train) [8][1500/3757] lr: 1.0000e-02 eta: 15:46:48 time: 0.1574 data_time: 0.0096 memory: 7124 grad_norm: 5.3987 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.5310 loss: 2.5310 2022/09/06 19:25:46 - mmengine - INFO - Epoch(train) [8][1600/3757] lr: 1.0000e-02 eta: 15:46:22 time: 0.1521 data_time: 0.0092 memory: 7124 grad_norm: 5.2113 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.1788 loss: 2.1788 2022/09/06 19:26:02 - mmengine - INFO - Epoch(train) [8][1700/3757] lr: 1.0000e-02 eta: 15:45:57 time: 0.1627 data_time: 0.0081 memory: 7124 grad_norm: 5.4367 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.5650 loss: 2.5650 2022/09/06 19:26:02 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:26:17 - mmengine - INFO - Epoch(train) [8][1800/3757] lr: 1.0000e-02 eta: 15:45:31 time: 0.1521 data_time: 0.0092 memory: 7124 grad_norm: 5.3725 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 2.2761 loss: 2.2761 2022/09/06 19:26:33 - mmengine - INFO - Epoch(train) [8][1900/3757] lr: 1.0000e-02 eta: 15:45:07 time: 0.1520 data_time: 0.0105 memory: 7124 grad_norm: 5.1604 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 2.3731 loss: 2.3731 2022/09/06 19:26:49 - mmengine - INFO - Epoch(train) [8][2000/3757] lr: 1.0000e-02 eta: 15:44:41 time: 0.1585 data_time: 0.0094 memory: 7124 grad_norm: 5.1041 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.2407 loss: 2.2407 2022/09/06 19:27:04 - mmengine - INFO - Epoch(train) [8][2100/3757] lr: 1.0000e-02 eta: 15:44:14 time: 0.1582 data_time: 0.0116 memory: 7124 grad_norm: 5.3311 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.2514 loss: 2.2514 2022/09/06 19:27:20 - mmengine - INFO - Epoch(train) [8][2200/3757] lr: 1.0000e-02 eta: 15:43:48 time: 0.1544 data_time: 0.0109 memory: 7124 grad_norm: 5.2987 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.5193 loss: 2.5193 2022/09/06 19:27:35 - mmengine - INFO - Epoch(train) [8][2300/3757] lr: 1.0000e-02 eta: 15:43:25 time: 0.1657 data_time: 0.0083 memory: 7124 grad_norm: 5.2108 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.4849 loss: 2.4849 2022/09/06 19:27:51 - mmengine - INFO - Epoch(train) [8][2400/3757] lr: 1.0000e-02 eta: 15:43:00 time: 0.1621 data_time: 0.0113 memory: 7124 grad_norm: 5.1705 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 2.2589 loss: 2.2589 2022/09/06 19:28:07 - mmengine - INFO - Epoch(train) [8][2500/3757] lr: 1.0000e-02 eta: 15:42:36 time: 0.1534 data_time: 0.0078 memory: 7124 grad_norm: 5.1725 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.4109 loss: 2.4109 2022/09/06 19:28:22 - mmengine - INFO - Epoch(train) [8][2600/3757] lr: 1.0000e-02 eta: 15:42:11 time: 0.1575 data_time: 0.0089 memory: 7124 grad_norm: 5.3447 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.4612 loss: 2.4612 2022/09/06 19:28:38 - mmengine - INFO - Epoch(train) [8][2700/3757] lr: 1.0000e-02 eta: 15:41:46 time: 0.1533 data_time: 0.0092 memory: 7124 grad_norm: 5.2078 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.4977 loss: 2.4977 2022/09/06 19:28:38 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:28:54 - mmengine - INFO - Epoch(train) [8][2800/3757] lr: 1.0000e-02 eta: 15:41:22 time: 0.1567 data_time: 0.0094 memory: 7124 grad_norm: 5.1307 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.0645 loss: 2.0645 2022/09/06 19:29:09 - mmengine - INFO - Epoch(train) [8][2900/3757] lr: 1.0000e-02 eta: 15:40:59 time: 0.1555 data_time: 0.0084 memory: 7124 grad_norm: 5.2796 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.1732 loss: 2.1732 2022/09/06 19:29:25 - mmengine - INFO - Epoch(train) [8][3000/3757] lr: 1.0000e-02 eta: 15:40:34 time: 0.1555 data_time: 0.0101 memory: 7124 grad_norm: 5.1166 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.4083 loss: 2.4083 2022/09/06 19:29:40 - mmengine - INFO - Epoch(train) [8][3100/3757] lr: 1.0000e-02 eta: 15:40:09 time: 0.1524 data_time: 0.0086 memory: 7124 grad_norm: 5.1783 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.1515 loss: 2.1515 2022/09/06 19:29:56 - mmengine - INFO - Epoch(train) [8][3200/3757] lr: 1.0000e-02 eta: 15:39:45 time: 0.1540 data_time: 0.0100 memory: 7124 grad_norm: 5.2898 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.4023 loss: 2.4023 2022/09/06 19:30:12 - mmengine - INFO - Epoch(train) [8][3300/3757] lr: 1.0000e-02 eta: 15:39:19 time: 0.1580 data_time: 0.0122 memory: 7124 grad_norm: 5.1329 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.5365 loss: 2.5365 2022/09/06 19:30:27 - mmengine - INFO - Epoch(train) [8][3400/3757] lr: 1.0000e-02 eta: 15:38:53 time: 0.1504 data_time: 0.0078 memory: 7124 grad_norm: 5.0342 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1996 loss: 2.1996 2022/09/06 19:30:43 - mmengine - INFO - Epoch(train) [8][3500/3757] lr: 1.0000e-02 eta: 15:38:32 time: 0.1625 data_time: 0.0092 memory: 7124 grad_norm: 5.1249 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 2.3297 loss: 2.3297 2022/09/06 19:30:58 - mmengine - INFO - Epoch(train) [8][3600/3757] lr: 1.0000e-02 eta: 15:38:07 time: 0.1550 data_time: 0.0095 memory: 7124 grad_norm: 5.2775 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.1399 loss: 2.1399 2022/09/06 19:31:14 - mmengine - INFO - Epoch(train) [8][3700/3757] lr: 1.0000e-02 eta: 15:37:42 time: 0.1518 data_time: 0.0105 memory: 7124 grad_norm: 5.2614 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.2913 loss: 2.2913 2022/09/06 19:31:14 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:31:23 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:31:23 - mmengine - INFO - Epoch(train) [8][3757/3757] lr: 1.0000e-02 eta: 15:37:34 time: 0.1412 data_time: 0.0069 memory: 7124 grad_norm: 5.2703 top1_acc: 0.5714 top5_acc: 0.8571 loss_cls: 2.1755 loss: 2.1755 2022/09/06 19:31:23 - mmengine - INFO - Saving checkpoint at 8 epochs 2022/09/06 19:33:39 - mmengine - INFO - Epoch(val) [8][100/310] eta: 0:04:01 time: 1.1491 data_time: 0.8473 memory: 7627 2022/09/06 19:35:56 - mmengine - INFO - Epoch(val) [8][200/310] eta: 0:02:28 time: 1.3522 data_time: 1.0452 memory: 7627 2022/09/06 19:38:01 - mmengine - INFO - Epoch(val) [8][300/310] eta: 0:00:11 time: 1.1463 data_time: 0.8438 memory: 7627 2022/09/06 19:38:21 - mmengine - INFO - Epoch(val) [8][310/310] acc/top1: 0.5736 acc/top5: 0.8191 acc/mean1: 0.5734 2022/09/06 19:38:21 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_7.pth is removed 2022/09/06 19:38:23 - mmengine - INFO - The best checkpoint with 0.5736 acc/top1 at 8 epoch is saved to best_acc/top1_epoch_8.pth. 2022/09/06 19:38:39 - mmengine - INFO - Epoch(train) [9][100/3757] lr: 1.0000e-02 eta: 15:36:42 time: 0.1549 data_time: 0.0097 memory: 7627 grad_norm: 5.4015 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.3800 loss: 2.3800 2022/09/06 19:38:55 - mmengine - INFO - Epoch(train) [9][200/3757] lr: 1.0000e-02 eta: 15:36:21 time: 0.1575 data_time: 0.0106 memory: 7124 grad_norm: 5.2310 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.2523 loss: 2.2523 2022/09/06 19:39:11 - mmengine - INFO - Epoch(train) [9][300/3757] lr: 1.0000e-02 eta: 15:35:57 time: 0.1545 data_time: 0.0112 memory: 7124 grad_norm: 5.2216 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.1226 loss: 2.1226 2022/09/06 19:39:27 - mmengine - INFO - Epoch(train) [9][400/3757] lr: 1.0000e-02 eta: 15:35:36 time: 0.1535 data_time: 0.0091 memory: 7124 grad_norm: 5.2511 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.2416 loss: 2.2416 2022/09/06 19:39:43 - mmengine - INFO - Epoch(train) [9][500/3757] lr: 1.0000e-02 eta: 15:35:14 time: 0.1632 data_time: 0.0089 memory: 7124 grad_norm: 5.0751 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.1909 loss: 2.1909 2022/09/06 19:39:58 - mmengine - INFO - Epoch(train) [9][600/3757] lr: 1.0000e-02 eta: 15:34:50 time: 0.1528 data_time: 0.0084 memory: 7124 grad_norm: 5.1825 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.3146 loss: 2.3146 2022/09/06 19:40:14 - mmengine - INFO - Epoch(train) [9][700/3757] lr: 1.0000e-02 eta: 15:34:27 time: 0.1561 data_time: 0.0099 memory: 7124 grad_norm: 5.3848 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.2727 loss: 2.2727 2022/09/06 19:40:29 - mmengine - INFO - Epoch(train) [9][800/3757] lr: 1.0000e-02 eta: 15:34:02 time: 0.1526 data_time: 0.0090 memory: 7124 grad_norm: 5.0705 top1_acc: 0.0000 top5_acc: 0.6250 loss_cls: 2.2975 loss: 2.2975 2022/09/06 19:40:45 - mmengine - INFO - Epoch(train) [9][900/3757] lr: 1.0000e-02 eta: 15:33:38 time: 0.1502 data_time: 0.0098 memory: 7124 grad_norm: 5.3167 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 2.2364 loss: 2.2364 2022/09/06 19:40:52 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:41:00 - mmengine - INFO - Epoch(train) [9][1000/3757] lr: 1.0000e-02 eta: 15:33:12 time: 0.1536 data_time: 0.0098 memory: 7124 grad_norm: 5.3127 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1788 loss: 2.1788 2022/09/06 19:41:16 - mmengine - INFO - Epoch(train) [9][1100/3757] lr: 1.0000e-02 eta: 15:32:50 time: 0.1547 data_time: 0.0109 memory: 7124 grad_norm: 5.0643 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 2.2877 loss: 2.2877 2022/09/06 19:41:32 - mmengine - INFO - Epoch(train) [9][1200/3757] lr: 1.0000e-02 eta: 15:32:25 time: 0.1561 data_time: 0.0087 memory: 7124 grad_norm: 5.1909 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0039 loss: 2.0039 2022/09/06 19:41:47 - mmengine - INFO - Epoch(train) [9][1300/3757] lr: 1.0000e-02 eta: 15:32:01 time: 0.1519 data_time: 0.0098 memory: 7124 grad_norm: 5.0095 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.3536 loss: 2.3536 2022/09/06 19:42:03 - mmengine - INFO - Epoch(train) [9][1400/3757] lr: 1.0000e-02 eta: 15:31:39 time: 0.1572 data_time: 0.0102 memory: 7124 grad_norm: 5.2813 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.0520 loss: 2.0520 2022/09/06 19:42:18 - mmengine - INFO - Epoch(train) [9][1500/3757] lr: 1.0000e-02 eta: 15:31:16 time: 0.1541 data_time: 0.0099 memory: 7124 grad_norm: 4.9584 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 2.1926 loss: 2.1926 2022/09/06 19:42:34 - mmengine - INFO - Epoch(train) [9][1600/3757] lr: 1.0000e-02 eta: 15:30:53 time: 0.1548 data_time: 0.0105 memory: 7124 grad_norm: 5.2239 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.1120 loss: 2.1120 2022/09/06 19:42:49 - mmengine - INFO - Epoch(train) [9][1700/3757] lr: 1.0000e-02 eta: 15:30:27 time: 0.1519 data_time: 0.0095 memory: 7124 grad_norm: 5.1519 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.1242 loss: 2.1242 2022/09/06 19:43:05 - mmengine - INFO - Epoch(train) [9][1800/3757] lr: 1.0000e-02 eta: 15:30:04 time: 0.1581 data_time: 0.0099 memory: 7124 grad_norm: 5.1632 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.4136 loss: 2.4136 2022/09/06 19:43:21 - mmengine - INFO - Epoch(train) [9][1900/3757] lr: 1.0000e-02 eta: 15:29:40 time: 0.1542 data_time: 0.0094 memory: 7124 grad_norm: 5.0991 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.1859 loss: 2.1859 2022/09/06 19:43:28 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:43:36 - mmengine - INFO - Epoch(train) [9][2000/3757] lr: 1.0000e-02 eta: 15:29:18 time: 0.1580 data_time: 0.0114 memory: 7124 grad_norm: 5.2597 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.3201 loss: 2.3201 2022/09/06 19:43:52 - mmengine - INFO - Epoch(train) [9][2100/3757] lr: 1.0000e-02 eta: 15:28:56 time: 0.1526 data_time: 0.0083 memory: 7124 grad_norm: 5.1680 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 2.2317 loss: 2.2317 2022/09/06 19:44:08 - mmengine - INFO - Epoch(train) [9][2200/3757] lr: 1.0000e-02 eta: 15:28:36 time: 0.1506 data_time: 0.0102 memory: 7124 grad_norm: 5.1567 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.1270 loss: 2.1270 2022/09/06 19:44:23 - mmengine - INFO - Epoch(train) [9][2300/3757] lr: 1.0000e-02 eta: 15:28:13 time: 0.1589 data_time: 0.0116 memory: 7124 grad_norm: 5.2529 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.3283 loss: 2.3283 2022/09/06 19:44:39 - mmengine - INFO - Epoch(train) [9][2400/3757] lr: 1.0000e-02 eta: 15:27:47 time: 0.1508 data_time: 0.0106 memory: 7124 grad_norm: 5.2769 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 2.2354 loss: 2.2354 2022/09/06 19:44:54 - mmengine - INFO - Epoch(train) [9][2500/3757] lr: 1.0000e-02 eta: 15:27:24 time: 0.1543 data_time: 0.0094 memory: 7124 grad_norm: 5.2661 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.1253 loss: 2.1253 2022/09/06 19:45:10 - mmengine - INFO - Epoch(train) [9][2600/3757] lr: 1.0000e-02 eta: 15:27:04 time: 0.1708 data_time: 0.0085 memory: 7124 grad_norm: 5.2023 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.4957 loss: 2.4957 2022/09/06 19:45:26 - mmengine - INFO - Epoch(train) [9][2700/3757] lr: 1.0000e-02 eta: 15:26:42 time: 0.1490 data_time: 0.0102 memory: 7124 grad_norm: 5.0604 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.3533 loss: 2.3533 2022/09/06 19:45:41 - mmengine - INFO - Epoch(train) [9][2800/3757] lr: 1.0000e-02 eta: 15:26:18 time: 0.1504 data_time: 0.0102 memory: 7124 grad_norm: 4.9602 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.0354 loss: 2.0354 2022/09/06 19:45:57 - mmengine - INFO - Epoch(train) [9][2900/3757] lr: 1.0000e-02 eta: 15:25:56 time: 0.1507 data_time: 0.0097 memory: 7124 grad_norm: 5.1772 top1_acc: 0.1250 top5_acc: 0.7500 loss_cls: 2.2892 loss: 2.2892 2022/09/06 19:46:04 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:46:13 - mmengine - INFO - Epoch(train) [9][3000/3757] lr: 1.0000e-02 eta: 15:25:34 time: 0.1528 data_time: 0.0099 memory: 7124 grad_norm: 5.1744 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.3350 loss: 2.3350 2022/09/06 19:46:28 - mmengine - INFO - Epoch(train) [9][3100/3757] lr: 1.0000e-02 eta: 15:25:12 time: 0.1549 data_time: 0.0093 memory: 7124 grad_norm: 5.0764 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.2928 loss: 2.2928 2022/09/06 19:46:44 - mmengine - INFO - Epoch(train) [9][3200/3757] lr: 1.0000e-02 eta: 15:24:49 time: 0.1601 data_time: 0.0085 memory: 7124 grad_norm: 5.1184 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.9274 loss: 1.9274 2022/09/06 19:46:59 - mmengine - INFO - Epoch(train) [9][3300/3757] lr: 1.0000e-02 eta: 15:24:24 time: 0.1550 data_time: 0.0096 memory: 7124 grad_norm: 5.1582 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9827 loss: 1.9827 2022/09/06 19:47:15 - mmengine - INFO - Epoch(train) [9][3400/3757] lr: 1.0000e-02 eta: 15:24:01 time: 0.1541 data_time: 0.0110 memory: 7124 grad_norm: 5.3400 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9512 loss: 1.9512 2022/09/06 19:47:30 - mmengine - INFO - Epoch(train) [9][3500/3757] lr: 1.0000e-02 eta: 15:23:36 time: 0.1527 data_time: 0.0101 memory: 7124 grad_norm: 5.0641 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.0982 loss: 2.0982 2022/09/06 19:47:46 - mmengine - INFO - Epoch(train) [9][3600/3757] lr: 1.0000e-02 eta: 15:23:13 time: 0.1508 data_time: 0.0090 memory: 7124 grad_norm: 5.1559 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.1061 loss: 2.1061 2022/09/06 19:48:02 - mmengine - INFO - Epoch(train) [9][3700/3757] lr: 1.0000e-02 eta: 15:22:52 time: 0.1615 data_time: 0.0154 memory: 7124 grad_norm: 5.1715 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.2625 loss: 2.2625 2022/09/06 19:48:10 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:48:10 - mmengine - INFO - Epoch(train) [9][3757/3757] lr: 1.0000e-02 eta: 15:22:43 time: 0.1345 data_time: 0.0065 memory: 7124 grad_norm: 4.9565 top1_acc: 0.8571 top5_acc: 1.0000 loss_cls: 2.0784 loss: 2.0784 2022/09/06 19:48:10 - mmengine - INFO - Saving checkpoint at 9 epochs 2022/09/06 19:50:26 - mmengine - INFO - Epoch(val) [9][100/310] eta: 0:03:49 time: 1.0920 data_time: 0.7910 memory: 7627 2022/09/06 19:52:45 - mmengine - INFO - Epoch(val) [9][200/310] eta: 0:02:28 time: 1.3484 data_time: 1.0470 memory: 7627 2022/09/06 19:54:50 - mmengine - INFO - Epoch(val) [9][300/310] eta: 0:00:11 time: 1.1552 data_time: 0.8506 memory: 7627 2022/09/06 19:55:08 - mmengine - INFO - Epoch(val) [9][310/310] acc/top1: 0.5774 acc/top5: 0.8223 acc/mean1: 0.5774 2022/09/06 19:55:08 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_8.pth is removed 2022/09/06 19:55:11 - mmengine - INFO - The best checkpoint with 0.5774 acc/top1 at 9 epoch is saved to best_acc/top1_epoch_9.pth. 2022/09/06 19:55:27 - mmengine - INFO - Epoch(train) [10][100/3757] lr: 1.0000e-02 eta: 15:21:55 time: 0.1538 data_time: 0.0093 memory: 7627 grad_norm: 5.0715 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.2686 loss: 2.2686 2022/09/06 19:55:41 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:55:43 - mmengine - INFO - Epoch(train) [10][200/3757] lr: 1.0000e-02 eta: 15:21:35 time: 0.1581 data_time: 0.0097 memory: 7124 grad_norm: 5.2202 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.9732 loss: 1.9732 2022/09/06 19:55:58 - mmengine - INFO - Epoch(train) [10][300/3757] lr: 1.0000e-02 eta: 15:21:14 time: 0.1548 data_time: 0.0098 memory: 7124 grad_norm: 5.2853 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.4649 loss: 2.4649 2022/09/06 19:56:14 - mmengine - INFO - Epoch(train) [10][400/3757] lr: 1.0000e-02 eta: 15:20:55 time: 0.1585 data_time: 0.0089 memory: 7124 grad_norm: 5.3781 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.2334 loss: 2.2334 2022/09/06 19:56:31 - mmengine - INFO - Epoch(train) [10][500/3757] lr: 1.0000e-02 eta: 15:20:38 time: 0.1726 data_time: 0.0080 memory: 7124 grad_norm: 5.3574 top1_acc: 0.1250 top5_acc: 0.5000 loss_cls: 2.2671 loss: 2.2671 2022/09/06 19:56:46 - mmengine - INFO - Epoch(train) [10][600/3757] lr: 1.0000e-02 eta: 15:20:19 time: 0.1556 data_time: 0.0093 memory: 7124 grad_norm: 5.2523 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0671 loss: 2.0671 2022/09/06 19:57:02 - mmengine - INFO - Epoch(train) [10][700/3757] lr: 1.0000e-02 eta: 15:19:58 time: 0.1565 data_time: 0.0103 memory: 7124 grad_norm: 4.8919 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9465 loss: 1.9465 2022/09/06 19:57:18 - mmengine - INFO - Epoch(train) [10][800/3757] lr: 1.0000e-02 eta: 15:19:36 time: 0.1553 data_time: 0.0100 memory: 7124 grad_norm: 5.2803 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.1808 loss: 2.1808 2022/09/06 19:57:34 - mmengine - INFO - Epoch(train) [10][900/3757] lr: 1.0000e-02 eta: 15:19:18 time: 0.1586 data_time: 0.0094 memory: 7124 grad_norm: 5.2565 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.1431 loss: 2.1431 2022/09/06 19:57:49 - mmengine - INFO - Epoch(train) [10][1000/3757] lr: 1.0000e-02 eta: 15:18:56 time: 0.1551 data_time: 0.0095 memory: 7124 grad_norm: 5.2303 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 2.1238 loss: 2.1238 2022/09/06 19:58:05 - mmengine - INFO - Epoch(train) [10][1100/3757] lr: 1.0000e-02 eta: 15:18:36 time: 0.1604 data_time: 0.0106 memory: 7124 grad_norm: 5.2036 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.1208 loss: 2.1208 2022/09/06 19:58:19 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 19:58:21 - mmengine - INFO - Epoch(train) [10][1200/3757] lr: 1.0000e-02 eta: 15:18:15 time: 0.1555 data_time: 0.0088 memory: 7124 grad_norm: 5.3085 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.9726 loss: 1.9726 2022/09/06 19:58:37 - mmengine - INFO - Epoch(train) [10][1300/3757] lr: 1.0000e-02 eta: 15:17:57 time: 0.1552 data_time: 0.0100 memory: 7124 grad_norm: 5.1383 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.2003 loss: 2.2003 2022/09/06 19:58:52 - mmengine - INFO - Epoch(train) [10][1400/3757] lr: 1.0000e-02 eta: 15:17:34 time: 0.1531 data_time: 0.0093 memory: 7124 grad_norm: 5.0617 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.0456 loss: 2.0456 2022/09/06 19:59:08 - mmengine - INFO - Epoch(train) [10][1500/3757] lr: 1.0000e-02 eta: 15:17:13 time: 0.1569 data_time: 0.0096 memory: 7124 grad_norm: 5.1337 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.3974 loss: 2.3974 2022/09/06 19:59:24 - mmengine - INFO - Epoch(train) [10][1600/3757] lr: 1.0000e-02 eta: 15:16:53 time: 0.1585 data_time: 0.0091 memory: 7124 grad_norm: 4.9731 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.0139 loss: 2.0139 2022/09/06 19:59:40 - mmengine - INFO - Epoch(train) [10][1700/3757] lr: 1.0000e-02 eta: 15:16:35 time: 0.1707 data_time: 0.0109 memory: 7124 grad_norm: 5.0832 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 2.2301 loss: 2.2301 2022/09/06 19:59:55 - mmengine - INFO - Epoch(train) [10][1800/3757] lr: 1.0000e-02 eta: 15:16:14 time: 0.1600 data_time: 0.0092 memory: 7124 grad_norm: 5.1792 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1554 loss: 2.1554 2022/09/06 20:00:11 - mmengine - INFO - Epoch(train) [10][1900/3757] lr: 1.0000e-02 eta: 15:15:53 time: 0.1556 data_time: 0.0092 memory: 7124 grad_norm: 5.0691 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 2.1494 loss: 2.1494 2022/09/06 20:00:27 - mmengine - INFO - Epoch(train) [10][2000/3757] lr: 1.0000e-02 eta: 15:15:33 time: 0.1532 data_time: 0.0090 memory: 7124 grad_norm: 5.2444 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.1355 loss: 2.1355 2022/09/06 20:00:43 - mmengine - INFO - Epoch(train) [10][2100/3757] lr: 1.0000e-02 eta: 15:15:13 time: 0.1526 data_time: 0.0095 memory: 7124 grad_norm: 5.1486 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 2.2929 loss: 2.2929 2022/09/06 20:00:56 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:00:58 - mmengine - INFO - Epoch(train) [10][2200/3757] lr: 1.0000e-02 eta: 15:14:52 time: 0.1519 data_time: 0.0101 memory: 7124 grad_norm: 5.2920 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.8224 loss: 1.8224 2022/09/06 20:01:14 - mmengine - INFO - Epoch(train) [10][2300/3757] lr: 1.0000e-02 eta: 15:14:32 time: 0.1603 data_time: 0.0097 memory: 7124 grad_norm: 5.3849 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1250 loss: 2.1250 2022/09/06 20:01:30 - mmengine - INFO - Epoch(train) [10][2400/3757] lr: 1.0000e-02 eta: 15:14:11 time: 0.1588 data_time: 0.0104 memory: 7124 grad_norm: 5.2266 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 2.0949 loss: 2.0949 2022/09/06 20:01:46 - mmengine - INFO - Epoch(train) [10][2500/3757] lr: 1.0000e-02 eta: 15:13:51 time: 0.1606 data_time: 0.0095 memory: 7124 grad_norm: 5.2888 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.1118 loss: 2.1118 2022/09/06 20:02:02 - mmengine - INFO - Epoch(train) [10][2600/3757] lr: 1.0000e-02 eta: 15:13:34 time: 0.1545 data_time: 0.0104 memory: 7124 grad_norm: 5.3104 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.5222 loss: 2.5222 2022/09/06 20:02:17 - mmengine - INFO - Epoch(train) [10][2700/3757] lr: 1.0000e-02 eta: 15:13:15 time: 0.1522 data_time: 0.0100 memory: 7124 grad_norm: 5.1369 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.0157 loss: 2.0157 2022/09/06 20:02:33 - mmengine - INFO - Epoch(train) [10][2800/3757] lr: 1.0000e-02 eta: 15:12:54 time: 0.1557 data_time: 0.0097 memory: 7124 grad_norm: 5.2045 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.8637 loss: 1.8637 2022/09/06 20:02:49 - mmengine - INFO - Epoch(train) [10][2900/3757] lr: 1.0000e-02 eta: 15:12:33 time: 0.1585 data_time: 0.0125 memory: 7124 grad_norm: 5.3788 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.9643 loss: 1.9643 2022/09/06 20:03:04 - mmengine - INFO - Epoch(train) [10][3000/3757] lr: 1.0000e-02 eta: 15:12:12 time: 0.1540 data_time: 0.0106 memory: 7124 grad_norm: 5.0686 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7844 loss: 1.7844 2022/09/06 20:03:20 - mmengine - INFO - Epoch(train) [10][3100/3757] lr: 1.0000e-02 eta: 15:11:51 time: 0.1555 data_time: 0.0099 memory: 7124 grad_norm: 5.0809 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.8375 loss: 1.8375 2022/09/06 20:03:34 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:03:36 - mmengine - INFO - Epoch(train) [10][3200/3757] lr: 1.0000e-02 eta: 15:11:32 time: 0.1602 data_time: 0.0103 memory: 7124 grad_norm: 5.1604 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.2842 loss: 2.2842 2022/09/06 20:03:51 - mmengine - INFO - Epoch(train) [10][3300/3757] lr: 1.0000e-02 eta: 15:11:10 time: 0.1527 data_time: 0.0097 memory: 7124 grad_norm: 5.0293 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.2679 loss: 2.2679 2022/09/06 20:04:07 - mmengine - INFO - Epoch(train) [10][3400/3757] lr: 1.0000e-02 eta: 15:10:51 time: 0.1538 data_time: 0.0100 memory: 7124 grad_norm: 5.0616 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.1527 loss: 2.1527 2022/09/06 20:04:23 - mmengine - INFO - Epoch(train) [10][3500/3757] lr: 1.0000e-02 eta: 15:10:30 time: 0.1571 data_time: 0.0098 memory: 7124 grad_norm: 5.2175 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.3677 loss: 2.3677 2022/09/06 20:04:39 - mmengine - INFO - Epoch(train) [10][3600/3757] lr: 1.0000e-02 eta: 15:10:10 time: 0.1551 data_time: 0.0095 memory: 7124 grad_norm: 4.9331 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 2.4143 loss: 2.4143 2022/09/06 20:04:54 - mmengine - INFO - Epoch(train) [10][3700/3757] lr: 1.0000e-02 eta: 15:09:50 time: 0.1615 data_time: 0.0084 memory: 7124 grad_norm: 5.3047 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.3444 loss: 2.3444 2022/09/06 20:05:03 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:05:03 - mmengine - INFO - Epoch(train) [10][3757/3757] lr: 1.0000e-02 eta: 15:09:41 time: 0.1352 data_time: 0.0074 memory: 7124 grad_norm: 5.3005 top1_acc: 0.5714 top5_acc: 0.7143 loss_cls: 2.3401 loss: 2.3401 2022/09/06 20:05:03 - mmengine - INFO - Saving checkpoint at 10 epochs 2022/09/06 20:07:22 - mmengine - INFO - Epoch(val) [10][100/310] eta: 0:04:09 time: 1.1904 data_time: 0.8881 memory: 7627 2022/09/06 20:09:39 - mmengine - INFO - Epoch(val) [10][200/310] eta: 0:02:24 time: 1.3099 data_time: 1.0097 memory: 7627 2022/09/06 20:11:43 - mmengine - INFO - Epoch(val) [10][300/310] eta: 0:00:12 time: 1.2020 data_time: 0.9052 memory: 7627 2022/09/06 20:12:02 - mmengine - INFO - Epoch(val) [10][310/310] acc/top1: 0.5917 acc/top5: 0.8283 acc/mean1: 0.5916 2022/09/06 20:12:02 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_9.pth is removed 2022/09/06 20:12:04 - mmengine - INFO - The best checkpoint with 0.5917 acc/top1 at 10 epoch is saved to best_acc/top1_epoch_10.pth. 2022/09/06 20:12:21 - mmengine - INFO - Epoch(train) [11][100/3757] lr: 1.0000e-02 eta: 15:09:00 time: 0.1572 data_time: 0.0091 memory: 7627 grad_norm: 5.0781 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 2.1642 loss: 2.1642 2022/09/06 20:12:36 - mmengine - INFO - Epoch(train) [11][200/3757] lr: 1.0000e-02 eta: 15:08:40 time: 0.1573 data_time: 0.0094 memory: 7124 grad_norm: 4.9346 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.5161 loss: 2.5161 2022/09/06 20:12:52 - mmengine - INFO - Epoch(train) [11][300/3757] lr: 1.0000e-02 eta: 15:08:20 time: 0.1544 data_time: 0.0097 memory: 7124 grad_norm: 5.0612 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 2.0785 loss: 2.0785 2022/09/06 20:13:08 - mmengine - INFO - Epoch(train) [11][400/3757] lr: 1.0000e-02 eta: 15:07:59 time: 0.1561 data_time: 0.0104 memory: 7124 grad_norm: 5.0186 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.0388 loss: 2.0388 2022/09/06 20:13:12 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:13:23 - mmengine - INFO - Epoch(train) [11][500/3757] lr: 1.0000e-02 eta: 15:07:39 time: 0.1582 data_time: 0.0102 memory: 7124 grad_norm: 5.0469 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9263 loss: 1.9263 2022/09/06 20:13:39 - mmengine - INFO - Epoch(train) [11][600/3757] lr: 1.0000e-02 eta: 15:07:21 time: 0.1607 data_time: 0.0089 memory: 7124 grad_norm: 4.9193 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9362 loss: 1.9362 2022/09/06 20:13:55 - mmengine - INFO - Epoch(train) [11][700/3757] lr: 1.0000e-02 eta: 15:07:00 time: 0.1588 data_time: 0.0095 memory: 7124 grad_norm: 5.1286 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9366 loss: 1.9366 2022/09/06 20:14:10 - mmengine - INFO - Epoch(train) [11][800/3757] lr: 1.0000e-02 eta: 15:06:40 time: 0.1608 data_time: 0.0111 memory: 7124 grad_norm: 5.0835 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 2.3670 loss: 2.3670 2022/09/06 20:14:26 - mmengine - INFO - Epoch(train) [11][900/3757] lr: 1.0000e-02 eta: 15:06:20 time: 0.1574 data_time: 0.0099 memory: 7124 grad_norm: 5.1866 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.1252 loss: 2.1252 2022/09/06 20:14:42 - mmengine - INFO - Epoch(train) [11][1000/3757] lr: 1.0000e-02 eta: 15:06:01 time: 0.1533 data_time: 0.0088 memory: 7124 grad_norm: 5.0317 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7252 loss: 1.7252 2022/09/06 20:14:58 - mmengine - INFO - Epoch(train) [11][1100/3757] lr: 1.0000e-02 eta: 15:05:42 time: 0.1559 data_time: 0.0099 memory: 7124 grad_norm: 5.0975 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.0032 loss: 2.0032 2022/09/06 20:15:14 - mmengine - INFO - Epoch(train) [11][1200/3757] lr: 1.0000e-02 eta: 15:05:23 time: 0.1587 data_time: 0.0095 memory: 7124 grad_norm: 5.1426 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.2600 loss: 2.2600 2022/09/06 20:15:29 - mmengine - INFO - Epoch(train) [11][1300/3757] lr: 1.0000e-02 eta: 15:05:03 time: 0.1544 data_time: 0.0101 memory: 7124 grad_norm: 5.1275 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9002 loss: 1.9002 2022/09/06 20:15:45 - mmengine - INFO - Epoch(train) [11][1400/3757] lr: 1.0000e-02 eta: 15:04:43 time: 0.1592 data_time: 0.0106 memory: 7124 grad_norm: 5.2020 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.2740 loss: 2.2740 2022/09/06 20:15:50 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:16:01 - mmengine - INFO - Epoch(train) [11][1500/3757] lr: 1.0000e-02 eta: 15:04:22 time: 0.1534 data_time: 0.0114 memory: 7124 grad_norm: 5.0578 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.8721 loss: 1.8721 2022/09/06 20:16:16 - mmengine - INFO - Epoch(train) [11][1600/3757] lr: 1.0000e-02 eta: 15:04:01 time: 0.1565 data_time: 0.0092 memory: 7124 grad_norm: 5.2393 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.2439 loss: 2.2439 2022/09/06 20:16:32 - mmengine - INFO - Epoch(train) [11][1700/3757] lr: 1.0000e-02 eta: 15:03:41 time: 0.1539 data_time: 0.0095 memory: 7124 grad_norm: 5.1088 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.1923 loss: 2.1923 2022/09/06 20:16:47 - mmengine - INFO - Epoch(train) [11][1800/3757] lr: 1.0000e-02 eta: 15:03:20 time: 0.1591 data_time: 0.0097 memory: 7124 grad_norm: 5.1867 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.2003 loss: 2.2003 2022/09/06 20:17:03 - mmengine - INFO - Epoch(train) [11][1900/3757] lr: 1.0000e-02 eta: 15:03:01 time: 0.1587 data_time: 0.0087 memory: 7124 grad_norm: 5.1081 top1_acc: 0.0000 top5_acc: 0.7500 loss_cls: 2.2203 loss: 2.2203 2022/09/06 20:17:19 - mmengine - INFO - Epoch(train) [11][2000/3757] lr: 1.0000e-02 eta: 15:02:42 time: 0.1561 data_time: 0.0091 memory: 7124 grad_norm: 4.8891 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.2592 loss: 2.2592 2022/09/06 20:17:35 - mmengine - INFO - Epoch(train) [11][2100/3757] lr: 1.0000e-02 eta: 15:02:23 time: 0.1607 data_time: 0.0106 memory: 7124 grad_norm: 5.1749 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1272 loss: 2.1272 2022/09/06 20:17:51 - mmengine - INFO - Epoch(train) [11][2200/3757] lr: 1.0000e-02 eta: 15:02:04 time: 0.1555 data_time: 0.0096 memory: 7124 grad_norm: 5.2413 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8415 loss: 1.8415 2022/09/06 20:18:06 - mmengine - INFO - Epoch(train) [11][2300/3757] lr: 1.0000e-02 eta: 15:01:45 time: 0.1540 data_time: 0.0104 memory: 7124 grad_norm: 5.0058 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.9588 loss: 1.9588 2022/09/06 20:18:22 - mmengine - INFO - Epoch(train) [11][2400/3757] lr: 1.0000e-02 eta: 15:01:27 time: 0.1569 data_time: 0.0089 memory: 7124 grad_norm: 4.9350 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9330 loss: 1.9330 2022/09/06 20:18:27 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:18:38 - mmengine - INFO - Epoch(train) [11][2500/3757] lr: 1.0000e-02 eta: 15:01:09 time: 0.1523 data_time: 0.0098 memory: 7124 grad_norm: 5.3210 top1_acc: 0.0000 top5_acc: 0.5000 loss_cls: 1.9254 loss: 1.9254 2022/09/06 20:18:54 - mmengine - INFO - Epoch(train) [11][2600/3757] lr: 1.0000e-02 eta: 15:00:49 time: 0.1561 data_time: 0.0085 memory: 7124 grad_norm: 5.0119 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.9867 loss: 1.9867 2022/09/06 20:19:10 - mmengine - INFO - Epoch(train) [11][2700/3757] lr: 1.0000e-02 eta: 15:00:29 time: 0.1544 data_time: 0.0108 memory: 7124 grad_norm: 5.0570 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.0461 loss: 2.0461 2022/09/06 20:19:25 - mmengine - INFO - Epoch(train) [11][2800/3757] lr: 1.0000e-02 eta: 15:00:11 time: 0.1623 data_time: 0.0123 memory: 7124 grad_norm: 5.0298 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 2.3466 loss: 2.3466 2022/09/06 20:19:41 - mmengine - INFO - Epoch(train) [11][2900/3757] lr: 1.0000e-02 eta: 14:59:51 time: 0.1572 data_time: 0.0104 memory: 7124 grad_norm: 5.1655 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9724 loss: 1.9724 2022/09/06 20:19:57 - mmengine - INFO - Epoch(train) [11][3000/3757] lr: 1.0000e-02 eta: 14:59:33 time: 0.1532 data_time: 0.0090 memory: 7124 grad_norm: 5.1484 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.1505 loss: 2.1505 2022/09/06 20:20:13 - mmengine - INFO - Epoch(train) [11][3100/3757] lr: 1.0000e-02 eta: 14:59:15 time: 0.1538 data_time: 0.0097 memory: 7124 grad_norm: 4.8654 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.2511 loss: 2.2511 2022/09/06 20:20:29 - mmengine - INFO - Epoch(train) [11][3200/3757] lr: 1.0000e-02 eta: 14:58:56 time: 0.1535 data_time: 0.0105 memory: 7124 grad_norm: 5.2949 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9288 loss: 1.9288 2022/09/06 20:20:44 - mmengine - INFO - Epoch(train) [11][3300/3757] lr: 1.0000e-02 eta: 14:58:35 time: 0.1552 data_time: 0.0106 memory: 7124 grad_norm: 5.0104 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.1867 loss: 2.1867 2022/09/06 20:21:00 - mmengine - INFO - Epoch(train) [11][3400/3757] lr: 1.0000e-02 eta: 14:58:17 time: 0.1578 data_time: 0.0103 memory: 7124 grad_norm: 5.0852 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.0674 loss: 2.0674 2022/09/06 20:21:05 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:21:16 - mmengine - INFO - Epoch(train) [11][3500/3757] lr: 1.0000e-02 eta: 14:57:58 time: 0.1559 data_time: 0.0092 memory: 7124 grad_norm: 5.1253 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1917 loss: 2.1917 2022/09/06 20:21:32 - mmengine - INFO - Epoch(train) [11][3600/3757] lr: 1.0000e-02 eta: 14:57:38 time: 0.1544 data_time: 0.0102 memory: 7124 grad_norm: 5.1263 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.1855 loss: 2.1855 2022/09/06 20:21:47 - mmengine - INFO - Epoch(train) [11][3700/3757] lr: 1.0000e-02 eta: 14:57:19 time: 0.1580 data_time: 0.0107 memory: 7124 grad_norm: 5.1691 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.9859 loss: 1.9859 2022/09/06 20:21:56 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:21:56 - mmengine - INFO - Epoch(train) [11][3757/3757] lr: 1.0000e-02 eta: 14:57:11 time: 0.1371 data_time: 0.0076 memory: 7124 grad_norm: 5.0632 top1_acc: 0.7143 top5_acc: 1.0000 loss_cls: 2.2461 loss: 2.2461 2022/09/06 20:21:56 - mmengine - INFO - Saving checkpoint at 11 epochs 2022/09/06 20:24:16 - mmengine - INFO - Epoch(val) [11][100/310] eta: 0:04:22 time: 1.2518 data_time: 0.9499 memory: 7627 2022/09/06 20:26:29 - mmengine - INFO - Epoch(val) [11][200/310] eta: 0:02:00 time: 1.0993 data_time: 0.7969 memory: 7627 2022/09/06 20:28:36 - mmengine - INFO - Epoch(val) [11][300/310] eta: 0:00:12 time: 1.2633 data_time: 0.9584 memory: 7627 2022/09/06 20:28:56 - mmengine - INFO - Epoch(val) [11][310/310] acc/top1: 0.5995 acc/top5: 0.8383 acc/mean1: 0.5994 2022/09/06 20:28:56 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_10.pth is removed 2022/09/06 20:28:58 - mmengine - INFO - The best checkpoint with 0.5995 acc/top1 at 11 epoch is saved to best_acc/top1_epoch_11.pth. 2022/09/06 20:29:14 - mmengine - INFO - Epoch(train) [12][100/3757] lr: 1.0000e-02 eta: 14:56:33 time: 0.1524 data_time: 0.0103 memory: 7627 grad_norm: 5.0846 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.2824 loss: 2.2824 2022/09/06 20:29:30 - mmengine - INFO - Epoch(train) [12][200/3757] lr: 1.0000e-02 eta: 14:56:13 time: 0.1554 data_time: 0.0099 memory: 7124 grad_norm: 5.1272 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.1565 loss: 2.1565 2022/09/06 20:29:46 - mmengine - INFO - Epoch(train) [12][300/3757] lr: 1.0000e-02 eta: 14:55:54 time: 0.1556 data_time: 0.0095 memory: 7124 grad_norm: 5.2146 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.1685 loss: 2.1685 2022/09/06 20:30:01 - mmengine - INFO - Epoch(train) [12][400/3757] lr: 1.0000e-02 eta: 14:55:34 time: 0.1583 data_time: 0.0096 memory: 7124 grad_norm: 5.0556 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.0576 loss: 2.0576 2022/09/06 20:30:17 - mmengine - INFO - Epoch(train) [12][500/3757] lr: 1.0000e-02 eta: 14:55:15 time: 0.1597 data_time: 0.0104 memory: 7124 grad_norm: 5.0449 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.1244 loss: 2.1244 2022/09/06 20:30:33 - mmengine - INFO - Epoch(train) [12][600/3757] lr: 1.0000e-02 eta: 14:54:55 time: 0.1553 data_time: 0.0098 memory: 7124 grad_norm: 4.9461 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.9790 loss: 1.9790 2022/09/06 20:30:44 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:30:48 - mmengine - INFO - Epoch(train) [12][700/3757] lr: 1.0000e-02 eta: 14:54:35 time: 0.1545 data_time: 0.0118 memory: 7124 grad_norm: 5.1550 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1080 loss: 2.1080 2022/09/06 20:31:04 - mmengine - INFO - Epoch(train) [12][800/3757] lr: 1.0000e-02 eta: 14:54:15 time: 0.1560 data_time: 0.0093 memory: 7124 grad_norm: 5.4224 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.4040 loss: 2.4040 2022/09/06 20:31:20 - mmengine - INFO - Epoch(train) [12][900/3757] lr: 1.0000e-02 eta: 14:53:57 time: 0.1553 data_time: 0.0109 memory: 7124 grad_norm: 5.0966 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9351 loss: 1.9351 2022/09/06 20:31:35 - mmengine - INFO - Epoch(train) [12][1000/3757] lr: 1.0000e-02 eta: 14:53:38 time: 0.1589 data_time: 0.0097 memory: 7124 grad_norm: 5.2471 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.0680 loss: 2.0680 2022/09/06 20:31:51 - mmengine - INFO - Epoch(train) [12][1100/3757] lr: 1.0000e-02 eta: 14:53:18 time: 0.1563 data_time: 0.0104 memory: 7124 grad_norm: 5.3456 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.0782 loss: 2.0782 2022/09/06 20:32:07 - mmengine - INFO - Epoch(train) [12][1200/3757] lr: 1.0000e-02 eta: 14:52:58 time: 0.1554 data_time: 0.0106 memory: 7124 grad_norm: 5.1633 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9768 loss: 1.9768 2022/09/06 20:32:22 - mmengine - INFO - Epoch(train) [12][1300/3757] lr: 1.0000e-02 eta: 14:52:40 time: 0.1565 data_time: 0.0102 memory: 7124 grad_norm: 5.0841 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.0378 loss: 2.0378 2022/09/06 20:32:38 - mmengine - INFO - Epoch(train) [12][1400/3757] lr: 1.0000e-02 eta: 14:52:20 time: 0.1605 data_time: 0.0092 memory: 7124 grad_norm: 5.1057 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.1207 loss: 2.1207 2022/09/06 20:32:54 - mmengine - INFO - Epoch(train) [12][1500/3757] lr: 1.0000e-02 eta: 14:52:01 time: 0.1551 data_time: 0.0107 memory: 7124 grad_norm: 5.2655 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.1185 loss: 2.1185 2022/09/06 20:33:09 - mmengine - INFO - Epoch(train) [12][1600/3757] lr: 1.0000e-02 eta: 14:51:42 time: 0.1531 data_time: 0.0105 memory: 7124 grad_norm: 5.0507 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 2.2014 loss: 2.2014 2022/09/06 20:33:21 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:33:25 - mmengine - INFO - Epoch(train) [12][1700/3757] lr: 1.0000e-02 eta: 14:51:22 time: 0.1572 data_time: 0.0096 memory: 7124 grad_norm: 4.8571 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0102 loss: 2.0102 2022/09/06 20:33:41 - mmengine - INFO - Epoch(train) [12][1800/3757] lr: 1.0000e-02 eta: 14:51:03 time: 0.1540 data_time: 0.0103 memory: 7124 grad_norm: 5.1768 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.1652 loss: 2.1652 2022/09/06 20:33:56 - mmengine - INFO - Epoch(train) [12][1900/3757] lr: 1.0000e-02 eta: 14:50:44 time: 0.1598 data_time: 0.0098 memory: 7124 grad_norm: 5.1160 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1481 loss: 2.1481 2022/09/06 20:34:12 - mmengine - INFO - Epoch(train) [12][2000/3757] lr: 1.0000e-02 eta: 14:50:25 time: 0.1557 data_time: 0.0103 memory: 7124 grad_norm: 4.9645 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.1494 loss: 2.1494 2022/09/06 20:34:28 - mmengine - INFO - Epoch(train) [12][2100/3757] lr: 1.0000e-02 eta: 14:50:06 time: 0.1531 data_time: 0.0108 memory: 7124 grad_norm: 5.0165 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.3751 loss: 2.3751 2022/09/06 20:34:44 - mmengine - INFO - Epoch(train) [12][2200/3757] lr: 1.0000e-02 eta: 14:49:48 time: 0.1570 data_time: 0.0098 memory: 7124 grad_norm: 5.0068 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.1901 loss: 2.1901 2022/09/06 20:35:00 - mmengine - INFO - Epoch(train) [12][2300/3757] lr: 1.0000e-02 eta: 14:49:30 time: 0.1636 data_time: 0.0087 memory: 7124 grad_norm: 5.0852 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.1364 loss: 2.1364 2022/09/06 20:35:15 - mmengine - INFO - Epoch(train) [12][2400/3757] lr: 1.0000e-02 eta: 14:49:10 time: 0.1530 data_time: 0.0095 memory: 7124 grad_norm: 5.0978 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 2.3130 loss: 2.3130 2022/09/06 20:35:31 - mmengine - INFO - Epoch(train) [12][2500/3757] lr: 1.0000e-02 eta: 14:48:53 time: 0.1552 data_time: 0.0101 memory: 7124 grad_norm: 5.0776 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9039 loss: 1.9039 2022/09/06 20:35:47 - mmengine - INFO - Epoch(train) [12][2600/3757] lr: 1.0000e-02 eta: 14:48:34 time: 0.1573 data_time: 0.0109 memory: 7124 grad_norm: 5.2055 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.2230 loss: 2.2230 2022/09/06 20:35:58 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:36:03 - mmengine - INFO - Epoch(train) [12][2700/3757] lr: 1.0000e-02 eta: 14:48:16 time: 0.1566 data_time: 0.0098 memory: 7124 grad_norm: 5.3448 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.9287 loss: 1.9287 2022/09/06 20:36:18 - mmengine - INFO - Epoch(train) [12][2800/3757] lr: 1.0000e-02 eta: 14:47:57 time: 0.1576 data_time: 0.0099 memory: 7124 grad_norm: 5.0399 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.4200 loss: 2.4200 2022/09/06 20:36:34 - mmengine - INFO - Epoch(train) [12][2900/3757] lr: 1.0000e-02 eta: 14:47:38 time: 0.1554 data_time: 0.0096 memory: 7124 grad_norm: 5.1366 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.2997 loss: 2.2997 2022/09/06 20:36:50 - mmengine - INFO - Epoch(train) [12][3000/3757] lr: 1.0000e-02 eta: 14:47:19 time: 0.1650 data_time: 0.0105 memory: 7124 grad_norm: 5.0772 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8504 loss: 1.8504 2022/09/06 20:37:06 - mmengine - INFO - Epoch(train) [12][3100/3757] lr: 1.0000e-02 eta: 14:47:01 time: 0.1545 data_time: 0.0107 memory: 7124 grad_norm: 5.0019 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.0063 loss: 2.0063 2022/09/06 20:37:21 - mmengine - INFO - Epoch(train) [12][3200/3757] lr: 1.0000e-02 eta: 14:46:41 time: 0.1544 data_time: 0.0101 memory: 7124 grad_norm: 5.4816 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.9316 loss: 1.9316 2022/09/06 20:37:37 - mmengine - INFO - Epoch(train) [12][3300/3757] lr: 1.0000e-02 eta: 14:46:22 time: 0.1610 data_time: 0.0088 memory: 7124 grad_norm: 4.9857 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.0737 loss: 2.0737 2022/09/06 20:37:53 - mmengine - INFO - Epoch(train) [12][3400/3757] lr: 1.0000e-02 eta: 14:46:04 time: 0.1526 data_time: 0.0095 memory: 7124 grad_norm: 5.3671 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.1817 loss: 2.1817 2022/09/06 20:38:09 - mmengine - INFO - Epoch(train) [12][3500/3757] lr: 1.0000e-02 eta: 14:45:48 time: 0.1703 data_time: 0.0205 memory: 7124 grad_norm: 4.9963 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.0558 loss: 2.0558 2022/09/06 20:38:24 - mmengine - INFO - Epoch(train) [12][3600/3757] lr: 1.0000e-02 eta: 14:45:28 time: 0.1540 data_time: 0.0094 memory: 7124 grad_norm: 5.1218 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.9368 loss: 1.9368 2022/09/06 20:38:36 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:38:40 - mmengine - INFO - Epoch(train) [12][3700/3757] lr: 1.0000e-02 eta: 14:45:10 time: 0.1552 data_time: 0.0094 memory: 7124 grad_norm: 4.8677 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.9121 loss: 1.9121 2022/09/06 20:38:49 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:38:49 - mmengine - INFO - Epoch(train) [12][3757/3757] lr: 1.0000e-02 eta: 14:45:02 time: 0.1406 data_time: 0.0067 memory: 7124 grad_norm: 5.1219 top1_acc: 0.5714 top5_acc: 1.0000 loss_cls: 2.0714 loss: 2.0714 2022/09/06 20:38:49 - mmengine - INFO - Saving checkpoint at 12 epochs 2022/09/06 20:41:08 - mmengine - INFO - Epoch(val) [12][100/310] eta: 0:04:07 time: 1.1786 data_time: 0.8777 memory: 7627 2022/09/06 20:43:26 - mmengine - INFO - Epoch(val) [12][200/310] eta: 0:02:30 time: 1.3703 data_time: 1.0697 memory: 7627 2022/09/06 20:45:30 - mmengine - INFO - Epoch(val) [12][300/310] eta: 0:00:11 time: 1.1131 data_time: 0.8129 memory: 7627 2022/09/06 20:45:46 - mmengine - INFO - Epoch(val) [12][310/310] acc/top1: 0.6037 acc/top5: 0.8377 acc/mean1: 0.6036 2022/09/06 20:45:46 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_11.pth is removed 2022/09/06 20:45:47 - mmengine - INFO - The best checkpoint with 0.6037 acc/top1 at 12 epoch is saved to best_acc/top1_epoch_12.pth. 2022/09/06 20:46:04 - mmengine - INFO - Epoch(train) [13][100/3757] lr: 1.0000e-02 eta: 14:44:26 time: 0.1595 data_time: 0.0095 memory: 7627 grad_norm: 5.2135 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.9188 loss: 1.9188 2022/09/06 20:46:20 - mmengine - INFO - Epoch(train) [13][200/3757] lr: 1.0000e-02 eta: 14:44:07 time: 0.1590 data_time: 0.0101 memory: 7124 grad_norm: 5.1506 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.9774 loss: 1.9774 2022/09/06 20:46:35 - mmengine - INFO - Epoch(train) [13][300/3757] lr: 1.0000e-02 eta: 14:43:48 time: 0.1593 data_time: 0.0110 memory: 7124 grad_norm: 5.0751 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.2041 loss: 2.2041 2022/09/06 20:46:51 - mmengine - INFO - Epoch(train) [13][400/3757] lr: 1.0000e-02 eta: 14:43:30 time: 0.1572 data_time: 0.0103 memory: 7124 grad_norm: 4.8818 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0765 loss: 2.0765 2022/09/06 20:47:07 - mmengine - INFO - Epoch(train) [13][500/3757] lr: 1.0000e-02 eta: 14:43:12 time: 0.1546 data_time: 0.0103 memory: 7124 grad_norm: 5.0775 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.1929 loss: 2.1929 2022/09/06 20:47:23 - mmengine - INFO - Epoch(train) [13][600/3757] lr: 1.0000e-02 eta: 14:42:56 time: 0.1566 data_time: 0.0095 memory: 7124 grad_norm: 5.1119 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8310 loss: 1.8310 2022/09/06 20:47:39 - mmengine - INFO - Epoch(train) [13][700/3757] lr: 1.0000e-02 eta: 14:42:39 time: 0.1711 data_time: 0.0093 memory: 7124 grad_norm: 4.7410 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.0604 loss: 2.0604 2022/09/06 20:47:55 - mmengine - INFO - Epoch(train) [13][800/3757] lr: 1.0000e-02 eta: 14:42:21 time: 0.1607 data_time: 0.0097 memory: 7124 grad_norm: 5.2481 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.1052 loss: 2.1052 2022/09/06 20:48:10 - mmengine - INFO - Epoch(train) [13][900/3757] lr: 1.0000e-02 eta: 14:42:02 time: 0.1609 data_time: 0.0109 memory: 7124 grad_norm: 5.2201 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 2.3446 loss: 2.3446 2022/09/06 20:48:13 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:48:26 - mmengine - INFO - Epoch(train) [13][1000/3757] lr: 1.0000e-02 eta: 14:41:43 time: 0.1555 data_time: 0.0100 memory: 7124 grad_norm: 4.9691 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.0332 loss: 2.0332 2022/09/06 20:48:42 - mmengine - INFO - Epoch(train) [13][1100/3757] lr: 1.0000e-02 eta: 14:41:26 time: 0.1569 data_time: 0.0104 memory: 7124 grad_norm: 5.2682 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.7981 loss: 1.7981 2022/09/06 20:48:58 - mmengine - INFO - Epoch(train) [13][1200/3757] lr: 1.0000e-02 eta: 14:41:08 time: 0.1558 data_time: 0.0106 memory: 7124 grad_norm: 4.9499 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 2.0434 loss: 2.0434 2022/09/06 20:49:13 - mmengine - INFO - Epoch(train) [13][1300/3757] lr: 1.0000e-02 eta: 14:40:49 time: 0.1563 data_time: 0.0094 memory: 7124 grad_norm: 5.0528 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.9087 loss: 1.9087 2022/09/06 20:49:29 - mmengine - INFO - Epoch(train) [13][1400/3757] lr: 1.0000e-02 eta: 14:40:31 time: 0.1571 data_time: 0.0102 memory: 7124 grad_norm: 5.0451 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9578 loss: 1.9578 2022/09/06 20:49:45 - mmengine - INFO - Epoch(train) [13][1500/3757] lr: 1.0000e-02 eta: 14:40:12 time: 0.1558 data_time: 0.0110 memory: 7124 grad_norm: 5.0277 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.1815 loss: 2.1815 2022/09/06 20:50:01 - mmengine - INFO - Epoch(train) [13][1600/3757] lr: 1.0000e-02 eta: 14:39:54 time: 0.1560 data_time: 0.0099 memory: 7124 grad_norm: 5.4595 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.9201 loss: 1.9201 2022/09/06 20:50:16 - mmengine - INFO - Epoch(train) [13][1700/3757] lr: 1.0000e-02 eta: 14:39:35 time: 0.1560 data_time: 0.0094 memory: 7124 grad_norm: 5.1977 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.9525 loss: 1.9525 2022/09/06 20:50:32 - mmengine - INFO - Epoch(train) [13][1800/3757] lr: 1.0000e-02 eta: 14:39:18 time: 0.1571 data_time: 0.0101 memory: 7124 grad_norm: 5.0082 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.9399 loss: 1.9399 2022/09/06 20:50:48 - mmengine - INFO - Epoch(train) [13][1900/3757] lr: 1.0000e-02 eta: 14:39:00 time: 0.1594 data_time: 0.0104 memory: 7124 grad_norm: 5.1040 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8786 loss: 1.8786 2022/09/06 20:50:51 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:51:04 - mmengine - INFO - Epoch(train) [13][2000/3757] lr: 1.0000e-02 eta: 14:38:42 time: 0.1566 data_time: 0.0088 memory: 7124 grad_norm: 5.0481 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.4409 loss: 2.4409 2022/09/06 20:51:20 - mmengine - INFO - Epoch(train) [13][2100/3757] lr: 1.0000e-02 eta: 14:38:24 time: 0.1564 data_time: 0.0091 memory: 7124 grad_norm: 5.1668 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 2.1674 loss: 2.1674 2022/09/06 20:51:36 - mmengine - INFO - Epoch(train) [13][2200/3757] lr: 1.0000e-02 eta: 14:38:07 time: 0.1565 data_time: 0.0101 memory: 7124 grad_norm: 4.9526 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.0751 loss: 2.0751 2022/09/06 20:51:51 - mmengine - INFO - Epoch(train) [13][2300/3757] lr: 1.0000e-02 eta: 14:37:49 time: 0.1574 data_time: 0.0110 memory: 7124 grad_norm: 5.0988 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.8582 loss: 1.8582 2022/09/06 20:52:07 - mmengine - INFO - Epoch(train) [13][2400/3757] lr: 1.0000e-02 eta: 14:37:30 time: 0.1569 data_time: 0.0100 memory: 7124 grad_norm: 5.1743 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.2085 loss: 2.2085 2022/09/06 20:52:23 - mmengine - INFO - Epoch(train) [13][2500/3757] lr: 1.0000e-02 eta: 14:37:11 time: 0.1530 data_time: 0.0093 memory: 7124 grad_norm: 5.1653 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 2.2062 loss: 2.2062 2022/09/06 20:52:38 - mmengine - INFO - Epoch(train) [13][2600/3757] lr: 1.0000e-02 eta: 14:36:53 time: 0.1569 data_time: 0.0108 memory: 7124 grad_norm: 5.1354 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.0523 loss: 2.0523 2022/09/06 20:52:54 - mmengine - INFO - Epoch(train) [13][2700/3757] lr: 1.0000e-02 eta: 14:36:35 time: 0.1548 data_time: 0.0092 memory: 7124 grad_norm: 5.0863 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.1943 loss: 2.1943 2022/09/06 20:53:10 - mmengine - INFO - Epoch(train) [13][2800/3757] lr: 1.0000e-02 eta: 14:36:17 time: 0.1623 data_time: 0.0104 memory: 7124 grad_norm: 5.1067 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1036 loss: 2.1036 2022/09/06 20:53:26 - mmengine - INFO - Epoch(train) [13][2900/3757] lr: 1.0000e-02 eta: 14:35:58 time: 0.1555 data_time: 0.0087 memory: 7124 grad_norm: 5.0373 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.1365 loss: 2.1365 2022/09/06 20:53:28 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:53:42 - mmengine - INFO - Epoch(train) [13][3000/3757] lr: 1.0000e-02 eta: 14:35:43 time: 0.1625 data_time: 0.0091 memory: 7124 grad_norm: 4.9450 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 2.2688 loss: 2.2688 2022/09/06 20:53:57 - mmengine - INFO - Epoch(train) [13][3100/3757] lr: 1.0000e-02 eta: 14:35:25 time: 0.1553 data_time: 0.0109 memory: 7124 grad_norm: 5.2026 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7744 loss: 1.7744 2022/09/06 20:54:13 - mmengine - INFO - Epoch(train) [13][3200/3757] lr: 1.0000e-02 eta: 14:35:07 time: 0.1549 data_time: 0.0089 memory: 7124 grad_norm: 5.1342 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.0725 loss: 2.0725 2022/09/06 20:54:29 - mmengine - INFO - Epoch(train) [13][3300/3757] lr: 1.0000e-02 eta: 14:34:49 time: 0.1536 data_time: 0.0100 memory: 7124 grad_norm: 5.2525 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 2.1241 loss: 2.1241 2022/09/06 20:54:45 - mmengine - INFO - Epoch(train) [13][3400/3757] lr: 1.0000e-02 eta: 14:34:30 time: 0.1567 data_time: 0.0099 memory: 7124 grad_norm: 5.2333 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8377 loss: 1.8377 2022/09/06 20:55:01 - mmengine - INFO - Epoch(train) [13][3500/3757] lr: 1.0000e-02 eta: 14:34:13 time: 0.1585 data_time: 0.0102 memory: 7124 grad_norm: 5.2287 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9981 loss: 1.9981 2022/09/06 20:55:16 - mmengine - INFO - Epoch(train) [13][3600/3757] lr: 1.0000e-02 eta: 14:33:56 time: 0.1561 data_time: 0.0087 memory: 7124 grad_norm: 5.1879 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.3687 loss: 2.3687 2022/09/06 20:55:32 - mmengine - INFO - Epoch(train) [13][3700/3757] lr: 1.0000e-02 eta: 14:33:38 time: 0.1560 data_time: 0.0099 memory: 7124 grad_norm: 5.0027 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9175 loss: 1.9175 2022/09/06 20:55:41 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 20:55:41 - mmengine - INFO - Epoch(train) [13][3757/3757] lr: 1.0000e-02 eta: 14:33:31 time: 0.1333 data_time: 0.0072 memory: 7124 grad_norm: 4.8410 top1_acc: 0.4286 top5_acc: 0.7143 loss_cls: 2.2878 loss: 2.2878 2022/09/06 20:55:41 - mmengine - INFO - Saving checkpoint at 13 epochs 2022/09/06 20:57:58 - mmengine - INFO - Epoch(val) [13][100/310] eta: 0:03:36 time: 1.0316 data_time: 0.7260 memory: 7627 2022/09/06 21:00:18 - mmengine - INFO - Epoch(val) [13][200/310] eta: 0:02:32 time: 1.3868 data_time: 1.0781 memory: 7627 2022/09/06 21:02:22 - mmengine - INFO - Epoch(val) [13][300/310] eta: 0:00:11 time: 1.1248 data_time: 0.8130 memory: 7627 2022/09/06 21:02:37 - mmengine - INFO - Epoch(val) [13][310/310] acc/top1: 0.6011 acc/top5: 0.8410 acc/mean1: 0.6010 2022/09/06 21:02:55 - mmengine - INFO - Epoch(train) [14][100/3757] lr: 1.0000e-02 eta: 14:33:04 time: 0.1566 data_time: 0.0092 memory: 7627 grad_norm: 5.2503 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9628 loss: 1.9628 2022/09/06 21:03:04 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:03:11 - mmengine - INFO - Epoch(train) [14][200/3757] lr: 1.0000e-02 eta: 14:32:46 time: 0.1566 data_time: 0.0095 memory: 7124 grad_norm: 5.2606 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9411 loss: 1.9411 2022/09/06 21:03:27 - mmengine - INFO - Epoch(train) [14][300/3757] lr: 1.0000e-02 eta: 14:32:27 time: 0.1570 data_time: 0.0090 memory: 7124 grad_norm: 5.2635 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.1151 loss: 2.1151 2022/09/06 21:03:42 - mmengine - INFO - Epoch(train) [14][400/3757] lr: 1.0000e-02 eta: 14:32:10 time: 0.1658 data_time: 0.0087 memory: 7124 grad_norm: 5.0666 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 1.9188 loss: 1.9188 2022/09/06 21:03:58 - mmengine - INFO - Epoch(train) [14][500/3757] lr: 1.0000e-02 eta: 14:31:51 time: 0.1564 data_time: 0.0110 memory: 7124 grad_norm: 5.1076 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7459 loss: 1.7459 2022/09/06 21:04:14 - mmengine - INFO - Epoch(train) [14][600/3757] lr: 1.0000e-02 eta: 14:31:34 time: 0.1553 data_time: 0.0096 memory: 7124 grad_norm: 5.1145 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9211 loss: 1.9211 2022/09/06 21:04:30 - mmengine - INFO - Epoch(train) [14][700/3757] lr: 1.0000e-02 eta: 14:31:16 time: 0.1596 data_time: 0.0096 memory: 7124 grad_norm: 5.2677 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.4201 loss: 2.4201 2022/09/06 21:04:45 - mmengine - INFO - Epoch(train) [14][800/3757] lr: 1.0000e-02 eta: 14:30:58 time: 0.1530 data_time: 0.0096 memory: 7124 grad_norm: 5.0456 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.0825 loss: 2.0825 2022/09/06 21:05:01 - mmengine - INFO - Epoch(train) [14][900/3757] lr: 1.0000e-02 eta: 14:30:41 time: 0.1549 data_time: 0.0089 memory: 7124 grad_norm: 5.0698 top1_acc: 0.2500 top5_acc: 0.8750 loss_cls: 2.0831 loss: 2.0831 2022/09/06 21:05:17 - mmengine - INFO - Epoch(train) [14][1000/3757] lr: 1.0000e-02 eta: 14:30:23 time: 0.1557 data_time: 0.0106 memory: 7124 grad_norm: 5.1730 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 2.2897 loss: 2.2897 2022/09/06 21:05:33 - mmengine - INFO - Epoch(train) [14][1100/3757] lr: 1.0000e-02 eta: 14:30:05 time: 0.1553 data_time: 0.0101 memory: 7124 grad_norm: 5.3012 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9549 loss: 1.9549 2022/09/06 21:05:42 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:05:48 - mmengine - INFO - Epoch(train) [14][1200/3757] lr: 1.0000e-02 eta: 14:29:46 time: 0.1559 data_time: 0.0106 memory: 7124 grad_norm: 5.0421 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.1693 loss: 2.1693 2022/09/06 21:06:04 - mmengine - INFO - Epoch(train) [14][1300/3757] lr: 1.0000e-02 eta: 14:29:29 time: 0.1602 data_time: 0.0096 memory: 7124 grad_norm: 4.8993 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.9650 loss: 1.9650 2022/09/06 21:06:20 - mmengine - INFO - Epoch(train) [14][1400/3757] lr: 1.0000e-02 eta: 14:29:11 time: 0.1556 data_time: 0.0094 memory: 7124 grad_norm: 5.1104 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.2614 loss: 2.2614 2022/09/06 21:06:36 - mmengine - INFO - Epoch(train) [14][1500/3757] lr: 1.0000e-02 eta: 14:28:53 time: 0.1551 data_time: 0.0086 memory: 7124 grad_norm: 5.1851 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 2.2888 loss: 2.2888 2022/09/06 21:06:52 - mmengine - INFO - Epoch(train) [14][1600/3757] lr: 1.0000e-02 eta: 14:28:35 time: 0.1592 data_time: 0.0121 memory: 7124 grad_norm: 4.9288 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9846 loss: 1.9846 2022/09/06 21:07:07 - mmengine - INFO - Epoch(train) [14][1700/3757] lr: 1.0000e-02 eta: 14:28:18 time: 0.1535 data_time: 0.0106 memory: 7124 grad_norm: 5.1893 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.0663 loss: 2.0663 2022/09/06 21:07:23 - mmengine - INFO - Epoch(train) [14][1800/3757] lr: 1.0000e-02 eta: 14:28:01 time: 0.1624 data_time: 0.0101 memory: 7124 grad_norm: 5.2005 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.2835 loss: 2.2835 2022/09/06 21:07:39 - mmengine - INFO - Epoch(train) [14][1900/3757] lr: 1.0000e-02 eta: 14:27:43 time: 0.1559 data_time: 0.0096 memory: 7124 grad_norm: 5.0042 top1_acc: 0.2500 top5_acc: 0.8750 loss_cls: 2.0285 loss: 2.0285 2022/09/06 21:07:55 - mmengine - INFO - Epoch(train) [14][2000/3757] lr: 1.0000e-02 eta: 14:27:27 time: 0.1585 data_time: 0.0090 memory: 7124 grad_norm: 5.0445 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.1747 loss: 2.1747 2022/09/06 21:08:11 - mmengine - INFO - Epoch(train) [14][2100/3757] lr: 1.0000e-02 eta: 14:27:09 time: 0.1605 data_time: 0.0101 memory: 7124 grad_norm: 5.1445 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.0284 loss: 2.0284 2022/09/06 21:08:20 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:08:26 - mmengine - INFO - Epoch(train) [14][2200/3757] lr: 1.0000e-02 eta: 14:26:51 time: 0.1560 data_time: 0.0095 memory: 7124 grad_norm: 5.0261 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7871 loss: 1.7871 2022/09/06 21:08:42 - mmengine - INFO - Epoch(train) [14][2300/3757] lr: 1.0000e-02 eta: 14:26:33 time: 0.1565 data_time: 0.0100 memory: 7124 grad_norm: 5.3367 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.9532 loss: 1.9532 2022/09/06 21:08:58 - mmengine - INFO - Epoch(train) [14][2400/3757] lr: 1.0000e-02 eta: 14:26:15 time: 0.1543 data_time: 0.0102 memory: 7124 grad_norm: 5.0780 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7256 loss: 1.7256 2022/09/06 21:09:14 - mmengine - INFO - Epoch(train) [14][2500/3757] lr: 1.0000e-02 eta: 14:25:58 time: 0.1567 data_time: 0.0100 memory: 7124 grad_norm: 5.0917 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.9688 loss: 1.9688 2022/09/06 21:09:30 - mmengine - INFO - Epoch(train) [14][2600/3757] lr: 1.0000e-02 eta: 14:25:42 time: 0.1554 data_time: 0.0094 memory: 7124 grad_norm: 5.3462 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.0488 loss: 2.0488 2022/09/06 21:09:46 - mmengine - INFO - Epoch(train) [14][2700/3757] lr: 1.0000e-02 eta: 14:25:24 time: 0.1556 data_time: 0.0109 memory: 7124 grad_norm: 4.8765 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9686 loss: 1.9686 2022/09/06 21:10:02 - mmengine - INFO - Epoch(train) [14][2800/3757] lr: 1.0000e-02 eta: 14:25:07 time: 0.1640 data_time: 0.0116 memory: 7124 grad_norm: 5.0797 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.0203 loss: 2.0203 2022/09/06 21:10:17 - mmengine - INFO - Epoch(train) [14][2900/3757] lr: 1.0000e-02 eta: 14:24:49 time: 0.1554 data_time: 0.0105 memory: 7124 grad_norm: 5.0043 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.9536 loss: 1.9536 2022/09/06 21:10:33 - mmengine - INFO - Epoch(train) [14][3000/3757] lr: 1.0000e-02 eta: 14:24:33 time: 0.1552 data_time: 0.0107 memory: 7124 grad_norm: 5.1874 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9036 loss: 1.9036 2022/09/06 21:10:49 - mmengine - INFO - Epoch(train) [14][3100/3757] lr: 1.0000e-02 eta: 14:24:17 time: 0.1586 data_time: 0.0098 memory: 7124 grad_norm: 5.0898 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0807 loss: 2.0807 2022/09/06 21:10:59 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:11:05 - mmengine - INFO - Epoch(train) [14][3200/3757] lr: 1.0000e-02 eta: 14:23:59 time: 0.1549 data_time: 0.0083 memory: 7124 grad_norm: 5.1147 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.0937 loss: 2.0937 2022/09/06 21:11:21 - mmengine - INFO - Epoch(train) [14][3300/3757] lr: 1.0000e-02 eta: 14:23:42 time: 0.1592 data_time: 0.0093 memory: 7124 grad_norm: 4.9457 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9920 loss: 1.9920 2022/09/06 21:11:37 - mmengine - INFO - Epoch(train) [14][3400/3757] lr: 1.0000e-02 eta: 14:23:24 time: 0.1540 data_time: 0.0091 memory: 7124 grad_norm: 4.9741 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.2816 loss: 2.2816 2022/09/06 21:11:52 - mmengine - INFO - Epoch(train) [14][3500/3757] lr: 1.0000e-02 eta: 14:23:06 time: 0.1588 data_time: 0.0101 memory: 7124 grad_norm: 5.1786 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.9960 loss: 1.9960 2022/09/06 21:12:08 - mmengine - INFO - Epoch(train) [14][3600/3757] lr: 1.0000e-02 eta: 14:22:48 time: 0.1548 data_time: 0.0095 memory: 7124 grad_norm: 5.1017 top1_acc: 0.3750 top5_acc: 1.0000 loss_cls: 2.0620 loss: 2.0620 2022/09/06 21:12:24 - mmengine - INFO - Epoch(train) [14][3700/3757] lr: 1.0000e-02 eta: 14:22:33 time: 0.1604 data_time: 0.0105 memory: 7124 grad_norm: 5.1146 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.2844 loss: 2.2844 2022/09/06 21:12:33 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:12:33 - mmengine - INFO - Epoch(train) [14][3757/3757] lr: 1.0000e-02 eta: 14:22:25 time: 0.1346 data_time: 0.0071 memory: 7124 grad_norm: 5.0522 top1_acc: 0.2857 top5_acc: 0.4286 loss_cls: 2.0962 loss: 2.0962 2022/09/06 21:12:33 - mmengine - INFO - Saving checkpoint at 14 epochs 2022/09/06 21:14:53 - mmengine - INFO - Epoch(val) [14][100/310] eta: 0:04:15 time: 1.2166 data_time: 0.9129 memory: 7627 2022/09/06 21:17:09 - mmengine - INFO - Epoch(val) [14][200/310] eta: 0:02:22 time: 1.2946 data_time: 0.9916 memory: 7627 2022/09/06 21:19:13 - mmengine - INFO - Epoch(val) [14][300/310] eta: 0:00:11 time: 1.1327 data_time: 0.8301 memory: 7627 2022/09/06 21:19:33 - mmengine - INFO - Epoch(val) [14][310/310] acc/top1: 0.6126 acc/top5: 0.8391 acc/mean1: 0.6126 2022/09/06 21:19:34 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_12.pth is removed 2022/09/06 21:19:36 - mmengine - INFO - The best checkpoint with 0.6126 acc/top1 at 14 epoch is saved to best_acc/top1_epoch_14.pth. 2022/09/06 21:19:53 - mmengine - INFO - Epoch(train) [15][100/3757] lr: 1.0000e-02 eta: 14:21:54 time: 0.1651 data_time: 0.0088 memory: 7627 grad_norm: 5.2518 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.1538 loss: 2.1538 2022/09/06 21:20:09 - mmengine - INFO - Epoch(train) [15][200/3757] lr: 1.0000e-02 eta: 14:21:37 time: 0.1568 data_time: 0.0099 memory: 7124 grad_norm: 5.1450 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.1305 loss: 2.1305 2022/09/06 21:20:25 - mmengine - INFO - Epoch(train) [15][300/3757] lr: 1.0000e-02 eta: 14:21:20 time: 0.1554 data_time: 0.0096 memory: 7124 grad_norm: 5.1033 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.9800 loss: 1.9800 2022/09/06 21:20:40 - mmengine - INFO - Epoch(train) [15][400/3757] lr: 1.0000e-02 eta: 14:21:02 time: 0.1548 data_time: 0.0098 memory: 7124 grad_norm: 5.1739 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.0303 loss: 2.0303 2022/09/06 21:20:41 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:20:56 - mmengine - INFO - Epoch(train) [15][500/3757] lr: 1.0000e-02 eta: 14:20:47 time: 0.1552 data_time: 0.0107 memory: 7124 grad_norm: 5.0297 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 2.0288 loss: 2.0288 2022/09/06 21:21:12 - mmengine - INFO - Epoch(train) [15][600/3757] lr: 1.0000e-02 eta: 14:20:30 time: 0.1588 data_time: 0.0100 memory: 7124 grad_norm: 4.9972 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.7920 loss: 1.7920 2022/09/06 21:21:28 - mmengine - INFO - Epoch(train) [15][700/3757] lr: 1.0000e-02 eta: 14:20:14 time: 0.1533 data_time: 0.0104 memory: 7124 grad_norm: 5.0082 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8582 loss: 1.8582 2022/09/06 21:21:44 - mmengine - INFO - Epoch(train) [15][800/3757] lr: 1.0000e-02 eta: 14:19:57 time: 0.1616 data_time: 0.0084 memory: 7124 grad_norm: 5.1046 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 2.1545 loss: 2.1545 2022/09/06 21:22:00 - mmengine - INFO - Epoch(train) [15][900/3757] lr: 1.0000e-02 eta: 14:19:38 time: 0.1533 data_time: 0.0095 memory: 7124 grad_norm: 5.2701 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.7798 loss: 1.7798 2022/09/06 21:22:16 - mmengine - INFO - Epoch(train) [15][1000/3757] lr: 1.0000e-02 eta: 14:19:22 time: 0.1598 data_time: 0.0097 memory: 7124 grad_norm: 5.1765 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0117 loss: 2.0117 2022/09/06 21:22:32 - mmengine - INFO - Epoch(train) [15][1100/3757] lr: 1.0000e-02 eta: 14:19:05 time: 0.1579 data_time: 0.0094 memory: 7124 grad_norm: 4.9329 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.1311 loss: 2.1311 2022/09/06 21:22:47 - mmengine - INFO - Epoch(train) [15][1200/3757] lr: 1.0000e-02 eta: 14:18:47 time: 0.1579 data_time: 0.0089 memory: 7124 grad_norm: 5.0835 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.0755 loss: 2.0755 2022/09/06 21:23:03 - mmengine - INFO - Epoch(train) [15][1300/3757] lr: 1.0000e-02 eta: 14:18:30 time: 0.1636 data_time: 0.0106 memory: 7124 grad_norm: 4.9362 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.1430 loss: 2.1430 2022/09/06 21:23:19 - mmengine - INFO - Epoch(train) [15][1400/3757] lr: 1.0000e-02 eta: 14:18:13 time: 0.1578 data_time: 0.0102 memory: 7124 grad_norm: 5.1134 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6333 loss: 1.6333 2022/09/06 21:23:19 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:23:35 - mmengine - INFO - Epoch(train) [15][1500/3757] lr: 1.0000e-02 eta: 14:17:56 time: 0.1554 data_time: 0.0109 memory: 7124 grad_norm: 5.3505 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 2.1416 loss: 2.1416 2022/09/06 21:23:51 - mmengine - INFO - Epoch(train) [15][1600/3757] lr: 1.0000e-02 eta: 14:17:39 time: 0.1558 data_time: 0.0100 memory: 7124 grad_norm: 5.2913 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.6902 loss: 1.6902 2022/09/06 21:24:07 - mmengine - INFO - Epoch(train) [15][1700/3757] lr: 1.0000e-02 eta: 14:17:22 time: 0.1567 data_time: 0.0093 memory: 7124 grad_norm: 5.0491 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.1501 loss: 2.1501 2022/09/06 21:24:23 - mmengine - INFO - Epoch(train) [15][1800/3757] lr: 1.0000e-02 eta: 14:17:06 time: 0.1646 data_time: 0.0115 memory: 7124 grad_norm: 5.2335 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.3928 loss: 2.3928 2022/09/06 21:24:38 - mmengine - INFO - Epoch(train) [15][1900/3757] lr: 1.0000e-02 eta: 14:16:48 time: 0.1593 data_time: 0.0105 memory: 7124 grad_norm: 4.9130 top1_acc: 0.1250 top5_acc: 0.6250 loss_cls: 2.1016 loss: 2.1016 2022/09/06 21:24:54 - mmengine - INFO - Epoch(train) [15][2000/3757] lr: 1.0000e-02 eta: 14:16:31 time: 0.1611 data_time: 0.0086 memory: 7124 grad_norm: 5.0818 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0913 loss: 2.0913 2022/09/06 21:25:10 - mmengine - INFO - Epoch(train) [15][2100/3757] lr: 1.0000e-02 eta: 14:16:13 time: 0.1582 data_time: 0.0096 memory: 7124 grad_norm: 4.9057 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.2431 loss: 2.2431 2022/09/06 21:25:26 - mmengine - INFO - Epoch(train) [15][2200/3757] lr: 1.0000e-02 eta: 14:15:56 time: 0.1576 data_time: 0.0095 memory: 7124 grad_norm: 4.9600 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.0547 loss: 2.0547 2022/09/06 21:25:42 - mmengine - INFO - Epoch(train) [15][2300/3757] lr: 1.0000e-02 eta: 14:15:39 time: 0.1624 data_time: 0.0098 memory: 7124 grad_norm: 5.1356 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.4890 loss: 2.4890 2022/09/06 21:25:57 - mmengine - INFO - Epoch(train) [15][2400/3757] lr: 1.0000e-02 eta: 14:15:21 time: 0.1565 data_time: 0.0094 memory: 7124 grad_norm: 5.1759 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.1860 loss: 2.1860 2022/09/06 21:25:58 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:26:13 - mmengine - INFO - Epoch(train) [15][2500/3757] lr: 1.0000e-02 eta: 14:15:05 time: 0.1554 data_time: 0.0090 memory: 7124 grad_norm: 5.1068 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.4919 loss: 2.4919 2022/09/06 21:26:29 - mmengine - INFO - Epoch(train) [15][2600/3757] lr: 1.0000e-02 eta: 14:14:49 time: 0.1586 data_time: 0.0092 memory: 7124 grad_norm: 5.2267 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.1770 loss: 2.1770 2022/09/06 21:26:45 - mmengine - INFO - Epoch(train) [15][2700/3757] lr: 1.0000e-02 eta: 14:14:31 time: 0.1553 data_time: 0.0117 memory: 7124 grad_norm: 5.0021 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.9837 loss: 1.9837 2022/09/06 21:27:01 - mmengine - INFO - Epoch(train) [15][2800/3757] lr: 1.0000e-02 eta: 14:14:12 time: 0.1558 data_time: 0.0098 memory: 7124 grad_norm: 5.1265 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.9858 loss: 1.9858 2022/09/06 21:27:17 - mmengine - INFO - Epoch(train) [15][2900/3757] lr: 1.0000e-02 eta: 14:13:56 time: 0.1551 data_time: 0.0101 memory: 7124 grad_norm: 5.0148 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.9366 loss: 1.9366 2022/09/06 21:27:33 - mmengine - INFO - Epoch(train) [15][3000/3757] lr: 1.0000e-02 eta: 14:13:39 time: 0.1548 data_time: 0.0099 memory: 7124 grad_norm: 4.9658 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 2.2574 loss: 2.2574 2022/09/06 21:27:49 - mmengine - INFO - Epoch(train) [15][3100/3757] lr: 1.0000e-02 eta: 14:13:25 time: 0.1558 data_time: 0.0093 memory: 7124 grad_norm: 4.9715 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.0274 loss: 2.0274 2022/09/06 21:28:05 - mmengine - INFO - Epoch(train) [15][3200/3757] lr: 1.0000e-02 eta: 14:13:10 time: 0.1592 data_time: 0.0100 memory: 7124 grad_norm: 4.8593 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9001 loss: 1.9001 2022/09/06 21:28:21 - mmengine - INFO - Epoch(train) [15][3300/3757] lr: 1.0000e-02 eta: 14:12:52 time: 0.1592 data_time: 0.0090 memory: 7124 grad_norm: 5.2153 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.1343 loss: 2.1343 2022/09/06 21:28:37 - mmengine - INFO - Epoch(train) [15][3400/3757] lr: 1.0000e-02 eta: 14:12:35 time: 0.1556 data_time: 0.0097 memory: 7124 grad_norm: 5.1783 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.2871 loss: 2.2871 2022/09/06 21:28:37 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:28:52 - mmengine - INFO - Epoch(train) [15][3500/3757] lr: 1.0000e-02 eta: 14:12:18 time: 0.1593 data_time: 0.0100 memory: 7124 grad_norm: 5.1754 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7989 loss: 1.7989 2022/09/06 21:29:08 - mmengine - INFO - Epoch(train) [15][3600/3757] lr: 1.0000e-02 eta: 14:12:01 time: 0.1575 data_time: 0.0102 memory: 7124 grad_norm: 4.8945 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9389 loss: 1.9389 2022/09/06 21:29:24 - mmengine - INFO - Epoch(train) [15][3700/3757] lr: 1.0000e-02 eta: 14:11:44 time: 0.1620 data_time: 0.0102 memory: 7124 grad_norm: 5.0591 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.3533 loss: 2.3533 2022/09/06 21:29:33 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:29:33 - mmengine - INFO - Epoch(train) [15][3757/3757] lr: 1.0000e-02 eta: 14:11:36 time: 0.1383 data_time: 0.0071 memory: 7124 grad_norm: 4.9841 top1_acc: 0.5714 top5_acc: 0.8571 loss_cls: 2.0194 loss: 2.0194 2022/09/06 21:29:33 - mmengine - INFO - Saving checkpoint at 15 epochs 2022/09/06 21:31:52 - mmengine - INFO - Epoch(val) [15][100/310] eta: 0:04:08 time: 1.1838 data_time: 0.8693 memory: 7627 2022/09/06 21:34:08 - mmengine - INFO - Epoch(val) [15][200/310] eta: 0:02:15 time: 1.2321 data_time: 0.9257 memory: 7627 2022/09/06 21:36:13 - mmengine - INFO - Epoch(val) [15][300/310] eta: 0:00:12 time: 1.2463 data_time: 0.9436 memory: 7627 2022/09/06 21:36:31 - mmengine - INFO - Epoch(val) [15][310/310] acc/top1: 0.6178 acc/top5: 0.8485 acc/mean1: 0.6176 2022/09/06 21:36:31 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_14.pth is removed 2022/09/06 21:36:33 - mmengine - INFO - The best checkpoint with 0.6178 acc/top1 at 15 epoch is saved to best_acc/top1_epoch_15.pth. 2022/09/06 21:36:49 - mmengine - INFO - Epoch(train) [16][100/3757] lr: 1.0000e-02 eta: 14:11:04 time: 0.1549 data_time: 0.0113 memory: 7627 grad_norm: 5.2298 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.1129 loss: 2.1129 2022/09/06 21:37:05 - mmengine - INFO - Epoch(train) [16][200/3757] lr: 1.0000e-02 eta: 14:10:47 time: 0.1600 data_time: 0.0124 memory: 7124 grad_norm: 5.1758 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.9595 loss: 1.9595 2022/09/06 21:37:20 - mmengine - INFO - Epoch(train) [16][300/3757] lr: 1.0000e-02 eta: 14:10:28 time: 0.1556 data_time: 0.0097 memory: 7124 grad_norm: 5.1281 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.0215 loss: 2.0215 2022/09/06 21:37:36 - mmengine - INFO - Epoch(train) [16][400/3757] lr: 1.0000e-02 eta: 14:10:10 time: 0.1540 data_time: 0.0098 memory: 7124 grad_norm: 5.0338 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9184 loss: 1.9184 2022/09/06 21:37:52 - mmengine - INFO - Epoch(train) [16][500/3757] lr: 1.0000e-02 eta: 14:09:52 time: 0.1572 data_time: 0.0096 memory: 7124 grad_norm: 5.2999 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.9876 loss: 1.9876 2022/09/06 21:38:08 - mmengine - INFO - Epoch(train) [16][600/3757] lr: 1.0000e-02 eta: 14:09:35 time: 0.1567 data_time: 0.0099 memory: 7124 grad_norm: 5.0732 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.2212 loss: 2.2212 2022/09/06 21:38:15 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:38:23 - mmengine - INFO - Epoch(train) [16][700/3757] lr: 1.0000e-02 eta: 14:09:18 time: 0.1606 data_time: 0.0097 memory: 7124 grad_norm: 5.2267 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.0972 loss: 2.0972 2022/09/06 21:38:39 - mmengine - INFO - Epoch(train) [16][800/3757] lr: 1.0000e-02 eta: 14:09:02 time: 0.1557 data_time: 0.0112 memory: 7124 grad_norm: 5.0805 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.0219 loss: 2.0219 2022/09/06 21:38:55 - mmengine - INFO - Epoch(train) [16][900/3757] lr: 1.0000e-02 eta: 14:08:44 time: 0.1599 data_time: 0.0105 memory: 7124 grad_norm: 5.0361 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.0886 loss: 2.0886 2022/09/06 21:39:11 - mmengine - INFO - Epoch(train) [16][1000/3757] lr: 1.0000e-02 eta: 14:08:26 time: 0.1529 data_time: 0.0108 memory: 7124 grad_norm: 4.8973 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.9875 loss: 1.9875 2022/09/06 21:39:26 - mmengine - INFO - Epoch(train) [16][1100/3757] lr: 1.0000e-02 eta: 14:08:08 time: 0.1556 data_time: 0.0099 memory: 7124 grad_norm: 4.9728 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.3530 loss: 2.3530 2022/09/06 21:39:42 - mmengine - INFO - Epoch(train) [16][1200/3757] lr: 1.0000e-02 eta: 14:07:51 time: 0.1537 data_time: 0.0085 memory: 7124 grad_norm: 5.0037 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0857 loss: 2.0857 2022/09/06 21:39:58 - mmengine - INFO - Epoch(train) [16][1300/3757] lr: 1.0000e-02 eta: 14:07:33 time: 0.1548 data_time: 0.0095 memory: 7124 grad_norm: 5.2010 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0685 loss: 2.0685 2022/09/06 21:40:14 - mmengine - INFO - Epoch(train) [16][1400/3757] lr: 1.0000e-02 eta: 14:07:16 time: 0.1614 data_time: 0.0108 memory: 7124 grad_norm: 5.1430 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8291 loss: 1.8291 2022/09/06 21:40:29 - mmengine - INFO - Epoch(train) [16][1500/3757] lr: 1.0000e-02 eta: 14:06:58 time: 0.1553 data_time: 0.0096 memory: 7124 grad_norm: 5.2361 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.8494 loss: 1.8494 2022/09/06 21:40:45 - mmengine - INFO - Epoch(train) [16][1600/3757] lr: 1.0000e-02 eta: 14:06:40 time: 0.1594 data_time: 0.0105 memory: 7124 grad_norm: 5.0803 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.2289 loss: 2.2289 2022/09/06 21:40:52 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:41:01 - mmengine - INFO - Epoch(train) [16][1700/3757] lr: 1.0000e-02 eta: 14:06:22 time: 0.1541 data_time: 0.0097 memory: 7124 grad_norm: 5.2639 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.0854 loss: 2.0854 2022/09/06 21:41:16 - mmengine - INFO - Epoch(train) [16][1800/3757] lr: 1.0000e-02 eta: 14:06:05 time: 0.1574 data_time: 0.0090 memory: 7124 grad_norm: 5.2761 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.3944 loss: 2.3944 2022/09/06 21:41:32 - mmengine - INFO - Epoch(train) [16][1900/3757] lr: 1.0000e-02 eta: 14:05:47 time: 0.1531 data_time: 0.0095 memory: 7124 grad_norm: 5.3045 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.0617 loss: 2.0617 2022/09/06 21:41:48 - mmengine - INFO - Epoch(train) [16][2000/3757] lr: 1.0000e-02 eta: 14:05:29 time: 0.1544 data_time: 0.0102 memory: 7124 grad_norm: 4.9394 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.0285 loss: 2.0285 2022/09/06 21:42:03 - mmengine - INFO - Epoch(train) [16][2100/3757] lr: 1.0000e-02 eta: 14:05:12 time: 0.1622 data_time: 0.0101 memory: 7124 grad_norm: 5.1835 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9398 loss: 1.9398 2022/09/06 21:42:19 - mmengine - INFO - Epoch(train) [16][2200/3757] lr: 1.0000e-02 eta: 14:04:54 time: 0.1548 data_time: 0.0090 memory: 7124 grad_norm: 5.1225 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9472 loss: 1.9472 2022/09/06 21:42:35 - mmengine - INFO - Epoch(train) [16][2300/3757] lr: 1.0000e-02 eta: 14:04:37 time: 0.1635 data_time: 0.0093 memory: 7124 grad_norm: 5.2533 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.6365 loss: 1.6365 2022/09/06 21:42:51 - mmengine - INFO - Epoch(train) [16][2400/3757] lr: 1.0000e-02 eta: 14:04:19 time: 0.1552 data_time: 0.0106 memory: 7124 grad_norm: 4.9615 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.8105 loss: 1.8105 2022/09/06 21:43:07 - mmengine - INFO - Epoch(train) [16][2500/3757] lr: 1.0000e-02 eta: 14:04:03 time: 0.1527 data_time: 0.0096 memory: 7124 grad_norm: 5.2645 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.9822 loss: 1.9822 2022/09/06 21:43:22 - mmengine - INFO - Epoch(train) [16][2600/3757] lr: 1.0000e-02 eta: 14:03:45 time: 0.1565 data_time: 0.0091 memory: 7124 grad_norm: 5.2922 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0064 loss: 2.0064 2022/09/06 21:43:29 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:43:38 - mmengine - INFO - Epoch(train) [16][2700/3757] lr: 1.0000e-02 eta: 14:03:28 time: 0.1561 data_time: 0.0104 memory: 7124 grad_norm: 5.0532 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.1098 loss: 2.1098 2022/09/06 21:43:54 - mmengine - INFO - Epoch(train) [16][2800/3757] lr: 1.0000e-02 eta: 14:03:10 time: 0.1655 data_time: 0.0098 memory: 7124 grad_norm: 4.9594 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.0313 loss: 2.0313 2022/09/06 21:44:09 - mmengine - INFO - Epoch(train) [16][2900/3757] lr: 1.0000e-02 eta: 14:02:52 time: 0.1538 data_time: 0.0093 memory: 7124 grad_norm: 4.9397 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9835 loss: 1.9835 2022/09/06 21:44:25 - mmengine - INFO - Epoch(train) [16][3000/3757] lr: 1.0000e-02 eta: 14:02:35 time: 0.1587 data_time: 0.0106 memory: 7124 grad_norm: 5.0439 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.9878 loss: 1.9878 2022/09/06 21:44:41 - mmengine - INFO - Epoch(train) [16][3100/3757] lr: 1.0000e-02 eta: 14:02:17 time: 0.1546 data_time: 0.0090 memory: 7124 grad_norm: 5.1478 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1597 loss: 2.1597 2022/09/06 21:44:57 - mmengine - INFO - Epoch(train) [16][3200/3757] lr: 1.0000e-02 eta: 14:02:00 time: 0.1483 data_time: 0.0098 memory: 7124 grad_norm: 5.2361 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.9769 loss: 1.9769 2022/09/06 21:45:12 - mmengine - INFO - Epoch(train) [16][3300/3757] lr: 1.0000e-02 eta: 14:01:42 time: 0.1561 data_time: 0.0106 memory: 7124 grad_norm: 4.9770 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.9200 loss: 1.9200 2022/09/06 21:45:28 - mmengine - INFO - Epoch(train) [16][3400/3757] lr: 1.0000e-02 eta: 14:01:25 time: 0.1549 data_time: 0.0093 memory: 7124 grad_norm: 5.0797 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8314 loss: 1.8314 2022/09/06 21:45:44 - mmengine - INFO - Epoch(train) [16][3500/3757] lr: 1.0000e-02 eta: 14:01:08 time: 0.1638 data_time: 0.0088 memory: 7124 grad_norm: 5.1184 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.8699 loss: 1.8699 2022/09/06 21:45:59 - mmengine - INFO - Epoch(train) [16][3600/3757] lr: 1.0000e-02 eta: 14:00:50 time: 0.1549 data_time: 0.0110 memory: 7124 grad_norm: 5.2713 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.3215 loss: 2.3215 2022/09/06 21:46:07 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:46:15 - mmengine - INFO - Epoch(train) [16][3700/3757] lr: 1.0000e-02 eta: 14:00:33 time: 0.1626 data_time: 0.0131 memory: 7124 grad_norm: 5.1328 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.7766 loss: 1.7766 2022/09/06 21:46:24 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:46:24 - mmengine - INFO - Epoch(train) [16][3757/3757] lr: 1.0000e-02 eta: 14:00:26 time: 0.1401 data_time: 0.0071 memory: 7124 grad_norm: 5.2917 top1_acc: 0.7143 top5_acc: 1.0000 loss_cls: 1.9962 loss: 1.9962 2022/09/06 21:46:24 - mmengine - INFO - Saving checkpoint at 16 epochs 2022/09/06 21:48:43 - mmengine - INFO - Epoch(val) [16][100/310] eta: 0:04:11 time: 1.1981 data_time: 0.8894 memory: 7627 2022/09/06 21:51:00 - mmengine - INFO - Epoch(val) [16][200/310] eta: 0:02:26 time: 1.3284 data_time: 1.0235 memory: 7627 2022/09/06 21:53:04 - mmengine - INFO - Epoch(val) [16][300/310] eta: 0:00:11 time: 1.1232 data_time: 0.8241 memory: 7627 2022/09/06 21:53:19 - mmengine - INFO - Epoch(val) [16][310/310] acc/top1: 0.6077 acc/top5: 0.8394 acc/mean1: 0.6077 2022/09/06 21:53:36 - mmengine - INFO - Epoch(train) [17][100/3757] lr: 1.0000e-02 eta: 14:00:00 time: 0.1557 data_time: 0.0098 memory: 7627 grad_norm: 5.0576 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.9731 loss: 1.9731 2022/09/06 21:53:52 - mmengine - INFO - Epoch(train) [17][200/3757] lr: 1.0000e-02 eta: 13:59:42 time: 0.1542 data_time: 0.0091 memory: 7124 grad_norm: 5.2760 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.7611 loss: 1.7611 2022/09/06 21:54:08 - mmengine - INFO - Epoch(train) [17][300/3757] lr: 1.0000e-02 eta: 13:59:25 time: 0.1552 data_time: 0.0101 memory: 7124 grad_norm: 5.0951 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8567 loss: 1.8567 2022/09/06 21:54:23 - mmengine - INFO - Epoch(train) [17][400/3757] lr: 1.0000e-02 eta: 13:59:08 time: 0.1588 data_time: 0.0119 memory: 7124 grad_norm: 5.1503 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9935 loss: 1.9935 2022/09/06 21:54:39 - mmengine - INFO - Epoch(train) [17][500/3757] lr: 1.0000e-02 eta: 13:58:50 time: 0.1519 data_time: 0.0091 memory: 7124 grad_norm: 5.1529 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.0243 loss: 2.0243 2022/09/06 21:54:55 - mmengine - INFO - Epoch(train) [17][600/3757] lr: 1.0000e-02 eta: 13:58:33 time: 0.1631 data_time: 0.0117 memory: 7124 grad_norm: 4.9615 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.6398 loss: 1.6398 2022/09/06 21:55:11 - mmengine - INFO - Epoch(train) [17][700/3757] lr: 1.0000e-02 eta: 13:58:17 time: 0.1584 data_time: 0.0116 memory: 7124 grad_norm: 5.0257 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8909 loss: 1.8909 2022/09/06 21:55:26 - mmengine - INFO - Epoch(train) [17][800/3757] lr: 1.0000e-02 eta: 13:57:59 time: 0.1562 data_time: 0.0102 memory: 7124 grad_norm: 5.1939 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.9081 loss: 1.9081 2022/09/06 21:55:40 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:55:42 - mmengine - INFO - Epoch(train) [17][900/3757] lr: 1.0000e-02 eta: 13:57:41 time: 0.1555 data_time: 0.0097 memory: 7124 grad_norm: 5.2740 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8463 loss: 1.8463 2022/09/06 21:55:57 - mmengine - INFO - Epoch(train) [17][1000/3757] lr: 1.0000e-02 eta: 13:57:23 time: 0.1559 data_time: 0.0098 memory: 7124 grad_norm: 5.4985 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9176 loss: 1.9176 2022/09/06 21:56:13 - mmengine - INFO - Epoch(train) [17][1100/3757] lr: 1.0000e-02 eta: 13:57:05 time: 0.1562 data_time: 0.0105 memory: 7124 grad_norm: 5.0760 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7508 loss: 1.7508 2022/09/06 21:56:29 - mmengine - INFO - Epoch(train) [17][1200/3757] lr: 1.0000e-02 eta: 13:56:48 time: 0.1544 data_time: 0.0104 memory: 7124 grad_norm: 5.2479 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8953 loss: 1.8953 2022/09/06 21:56:45 - mmengine - INFO - Epoch(train) [17][1300/3757] lr: 1.0000e-02 eta: 13:56:30 time: 0.1590 data_time: 0.0094 memory: 7124 grad_norm: 5.1561 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.3719 loss: 2.3719 2022/09/06 21:57:00 - mmengine - INFO - Epoch(train) [17][1400/3757] lr: 1.0000e-02 eta: 13:56:13 time: 0.1547 data_time: 0.0105 memory: 7124 grad_norm: 5.2496 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.1174 loss: 2.1174 2022/09/06 21:57:16 - mmengine - INFO - Epoch(train) [17][1500/3757] lr: 1.0000e-02 eta: 13:55:55 time: 0.1595 data_time: 0.0113 memory: 7124 grad_norm: 5.2771 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.6514 loss: 1.6514 2022/09/06 21:57:32 - mmengine - INFO - Epoch(train) [17][1600/3757] lr: 1.0000e-02 eta: 13:55:38 time: 0.1550 data_time: 0.0092 memory: 7124 grad_norm: 5.1568 top1_acc: 0.2500 top5_acc: 1.0000 loss_cls: 1.6701 loss: 1.6701 2022/09/06 21:57:48 - mmengine - INFO - Epoch(train) [17][1700/3757] lr: 1.0000e-02 eta: 13:55:21 time: 0.1549 data_time: 0.0094 memory: 7124 grad_norm: 5.1253 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8334 loss: 1.8334 2022/09/06 21:58:03 - mmengine - INFO - Epoch(train) [17][1800/3757] lr: 1.0000e-02 eta: 13:55:04 time: 0.1579 data_time: 0.0109 memory: 7124 grad_norm: 5.2685 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.2998 loss: 2.2998 2022/09/06 21:58:17 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 21:58:19 - mmengine - INFO - Epoch(train) [17][1900/3757] lr: 1.0000e-02 eta: 13:54:46 time: 0.1556 data_time: 0.0102 memory: 7124 grad_norm: 4.9869 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.1848 loss: 2.1848 2022/09/06 21:58:35 - mmengine - INFO - Epoch(train) [17][2000/3757] lr: 1.0000e-02 eta: 13:54:30 time: 0.1619 data_time: 0.0110 memory: 7124 grad_norm: 5.1621 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8102 loss: 1.8102 2022/09/06 21:58:51 - mmengine - INFO - Epoch(train) [17][2100/3757] lr: 1.0000e-02 eta: 13:54:12 time: 0.1566 data_time: 0.0094 memory: 7124 grad_norm: 5.0177 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.2306 loss: 2.2306 2022/09/06 21:59:06 - mmengine - INFO - Epoch(train) [17][2200/3757] lr: 1.0000e-02 eta: 13:53:56 time: 0.1604 data_time: 0.0104 memory: 7124 grad_norm: 5.3184 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.1045 loss: 2.1045 2022/09/06 21:59:22 - mmengine - INFO - Epoch(train) [17][2300/3757] lr: 1.0000e-02 eta: 13:53:40 time: 0.1557 data_time: 0.0097 memory: 7124 grad_norm: 5.3089 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.9451 loss: 1.9451 2022/09/06 21:59:38 - mmengine - INFO - Epoch(train) [17][2400/3757] lr: 1.0000e-02 eta: 13:53:22 time: 0.1518 data_time: 0.0103 memory: 7124 grad_norm: 5.1363 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.3780 loss: 2.3780 2022/09/06 21:59:54 - mmengine - INFO - Epoch(train) [17][2500/3757] lr: 1.0000e-02 eta: 13:53:05 time: 0.1512 data_time: 0.0102 memory: 7124 grad_norm: 5.1925 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9551 loss: 1.9551 2022/09/06 22:00:10 - mmengine - INFO - Epoch(train) [17][2600/3757] lr: 1.0000e-02 eta: 13:52:50 time: 0.1562 data_time: 0.0104 memory: 7124 grad_norm: 5.1622 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.1614 loss: 2.1614 2022/09/06 22:00:26 - mmengine - INFO - Epoch(train) [17][2700/3757] lr: 1.0000e-02 eta: 13:52:33 time: 0.1557 data_time: 0.0096 memory: 7124 grad_norm: 5.2284 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.1505 loss: 2.1505 2022/09/06 22:00:42 - mmengine - INFO - Epoch(train) [17][2800/3757] lr: 1.0000e-02 eta: 13:52:16 time: 0.1573 data_time: 0.0092 memory: 7124 grad_norm: 5.1898 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.1139 loss: 2.1139 2022/09/06 22:00:55 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:00:57 - mmengine - INFO - Epoch(train) [17][2900/3757] lr: 1.0000e-02 eta: 13:51:58 time: 0.1588 data_time: 0.0101 memory: 7124 grad_norm: 5.0151 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.0557 loss: 2.0557 2022/09/06 22:01:13 - mmengine - INFO - Epoch(train) [17][3000/3757] lr: 1.0000e-02 eta: 13:51:40 time: 0.1532 data_time: 0.0088 memory: 7124 grad_norm: 5.1097 top1_acc: 0.1250 top5_acc: 0.6250 loss_cls: 2.1457 loss: 2.1457 2022/09/06 22:01:29 - mmengine - INFO - Epoch(train) [17][3100/3757] lr: 1.0000e-02 eta: 13:51:25 time: 0.1518 data_time: 0.0097 memory: 7124 grad_norm: 5.1250 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.9173 loss: 1.9173 2022/09/06 22:01:45 - mmengine - INFO - Epoch(train) [17][3200/3757] lr: 1.0000e-02 eta: 13:51:07 time: 0.1586 data_time: 0.0095 memory: 7124 grad_norm: 5.3067 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0145 loss: 2.0145 2022/09/06 22:02:00 - mmengine - INFO - Epoch(train) [17][3300/3757] lr: 1.0000e-02 eta: 13:50:49 time: 0.1529 data_time: 0.0099 memory: 7124 grad_norm: 5.0390 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.5830 loss: 1.5830 2022/09/06 22:02:16 - mmengine - INFO - Epoch(train) [17][3400/3757] lr: 1.0000e-02 eta: 13:50:32 time: 0.1597 data_time: 0.0124 memory: 7124 grad_norm: 5.2048 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.0697 loss: 2.0697 2022/09/06 22:02:31 - mmengine - INFO - Epoch(train) [17][3500/3757] lr: 1.0000e-02 eta: 13:50:14 time: 0.1553 data_time: 0.0094 memory: 7124 grad_norm: 5.0075 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.3900 loss: 2.3900 2022/09/06 22:02:47 - mmengine - INFO - Epoch(train) [17][3600/3757] lr: 1.0000e-02 eta: 13:49:57 time: 0.1596 data_time: 0.0094 memory: 7124 grad_norm: 5.2476 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8700 loss: 1.8700 2022/09/06 22:03:03 - mmengine - INFO - Epoch(train) [17][3700/3757] lr: 1.0000e-02 eta: 13:49:39 time: 0.1552 data_time: 0.0092 memory: 7124 grad_norm: 5.2459 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 2.0637 loss: 2.0637 2022/09/06 22:03:11 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:03:11 - mmengine - INFO - Epoch(train) [17][3757/3757] lr: 1.0000e-02 eta: 13:49:32 time: 0.1345 data_time: 0.0071 memory: 7124 grad_norm: 5.1416 top1_acc: 0.5714 top5_acc: 0.7143 loss_cls: 1.9099 loss: 1.9099 2022/09/06 22:03:11 - mmengine - INFO - Saving checkpoint at 17 epochs 2022/09/06 22:05:29 - mmengine - INFO - Epoch(val) [17][100/310] eta: 0:03:51 time: 1.1035 data_time: 0.7989 memory: 7627 2022/09/06 22:07:47 - mmengine - INFO - Epoch(val) [17][200/310] eta: 0:02:31 time: 1.3791 data_time: 1.0736 memory: 7627 2022/09/06 22:09:52 - mmengine - INFO - Epoch(val) [17][300/310] eta: 0:00:11 time: 1.1387 data_time: 0.8363 memory: 7627 2022/09/06 22:10:09 - mmengine - INFO - Epoch(val) [17][310/310] acc/top1: 0.6188 acc/top5: 0.8478 acc/mean1: 0.6185 2022/09/06 22:10:09 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_15.pth is removed 2022/09/06 22:10:11 - mmengine - INFO - The best checkpoint with 0.6188 acc/top1 at 17 epoch is saved to best_acc/top1_epoch_17.pth. 2022/09/06 22:10:28 - mmengine - INFO - Epoch(train) [18][100/3757] lr: 1.0000e-02 eta: 13:49:03 time: 0.1632 data_time: 0.0109 memory: 7627 grad_norm: 5.3062 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.8572 loss: 1.8572 2022/09/06 22:10:32 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:10:43 - mmengine - INFO - Epoch(train) [18][200/3757] lr: 1.0000e-02 eta: 13:48:45 time: 0.1532 data_time: 0.0108 memory: 7124 grad_norm: 5.1502 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.7675 loss: 1.7675 2022/09/06 22:10:59 - mmengine - INFO - Epoch(train) [18][300/3757] lr: 1.0000e-02 eta: 13:48:29 time: 0.1530 data_time: 0.0099 memory: 7124 grad_norm: 5.0785 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8281 loss: 1.8281 2022/09/06 22:11:15 - mmengine - INFO - Epoch(train) [18][400/3757] lr: 1.0000e-02 eta: 13:48:12 time: 0.1544 data_time: 0.0093 memory: 7124 grad_norm: 5.1884 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.6722 loss: 1.6722 2022/09/06 22:11:31 - mmengine - INFO - Epoch(train) [18][500/3757] lr: 1.0000e-02 eta: 13:47:55 time: 0.1545 data_time: 0.0105 memory: 7124 grad_norm: 5.1452 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8167 loss: 1.8167 2022/09/06 22:11:47 - mmengine - INFO - Epoch(train) [18][600/3757] lr: 1.0000e-02 eta: 13:47:38 time: 0.1627 data_time: 0.0110 memory: 7124 grad_norm: 4.9707 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9497 loss: 1.9497 2022/09/06 22:12:02 - mmengine - INFO - Epoch(train) [18][700/3757] lr: 1.0000e-02 eta: 13:47:21 time: 0.1572 data_time: 0.0109 memory: 7124 grad_norm: 5.4290 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.9220 loss: 1.9220 2022/09/06 22:12:18 - mmengine - INFO - Epoch(train) [18][800/3757] lr: 1.0000e-02 eta: 13:47:05 time: 0.1591 data_time: 0.0105 memory: 7124 grad_norm: 5.1474 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0951 loss: 2.0951 2022/09/06 22:12:34 - mmengine - INFO - Epoch(train) [18][900/3757] lr: 1.0000e-02 eta: 13:46:48 time: 0.1556 data_time: 0.0098 memory: 7124 grad_norm: 5.2403 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.8260 loss: 1.8260 2022/09/06 22:12:50 - mmengine - INFO - Epoch(train) [18][1000/3757] lr: 1.0000e-02 eta: 13:46:31 time: 0.1604 data_time: 0.0115 memory: 7124 grad_norm: 5.1986 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9217 loss: 1.9217 2022/09/06 22:13:06 - mmengine - INFO - Epoch(train) [18][1100/3757] lr: 1.0000e-02 eta: 13:46:14 time: 0.1559 data_time: 0.0104 memory: 7124 grad_norm: 5.2107 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.7119 loss: 1.7119 2022/09/06 22:13:11 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:13:21 - mmengine - INFO - Epoch(train) [18][1200/3757] lr: 1.0000e-02 eta: 13:45:57 time: 0.1571 data_time: 0.0105 memory: 7124 grad_norm: 5.0573 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.9714 loss: 1.9714 2022/09/06 22:13:37 - mmengine - INFO - Epoch(train) [18][1300/3757] lr: 1.0000e-02 eta: 13:45:40 time: 0.1563 data_time: 0.0113 memory: 7124 grad_norm: 5.0449 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.0800 loss: 2.0800 2022/09/06 22:13:53 - mmengine - INFO - Epoch(train) [18][1400/3757] lr: 1.0000e-02 eta: 13:45:24 time: 0.1567 data_time: 0.0109 memory: 7124 grad_norm: 5.2739 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9339 loss: 1.9339 2022/09/06 22:14:09 - mmengine - INFO - Epoch(train) [18][1500/3757] lr: 1.0000e-02 eta: 13:45:07 time: 0.1573 data_time: 0.0109 memory: 7124 grad_norm: 5.3812 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8567 loss: 1.8567 2022/09/06 22:14:25 - mmengine - INFO - Epoch(train) [18][1600/3757] lr: 1.0000e-02 eta: 13:44:50 time: 0.1591 data_time: 0.0103 memory: 7124 grad_norm: 5.2674 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.3225 loss: 2.3225 2022/09/06 22:14:40 - mmengine - INFO - Epoch(train) [18][1700/3757] lr: 1.0000e-02 eta: 13:44:33 time: 0.1547 data_time: 0.0115 memory: 7124 grad_norm: 5.1271 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7946 loss: 1.7946 2022/09/06 22:14:56 - mmengine - INFO - Epoch(train) [18][1800/3757] lr: 1.0000e-02 eta: 13:44:17 time: 0.1559 data_time: 0.0094 memory: 7124 grad_norm: 5.3283 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9425 loss: 1.9425 2022/09/06 22:15:12 - mmengine - INFO - Epoch(train) [18][1900/3757] lr: 1.0000e-02 eta: 13:44:01 time: 0.1526 data_time: 0.0107 memory: 7124 grad_norm: 5.0707 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.1360 loss: 2.1360 2022/09/06 22:15:28 - mmengine - INFO - Epoch(train) [18][2000/3757] lr: 1.0000e-02 eta: 13:43:44 time: 0.1590 data_time: 0.0115 memory: 7124 grad_norm: 5.1507 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.6629 loss: 1.6629 2022/09/06 22:15:44 - mmengine - INFO - Epoch(train) [18][2100/3757] lr: 1.0000e-02 eta: 13:43:27 time: 0.1569 data_time: 0.0098 memory: 7124 grad_norm: 5.0989 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.0395 loss: 2.0395 2022/09/06 22:15:49 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:16:00 - mmengine - INFO - Epoch(train) [18][2200/3757] lr: 1.0000e-02 eta: 13:43:13 time: 0.1667 data_time: 0.0100 memory: 7124 grad_norm: 5.1872 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.8594 loss: 1.8594 2022/09/06 22:16:16 - mmengine - INFO - Epoch(train) [18][2300/3757] lr: 1.0000e-02 eta: 13:42:55 time: 0.1574 data_time: 0.0107 memory: 7124 grad_norm: 5.1879 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.3220 loss: 2.3220 2022/09/06 22:16:32 - mmengine - INFO - Epoch(train) [18][2400/3757] lr: 1.0000e-02 eta: 13:42:39 time: 0.1567 data_time: 0.0098 memory: 7124 grad_norm: 5.1487 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.2144 loss: 2.2144 2022/09/06 22:16:48 - mmengine - INFO - Epoch(train) [18][2500/3757] lr: 1.0000e-02 eta: 13:42:23 time: 0.1601 data_time: 0.0121 memory: 7124 grad_norm: 4.8480 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9191 loss: 1.9191 2022/09/06 22:17:04 - mmengine - INFO - Epoch(train) [18][2600/3757] lr: 1.0000e-02 eta: 13:42:07 time: 0.1543 data_time: 0.0103 memory: 7124 grad_norm: 4.9681 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.3609 loss: 2.3609 2022/09/06 22:17:19 - mmengine - INFO - Epoch(train) [18][2700/3757] lr: 1.0000e-02 eta: 13:41:50 time: 0.1551 data_time: 0.0106 memory: 7124 grad_norm: 5.1996 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 1.9089 loss: 1.9089 2022/09/06 22:17:35 - mmengine - INFO - Epoch(train) [18][2800/3757] lr: 1.0000e-02 eta: 13:41:33 time: 0.1577 data_time: 0.0098 memory: 7124 grad_norm: 5.2991 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7573 loss: 1.7573 2022/09/06 22:17:51 - mmengine - INFO - Epoch(train) [18][2900/3757] lr: 1.0000e-02 eta: 13:41:16 time: 0.1546 data_time: 0.0104 memory: 7124 grad_norm: 5.2250 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 2.1872 loss: 2.1872 2022/09/06 22:18:07 - mmengine - INFO - Epoch(train) [18][3000/3757] lr: 1.0000e-02 eta: 13:41:00 time: 0.1609 data_time: 0.0108 memory: 7124 grad_norm: 5.1439 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.2072 loss: 2.2072 2022/09/06 22:18:23 - mmengine - INFO - Epoch(train) [18][3100/3757] lr: 1.0000e-02 eta: 13:40:43 time: 0.1524 data_time: 0.0102 memory: 7124 grad_norm: 5.4506 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8793 loss: 1.8793 2022/09/06 22:18:28 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:18:39 - mmengine - INFO - Epoch(train) [18][3200/3757] lr: 1.0000e-02 eta: 13:40:27 time: 0.1648 data_time: 0.0102 memory: 7124 grad_norm: 5.4503 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.5730 loss: 1.5730 2022/09/06 22:18:55 - mmengine - INFO - Epoch(train) [18][3300/3757] lr: 1.0000e-02 eta: 13:40:10 time: 0.1541 data_time: 0.0101 memory: 7124 grad_norm: 5.1254 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.1054 loss: 2.1054 2022/09/06 22:19:10 - mmengine - INFO - Epoch(train) [18][3400/3757] lr: 1.0000e-02 eta: 13:39:53 time: 0.1534 data_time: 0.0098 memory: 7124 grad_norm: 4.9844 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.9772 loss: 1.9772 2022/09/06 22:19:26 - mmengine - INFO - Epoch(train) [18][3500/3757] lr: 1.0000e-02 eta: 13:39:36 time: 0.1570 data_time: 0.0112 memory: 7124 grad_norm: 5.0534 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 2.1227 loss: 2.1227 2022/09/06 22:19:42 - mmengine - INFO - Epoch(train) [18][3600/3757] lr: 1.0000e-02 eta: 13:39:19 time: 0.1569 data_time: 0.0110 memory: 7124 grad_norm: 4.9119 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8987 loss: 1.8987 2022/09/06 22:19:57 - mmengine - INFO - Epoch(train) [18][3700/3757] lr: 1.0000e-02 eta: 13:39:02 time: 0.1523 data_time: 0.0100 memory: 7124 grad_norm: 5.1054 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.2930 loss: 2.2930 2022/09/06 22:20:06 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:20:06 - mmengine - INFO - Epoch(train) [18][3757/3757] lr: 1.0000e-02 eta: 13:38:55 time: 0.1347 data_time: 0.0072 memory: 7124 grad_norm: 4.9869 top1_acc: 0.4286 top5_acc: 0.5714 loss_cls: 1.9045 loss: 1.9045 2022/09/06 22:20:06 - mmengine - INFO - Saving checkpoint at 18 epochs 2022/09/06 22:22:26 - mmengine - INFO - Epoch(val) [18][100/310] eta: 0:04:12 time: 1.2033 data_time: 0.8888 memory: 7627 2022/09/06 22:24:43 - mmengine - INFO - Epoch(val) [18][200/310] eta: 0:02:16 time: 1.2421 data_time: 0.9395 memory: 7627 2022/09/06 22:26:47 - mmengine - INFO - Epoch(val) [18][300/310] eta: 0:00:12 time: 1.2051 data_time: 0.9052 memory: 7627 2022/09/06 22:27:05 - mmengine - INFO - Epoch(val) [18][310/310] acc/top1: 0.6195 acc/top5: 0.8490 acc/mean1: 0.6194 2022/09/06 22:27:05 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_17.pth is removed 2022/09/06 22:27:06 - mmengine - INFO - The best checkpoint with 0.6195 acc/top1 at 18 epoch is saved to best_acc/top1_epoch_18.pth. 2022/09/06 22:27:23 - mmengine - INFO - Epoch(train) [19][100/3757] lr: 1.0000e-02 eta: 13:38:26 time: 0.1535 data_time: 0.0092 memory: 7627 grad_norm: 5.1934 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.9431 loss: 1.9431 2022/09/06 22:27:39 - mmengine - INFO - Epoch(train) [19][200/3757] lr: 1.0000e-02 eta: 13:38:09 time: 0.1633 data_time: 0.0130 memory: 7124 grad_norm: 5.0975 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7966 loss: 1.7966 2022/09/06 22:27:54 - mmengine - INFO - Epoch(train) [19][300/3757] lr: 1.0000e-02 eta: 13:37:51 time: 0.1557 data_time: 0.0112 memory: 7124 grad_norm: 5.0650 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.9627 loss: 1.9627 2022/09/06 22:28:06 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:28:10 - mmengine - INFO - Epoch(train) [19][400/3757] lr: 1.0000e-02 eta: 13:37:34 time: 0.1567 data_time: 0.0112 memory: 7124 grad_norm: 5.2158 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7194 loss: 1.7194 2022/09/06 22:28:26 - mmengine - INFO - Epoch(train) [19][500/3757] lr: 1.0000e-02 eta: 13:37:17 time: 0.1545 data_time: 0.0091 memory: 7124 grad_norm: 5.1345 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8840 loss: 1.8840 2022/09/06 22:28:42 - mmengine - INFO - Epoch(train) [19][600/3757] lr: 1.0000e-02 eta: 13:37:00 time: 0.1512 data_time: 0.0091 memory: 7124 grad_norm: 5.0799 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 2.1939 loss: 2.1939 2022/09/06 22:28:57 - mmengine - INFO - Epoch(train) [19][700/3757] lr: 1.0000e-02 eta: 13:36:42 time: 0.1541 data_time: 0.0106 memory: 7124 grad_norm: 5.1174 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.9337 loss: 1.9337 2022/09/06 22:29:13 - mmengine - INFO - Epoch(train) [19][800/3757] lr: 1.0000e-02 eta: 13:36:25 time: 0.1528 data_time: 0.0097 memory: 7124 grad_norm: 4.9501 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.9103 loss: 1.9103 2022/09/06 22:29:28 - mmengine - INFO - Epoch(train) [19][900/3757] lr: 1.0000e-02 eta: 13:36:07 time: 0.1566 data_time: 0.0108 memory: 7124 grad_norm: 5.0974 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.7636 loss: 1.7636 2022/09/06 22:29:44 - mmengine - INFO - Epoch(train) [19][1000/3757] lr: 1.0000e-02 eta: 13:35:50 time: 0.1551 data_time: 0.0089 memory: 7124 grad_norm: 4.9851 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0188 loss: 2.0188 2022/09/06 22:29:59 - mmengine - INFO - Epoch(train) [19][1100/3757] lr: 1.0000e-02 eta: 13:35:32 time: 0.1558 data_time: 0.0100 memory: 7124 grad_norm: 5.1076 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.9720 loss: 1.9720 2022/09/06 22:30:15 - mmengine - INFO - Epoch(train) [19][1200/3757] lr: 1.0000e-02 eta: 13:35:14 time: 0.1557 data_time: 0.0100 memory: 7124 grad_norm: 5.0930 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.6046 loss: 1.6046 2022/09/06 22:30:31 - mmengine - INFO - Epoch(train) [19][1300/3757] lr: 1.0000e-02 eta: 13:34:57 time: 0.1535 data_time: 0.0104 memory: 7124 grad_norm: 5.0253 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 2.1581 loss: 2.1581 2022/09/06 22:30:42 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:30:46 - mmengine - INFO - Epoch(train) [19][1400/3757] lr: 1.0000e-02 eta: 13:34:40 time: 0.1537 data_time: 0.0108 memory: 7124 grad_norm: 5.0249 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.9014 loss: 1.9014 2022/09/06 22:31:02 - mmengine - INFO - Epoch(train) [19][1500/3757] lr: 1.0000e-02 eta: 13:34:22 time: 0.1550 data_time: 0.0096 memory: 7124 grad_norm: 5.4476 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.0812 loss: 2.0812 2022/09/06 22:31:18 - mmengine - INFO - Epoch(train) [19][1600/3757] lr: 1.0000e-02 eta: 13:34:06 time: 0.1620 data_time: 0.0098 memory: 7124 grad_norm: 5.0293 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0463 loss: 2.0463 2022/09/06 22:31:34 - mmengine - INFO - Epoch(train) [19][1700/3757] lr: 1.0000e-02 eta: 13:33:48 time: 0.1534 data_time: 0.0087 memory: 7124 grad_norm: 5.0085 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.1823 loss: 2.1823 2022/09/06 22:31:49 - mmengine - INFO - Epoch(train) [19][1800/3757] lr: 1.0000e-02 eta: 13:33:31 time: 0.1567 data_time: 0.0111 memory: 7124 grad_norm: 5.2542 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.9919 loss: 1.9919 2022/09/06 22:32:05 - mmengine - INFO - Epoch(train) [19][1900/3757] lr: 1.0000e-02 eta: 13:33:13 time: 0.1566 data_time: 0.0118 memory: 7124 grad_norm: 5.2383 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.1921 loss: 2.1921 2022/09/06 22:32:20 - mmengine - INFO - Epoch(train) [19][2000/3757] lr: 1.0000e-02 eta: 13:32:56 time: 0.1563 data_time: 0.0090 memory: 7124 grad_norm: 4.9784 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.2622 loss: 2.2622 2022/09/06 22:32:36 - mmengine - INFO - Epoch(train) [19][2100/3757] lr: 1.0000e-02 eta: 13:32:38 time: 0.1527 data_time: 0.0090 memory: 7124 grad_norm: 5.3867 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.0550 loss: 2.0550 2022/09/06 22:32:52 - mmengine - INFO - Epoch(train) [19][2200/3757] lr: 1.0000e-02 eta: 13:32:22 time: 0.1536 data_time: 0.0095 memory: 7124 grad_norm: 5.1465 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.2082 loss: 2.2082 2022/09/06 22:33:08 - mmengine - INFO - Epoch(train) [19][2300/3757] lr: 1.0000e-02 eta: 13:32:05 time: 0.1535 data_time: 0.0090 memory: 7124 grad_norm: 5.3372 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8925 loss: 1.8925 2022/09/06 22:33:19 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:33:24 - mmengine - INFO - Epoch(train) [19][2400/3757] lr: 1.0000e-02 eta: 13:31:50 time: 0.1744 data_time: 0.0100 memory: 7124 grad_norm: 5.1256 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8924 loss: 1.8924 2022/09/06 22:33:39 - mmengine - INFO - Epoch(train) [19][2500/3757] lr: 1.0000e-02 eta: 13:31:32 time: 0.1618 data_time: 0.0108 memory: 7124 grad_norm: 5.1608 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.0367 loss: 2.0367 2022/09/06 22:33:55 - mmengine - INFO - Epoch(train) [19][2600/3757] lr: 1.0000e-02 eta: 13:31:15 time: 0.1586 data_time: 0.0099 memory: 7124 grad_norm: 5.2604 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.1550 loss: 2.1550 2022/09/06 22:34:11 - mmengine - INFO - Epoch(train) [19][2700/3757] lr: 1.0000e-02 eta: 13:30:58 time: 0.1618 data_time: 0.0096 memory: 7124 grad_norm: 5.1726 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0089 loss: 2.0089 2022/09/06 22:34:26 - mmengine - INFO - Epoch(train) [19][2800/3757] lr: 1.0000e-02 eta: 13:30:41 time: 0.1553 data_time: 0.0102 memory: 7124 grad_norm: 5.3851 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8319 loss: 1.8319 2022/09/06 22:34:42 - mmengine - INFO - Epoch(train) [19][2900/3757] lr: 1.0000e-02 eta: 13:30:24 time: 0.1519 data_time: 0.0104 memory: 7124 grad_norm: 4.8705 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.7538 loss: 1.7538 2022/09/06 22:34:58 - mmengine - INFO - Epoch(train) [19][3000/3757] lr: 1.0000e-02 eta: 13:30:06 time: 0.1538 data_time: 0.0098 memory: 7124 grad_norm: 5.2502 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8715 loss: 1.8715 2022/09/06 22:35:13 - mmengine - INFO - Epoch(train) [19][3100/3757] lr: 1.0000e-02 eta: 13:29:49 time: 0.1618 data_time: 0.0109 memory: 7124 grad_norm: 5.1737 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.1781 loss: 2.1781 2022/09/06 22:35:29 - mmengine - INFO - Epoch(train) [19][3200/3757] lr: 1.0000e-02 eta: 13:29:32 time: 0.1652 data_time: 0.0152 memory: 7124 grad_norm: 5.2208 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8773 loss: 1.8773 2022/09/06 22:35:45 - mmengine - INFO - Epoch(train) [19][3300/3757] lr: 1.0000e-02 eta: 13:29:15 time: 0.1547 data_time: 0.0104 memory: 7124 grad_norm: 5.0578 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.8479 loss: 1.8479 2022/09/06 22:35:56 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:36:00 - mmengine - INFO - Epoch(train) [19][3400/3757] lr: 1.0000e-02 eta: 13:28:58 time: 0.1574 data_time: 0.0109 memory: 7124 grad_norm: 5.3764 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8547 loss: 1.8547 2022/09/06 22:36:16 - mmengine - INFO - Epoch(train) [19][3500/3757] lr: 1.0000e-02 eta: 13:28:40 time: 0.1527 data_time: 0.0100 memory: 7124 grad_norm: 5.1744 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9910 loss: 1.9910 2022/09/06 22:36:31 - mmengine - INFO - Epoch(train) [19][3600/3757] lr: 1.0000e-02 eta: 13:28:22 time: 0.1487 data_time: 0.0098 memory: 7124 grad_norm: 4.9451 top1_acc: 0.3750 top5_acc: 0.3750 loss_cls: 2.0282 loss: 2.0282 2022/09/06 22:36:47 - mmengine - INFO - Epoch(train) [19][3700/3757] lr: 1.0000e-02 eta: 13:28:05 time: 0.1572 data_time: 0.0115 memory: 7124 grad_norm: 5.1705 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9776 loss: 1.9776 2022/09/06 22:36:56 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:36:56 - mmengine - INFO - Epoch(train) [19][3757/3757] lr: 1.0000e-02 eta: 13:27:58 time: 0.1356 data_time: 0.0069 memory: 7124 grad_norm: 5.3943 top1_acc: 0.2857 top5_acc: 0.4286 loss_cls: 2.0862 loss: 2.0862 2022/09/06 22:36:56 - mmengine - INFO - Saving checkpoint at 19 epochs 2022/09/06 22:39:15 - mmengine - INFO - Epoch(val) [19][100/310] eta: 0:04:28 time: 1.2802 data_time: 0.9705 memory: 7627 2022/09/06 22:41:30 - mmengine - INFO - Epoch(val) [19][200/310] eta: 0:02:04 time: 1.1302 data_time: 0.8129 memory: 7627 2022/09/06 22:43:37 - mmengine - INFO - Epoch(val) [19][300/310] eta: 0:00:12 time: 1.2627 data_time: 0.9609 memory: 7627 2022/09/06 22:43:53 - mmengine - INFO - Epoch(val) [19][310/310] acc/top1: 0.6291 acc/top5: 0.8538 acc/mean1: 0.6289 2022/09/06 22:43:53 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_18.pth is removed 2022/09/06 22:43:55 - mmengine - INFO - The best checkpoint with 0.6291 acc/top1 at 19 epoch is saved to best_acc/top1_epoch_19.pth. 2022/09/06 22:44:12 - mmengine - INFO - Epoch(train) [20][100/3757] lr: 1.0000e-02 eta: 13:27:31 time: 0.1550 data_time: 0.0100 memory: 7627 grad_norm: 5.2351 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9344 loss: 1.9344 2022/09/06 22:44:27 - mmengine - INFO - Epoch(train) [20][200/3757] lr: 1.0000e-02 eta: 13:27:14 time: 0.1545 data_time: 0.0111 memory: 7124 grad_norm: 5.1511 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.0168 loss: 2.0168 2022/09/06 22:44:43 - mmengine - INFO - Epoch(train) [20][300/3757] lr: 1.0000e-02 eta: 13:26:57 time: 0.1571 data_time: 0.0096 memory: 7124 grad_norm: 5.4210 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.8270 loss: 1.8270 2022/09/06 22:44:59 - mmengine - INFO - Epoch(train) [20][400/3757] lr: 1.0000e-02 eta: 13:26:39 time: 0.1521 data_time: 0.0103 memory: 7124 grad_norm: 5.3352 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9809 loss: 1.9809 2022/09/06 22:45:14 - mmengine - INFO - Epoch(train) [20][500/3757] lr: 1.0000e-02 eta: 13:26:23 time: 0.1546 data_time: 0.0097 memory: 7124 grad_norm: 5.0189 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1321 loss: 2.1321 2022/09/06 22:45:31 - mmengine - INFO - Epoch(train) [20][600/3757] lr: 1.0000e-02 eta: 13:26:08 time: 0.1609 data_time: 0.0091 memory: 7124 grad_norm: 5.0026 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.5812 loss: 1.5812 2022/09/06 22:45:33 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:45:46 - mmengine - INFO - Epoch(train) [20][700/3757] lr: 1.0000e-02 eta: 13:25:50 time: 0.1561 data_time: 0.0110 memory: 7124 grad_norm: 5.2798 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8751 loss: 1.8751 2022/09/06 22:46:02 - mmengine - INFO - Epoch(train) [20][800/3757] lr: 1.0000e-02 eta: 13:25:33 time: 0.1625 data_time: 0.0157 memory: 7124 grad_norm: 5.3392 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.1120 loss: 2.1120 2022/09/06 22:46:17 - mmengine - INFO - Epoch(train) [20][900/3757] lr: 1.0000e-02 eta: 13:25:16 time: 0.1553 data_time: 0.0109 memory: 7124 grad_norm: 5.1026 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.9480 loss: 1.9480 2022/09/06 22:46:33 - mmengine - INFO - Epoch(train) [20][1000/3757] lr: 1.0000e-02 eta: 13:24:58 time: 0.1514 data_time: 0.0089 memory: 7124 grad_norm: 5.1718 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.1334 loss: 2.1334 2022/09/06 22:46:49 - mmengine - INFO - Epoch(train) [20][1100/3757] lr: 1.0000e-02 eta: 13:24:40 time: 0.1543 data_time: 0.0089 memory: 7124 grad_norm: 4.9429 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.6404 loss: 1.6404 2022/09/06 22:47:04 - mmengine - INFO - Epoch(train) [20][1200/3757] lr: 1.0000e-02 eta: 13:24:24 time: 0.1553 data_time: 0.0104 memory: 7124 grad_norm: 5.2271 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7866 loss: 1.7866 2022/09/06 22:47:20 - mmengine - INFO - Epoch(train) [20][1300/3757] lr: 1.0000e-02 eta: 13:24:06 time: 0.1580 data_time: 0.0108 memory: 7124 grad_norm: 5.1527 top1_acc: 0.2500 top5_acc: 0.8750 loss_cls: 1.7865 loss: 1.7865 2022/09/06 22:47:36 - mmengine - INFO - Epoch(train) [20][1400/3757] lr: 1.0000e-02 eta: 13:23:49 time: 0.1547 data_time: 0.0105 memory: 7124 grad_norm: 5.2146 top1_acc: 0.3750 top5_acc: 1.0000 loss_cls: 2.1399 loss: 2.1399 2022/09/06 22:47:51 - mmengine - INFO - Epoch(train) [20][1500/3757] lr: 1.0000e-02 eta: 13:23:32 time: 0.1551 data_time: 0.0100 memory: 7124 grad_norm: 5.0430 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.1152 loss: 2.1152 2022/09/06 22:48:07 - mmengine - INFO - Epoch(train) [20][1600/3757] lr: 1.0000e-02 eta: 13:23:15 time: 0.1589 data_time: 0.0100 memory: 7124 grad_norm: 5.1964 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.0101 loss: 2.0101 2022/09/06 22:48:10 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:48:23 - mmengine - INFO - Epoch(train) [20][1700/3757] lr: 1.0000e-02 eta: 13:22:59 time: 0.1537 data_time: 0.0109 memory: 7124 grad_norm: 5.0989 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.7314 loss: 1.7314 2022/09/06 22:48:39 - mmengine - INFO - Epoch(train) [20][1800/3757] lr: 1.0000e-02 eta: 13:22:42 time: 0.1557 data_time: 0.0099 memory: 7124 grad_norm: 5.1889 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.7605 loss: 1.7605 2022/09/06 22:48:54 - mmengine - INFO - Epoch(train) [20][1900/3757] lr: 1.0000e-02 eta: 13:22:25 time: 0.1547 data_time: 0.0107 memory: 7124 grad_norm: 5.0966 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.4487 loss: 2.4487 2022/09/06 22:49:10 - mmengine - INFO - Epoch(train) [20][2000/3757] lr: 1.0000e-02 eta: 13:22:07 time: 0.1547 data_time: 0.0101 memory: 7124 grad_norm: 5.3043 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.2784 loss: 2.2784 2022/09/06 22:49:25 - mmengine - INFO - Epoch(train) [20][2100/3757] lr: 1.0000e-02 eta: 13:21:50 time: 0.1535 data_time: 0.0096 memory: 7124 grad_norm: 5.1173 top1_acc: 0.3750 top5_acc: 0.3750 loss_cls: 2.0110 loss: 2.0110 2022/09/06 22:49:41 - mmengine - INFO - Epoch(train) [20][2200/3757] lr: 1.0000e-02 eta: 13:21:34 time: 0.1650 data_time: 0.0103 memory: 7124 grad_norm: 5.1939 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.0856 loss: 2.0856 2022/09/06 22:49:57 - mmengine - INFO - Epoch(train) [20][2300/3757] lr: 1.0000e-02 eta: 13:21:16 time: 0.1548 data_time: 0.0091 memory: 7124 grad_norm: 4.8771 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9469 loss: 1.9469 2022/09/06 22:50:12 - mmengine - INFO - Epoch(train) [20][2400/3757] lr: 1.0000e-02 eta: 13:20:59 time: 0.1570 data_time: 0.0095 memory: 7124 grad_norm: 5.2300 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.6842 loss: 1.6842 2022/09/06 22:50:28 - mmengine - INFO - Epoch(train) [20][2500/3757] lr: 1.0000e-02 eta: 13:20:42 time: 0.1551 data_time: 0.0100 memory: 7124 grad_norm: 5.1173 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8682 loss: 1.8682 2022/09/06 22:50:44 - mmengine - INFO - Epoch(train) [20][2600/3757] lr: 1.0000e-02 eta: 13:20:25 time: 0.1546 data_time: 0.0103 memory: 7124 grad_norm: 5.1682 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.0963 loss: 2.0963 2022/09/06 22:50:47 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:51:00 - mmengine - INFO - Epoch(train) [20][2700/3757] lr: 1.0000e-02 eta: 13:20:08 time: 0.1573 data_time: 0.0092 memory: 7124 grad_norm: 5.3496 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.7746 loss: 1.7746 2022/09/06 22:51:15 - mmengine - INFO - Epoch(train) [20][2800/3757] lr: 1.0000e-02 eta: 13:19:50 time: 0.1503 data_time: 0.0099 memory: 7124 grad_norm: 5.1226 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 2.1669 loss: 2.1669 2022/09/06 22:51:31 - mmengine - INFO - Epoch(train) [20][2900/3757] lr: 1.0000e-02 eta: 13:19:34 time: 0.1598 data_time: 0.0121 memory: 7124 grad_norm: 5.1851 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9881 loss: 1.9881 2022/09/06 22:51:47 - mmengine - INFO - Epoch(train) [20][3000/3757] lr: 1.0000e-02 eta: 13:19:17 time: 0.1547 data_time: 0.0105 memory: 7124 grad_norm: 5.0700 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8174 loss: 1.8174 2022/09/06 22:52:02 - mmengine - INFO - Epoch(train) [20][3100/3757] lr: 1.0000e-02 eta: 13:19:00 time: 0.1555 data_time: 0.0123 memory: 7124 grad_norm: 5.0342 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.6264 loss: 1.6264 2022/09/06 22:52:18 - mmengine - INFO - Epoch(train) [20][3200/3757] lr: 1.0000e-02 eta: 13:18:43 time: 0.1563 data_time: 0.0094 memory: 7124 grad_norm: 5.2992 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0124 loss: 2.0124 2022/09/06 22:52:34 - mmengine - INFO - Epoch(train) [20][3300/3757] lr: 1.0000e-02 eta: 13:18:27 time: 0.1507 data_time: 0.0085 memory: 7124 grad_norm: 5.1446 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.8794 loss: 1.8794 2022/09/06 22:52:50 - mmengine - INFO - Epoch(train) [20][3400/3757] lr: 1.0000e-02 eta: 13:18:10 time: 0.1590 data_time: 0.0107 memory: 7124 grad_norm: 5.1189 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 2.2153 loss: 2.2153 2022/09/06 22:53:05 - mmengine - INFO - Epoch(train) [20][3500/3757] lr: 1.0000e-02 eta: 13:17:52 time: 0.1548 data_time: 0.0110 memory: 7124 grad_norm: 5.1346 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 1.8762 loss: 1.8762 2022/09/06 22:53:21 - mmengine - INFO - Epoch(train) [20][3600/3757] lr: 1.0000e-02 eta: 13:17:35 time: 0.1577 data_time: 0.0112 memory: 7124 grad_norm: 5.3167 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7251 loss: 1.7251 2022/09/06 22:53:23 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:53:36 - mmengine - INFO - Epoch(train) [20][3700/3757] lr: 1.0000e-02 eta: 13:17:18 time: 0.1505 data_time: 0.0105 memory: 7124 grad_norm: 5.3306 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 2.0058 loss: 2.0058 2022/09/06 22:53:45 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 22:53:45 - mmengine - INFO - Epoch(train) [20][3757/3757] lr: 1.0000e-02 eta: 13:17:11 time: 0.1347 data_time: 0.0068 memory: 7124 grad_norm: 5.2506 top1_acc: 0.2857 top5_acc: 0.8571 loss_cls: 1.9799 loss: 1.9799 2022/09/06 22:53:45 - mmengine - INFO - Saving checkpoint at 20 epochs 2022/09/06 22:56:02 - mmengine - INFO - Epoch(val) [20][100/310] eta: 0:03:47 time: 1.0841 data_time: 0.7712 memory: 7627 2022/09/06 22:58:22 - mmengine - INFO - Epoch(val) [20][200/310] eta: 0:02:33 time: 1.3955 data_time: 1.0929 memory: 7627 2022/09/06 23:00:25 - mmengine - INFO - Epoch(val) [20][300/310] eta: 0:00:11 time: 1.1218 data_time: 0.8232 memory: 7627 2022/09/06 23:00:43 - mmengine - INFO - Epoch(val) [20][310/310] acc/top1: 0.6227 acc/top5: 0.8480 acc/mean1: 0.6225 2022/09/06 23:01:01 - mmengine - INFO - Epoch(train) [21][100/3757] lr: 1.0000e-02 eta: 13:16:48 time: 0.1609 data_time: 0.0114 memory: 7627 grad_norm: 4.9962 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7572 loss: 1.7572 2022/09/06 23:01:16 - mmengine - INFO - Epoch(train) [21][200/3757] lr: 1.0000e-02 eta: 13:16:31 time: 0.1561 data_time: 0.0105 memory: 7124 grad_norm: 5.1738 top1_acc: 0.1250 top5_acc: 0.5000 loss_cls: 1.6392 loss: 1.6392 2022/09/06 23:01:32 - mmengine - INFO - Epoch(train) [21][300/3757] lr: 1.0000e-02 eta: 13:16:14 time: 0.1528 data_time: 0.0093 memory: 7124 grad_norm: 5.1410 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.8290 loss: 1.8290 2022/09/06 23:01:48 - mmengine - INFO - Epoch(train) [21][400/3757] lr: 1.0000e-02 eta: 13:15:58 time: 0.1549 data_time: 0.0106 memory: 7124 grad_norm: 5.2489 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.9047 loss: 1.9047 2022/09/06 23:02:04 - mmengine - INFO - Epoch(train) [21][500/3757] lr: 1.0000e-02 eta: 13:15:41 time: 0.1607 data_time: 0.0100 memory: 7124 grad_norm: 5.1712 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8445 loss: 1.8445 2022/09/06 23:02:19 - mmengine - INFO - Epoch(train) [21][600/3757] lr: 1.0000e-02 eta: 13:15:23 time: 0.1569 data_time: 0.0103 memory: 7124 grad_norm: 5.3234 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.8202 loss: 1.8202 2022/09/06 23:02:35 - mmengine - INFO - Epoch(train) [21][700/3757] lr: 1.0000e-02 eta: 13:15:06 time: 0.1541 data_time: 0.0091 memory: 7124 grad_norm: 5.1993 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.6580 loss: 1.6580 2022/09/06 23:02:50 - mmengine - INFO - Epoch(train) [21][800/3757] lr: 1.0000e-02 eta: 13:14:49 time: 0.1561 data_time: 0.0099 memory: 7124 grad_norm: 5.0923 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.8218 loss: 1.8218 2022/09/06 23:03:00 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:03:06 - mmengine - INFO - Epoch(train) [21][900/3757] lr: 1.0000e-02 eta: 13:14:32 time: 0.1536 data_time: 0.0099 memory: 7124 grad_norm: 5.1796 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8546 loss: 1.8546 2022/09/06 23:03:22 - mmengine - INFO - Epoch(train) [21][1000/3757] lr: 1.0000e-02 eta: 13:14:16 time: 0.1597 data_time: 0.0102 memory: 7124 grad_norm: 5.2188 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8126 loss: 1.8126 2022/09/06 23:03:38 - mmengine - INFO - Epoch(train) [21][1100/3757] lr: 1.0000e-02 eta: 13:13:59 time: 0.1545 data_time: 0.0099 memory: 7124 grad_norm: 5.2437 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.8906 loss: 1.8906 2022/09/06 23:03:53 - mmengine - INFO - Epoch(train) [21][1200/3757] lr: 1.0000e-02 eta: 13:13:43 time: 0.1572 data_time: 0.0107 memory: 7124 grad_norm: 5.2707 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9374 loss: 1.9374 2022/09/06 23:04:09 - mmengine - INFO - Epoch(train) [21][1300/3757] lr: 1.0000e-02 eta: 13:13:26 time: 0.1551 data_time: 0.0103 memory: 7124 grad_norm: 5.1570 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.2475 loss: 2.2475 2022/09/06 23:04:25 - mmengine - INFO - Epoch(train) [21][1400/3757] lr: 1.0000e-02 eta: 13:13:09 time: 0.1542 data_time: 0.0100 memory: 7124 grad_norm: 5.0433 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.9523 loss: 1.9523 2022/09/06 23:04:41 - mmengine - INFO - Epoch(train) [21][1500/3757] lr: 1.0000e-02 eta: 13:12:53 time: 0.1573 data_time: 0.0094 memory: 7124 grad_norm: 5.1498 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.9322 loss: 1.9322 2022/09/06 23:04:57 - mmengine - INFO - Epoch(train) [21][1600/3757] lr: 1.0000e-02 eta: 13:12:37 time: 0.1563 data_time: 0.0094 memory: 7124 grad_norm: 5.0495 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7967 loss: 1.7967 2022/09/06 23:05:13 - mmengine - INFO - Epoch(train) [21][1700/3757] lr: 1.0000e-02 eta: 13:12:20 time: 0.1593 data_time: 0.0109 memory: 7124 grad_norm: 5.1983 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.0347 loss: 2.0347 2022/09/06 23:05:28 - mmengine - INFO - Epoch(train) [21][1800/3757] lr: 1.0000e-02 eta: 13:12:04 time: 0.1559 data_time: 0.0120 memory: 7124 grad_norm: 5.0696 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.9311 loss: 1.9311 2022/09/06 23:05:38 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:05:44 - mmengine - INFO - Epoch(train) [21][1900/3757] lr: 1.0000e-02 eta: 13:11:47 time: 0.1568 data_time: 0.0118 memory: 7124 grad_norm: 5.0267 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.6953 loss: 1.6953 2022/09/06 23:06:00 - mmengine - INFO - Epoch(train) [21][2000/3757] lr: 1.0000e-02 eta: 13:11:30 time: 0.1528 data_time: 0.0110 memory: 7124 grad_norm: 4.9161 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8098 loss: 1.8098 2022/09/06 23:06:15 - mmengine - INFO - Epoch(train) [21][2100/3757] lr: 1.0000e-02 eta: 13:11:12 time: 0.1513 data_time: 0.0090 memory: 7124 grad_norm: 5.1420 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9118 loss: 1.9118 2022/09/06 23:06:31 - mmengine - INFO - Epoch(train) [21][2200/3757] lr: 1.0000e-02 eta: 13:10:55 time: 0.1556 data_time: 0.0114 memory: 7124 grad_norm: 5.1851 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9727 loss: 1.9727 2022/09/06 23:06:47 - mmengine - INFO - Epoch(train) [21][2300/3757] lr: 1.0000e-02 eta: 13:10:38 time: 0.1536 data_time: 0.0105 memory: 7124 grad_norm: 5.1470 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9207 loss: 1.9207 2022/09/06 23:07:02 - mmengine - INFO - Epoch(train) [21][2400/3757] lr: 1.0000e-02 eta: 13:10:21 time: 0.1576 data_time: 0.0104 memory: 7124 grad_norm: 5.2998 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7885 loss: 1.7885 2022/09/06 23:07:18 - mmengine - INFO - Epoch(train) [21][2500/3757] lr: 1.0000e-02 eta: 13:10:04 time: 0.1592 data_time: 0.0152 memory: 7124 grad_norm: 4.9521 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.9244 loss: 1.9244 2022/09/06 23:07:34 - mmengine - INFO - Epoch(train) [21][2600/3757] lr: 1.0000e-02 eta: 13:09:48 time: 0.1565 data_time: 0.0095 memory: 7124 grad_norm: 5.0971 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9719 loss: 1.9719 2022/09/06 23:07:49 - mmengine - INFO - Epoch(train) [21][2700/3757] lr: 1.0000e-02 eta: 13:09:32 time: 0.1547 data_time: 0.0098 memory: 7124 grad_norm: 5.1814 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 2.0783 loss: 2.0783 2022/09/06 23:08:05 - mmengine - INFO - Epoch(train) [21][2800/3757] lr: 1.0000e-02 eta: 13:09:16 time: 0.1492 data_time: 0.0100 memory: 7124 grad_norm: 5.1781 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9551 loss: 1.9551 2022/09/06 23:08:15 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:08:21 - mmengine - INFO - Epoch(train) [21][2900/3757] lr: 1.0000e-02 eta: 13:08:58 time: 0.1569 data_time: 0.0089 memory: 7124 grad_norm: 5.1152 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.7594 loss: 1.7594 2022/09/06 23:08:37 - mmengine - INFO - Epoch(train) [21][3000/3757] lr: 1.0000e-02 eta: 13:08:42 time: 0.1573 data_time: 0.0106 memory: 7124 grad_norm: 5.1154 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.0266 loss: 2.0266 2022/09/06 23:08:52 - mmengine - INFO - Epoch(train) [21][3100/3757] lr: 1.0000e-02 eta: 13:08:25 time: 0.1539 data_time: 0.0098 memory: 7124 grad_norm: 5.1159 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.9328 loss: 1.9328 2022/09/06 23:09:08 - mmengine - INFO - Epoch(train) [21][3200/3757] lr: 1.0000e-02 eta: 13:08:07 time: 0.1524 data_time: 0.0095 memory: 7124 grad_norm: 5.0276 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.0795 loss: 2.0795 2022/09/06 23:09:24 - mmengine - INFO - Epoch(train) [21][3300/3757] lr: 1.0000e-02 eta: 13:07:51 time: 0.1609 data_time: 0.0109 memory: 7124 grad_norm: 5.0383 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.0235 loss: 2.0235 2022/09/06 23:09:39 - mmengine - INFO - Epoch(train) [21][3400/3757] lr: 1.0000e-02 eta: 13:07:33 time: 0.1554 data_time: 0.0097 memory: 7124 grad_norm: 5.2530 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9102 loss: 1.9102 2022/09/06 23:09:55 - mmengine - INFO - Epoch(train) [21][3500/3757] lr: 1.0000e-02 eta: 13:07:17 time: 0.1525 data_time: 0.0103 memory: 7124 grad_norm: 5.1759 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.6164 loss: 1.6164 2022/09/06 23:10:11 - mmengine - INFO - Epoch(train) [21][3600/3757] lr: 1.0000e-02 eta: 13:07:00 time: 0.1534 data_time: 0.0096 memory: 7124 grad_norm: 5.2578 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.0122 loss: 2.0122 2022/09/06 23:10:27 - mmengine - INFO - Epoch(train) [21][3700/3757] lr: 1.0000e-02 eta: 13:06:44 time: 0.1538 data_time: 0.0089 memory: 7124 grad_norm: 5.1262 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7024 loss: 1.7024 2022/09/06 23:10:35 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:10:35 - mmengine - INFO - Epoch(train) [21][3757/3757] lr: 1.0000e-02 eta: 13:06:37 time: 0.1341 data_time: 0.0069 memory: 7124 grad_norm: 5.1172 top1_acc: 0.5714 top5_acc: 0.8571 loss_cls: 1.9298 loss: 1.9298 2022/09/06 23:10:35 - mmengine - INFO - Saving checkpoint at 21 epochs 2022/09/06 23:12:52 - mmengine - INFO - Epoch(val) [21][100/310] eta: 0:03:49 time: 1.0925 data_time: 0.7919 memory: 7627 2022/09/06 23:15:11 - mmengine - INFO - Epoch(val) [21][200/310] eta: 0:02:33 time: 1.3938 data_time: 1.0968 memory: 7627 2022/09/06 23:17:15 - mmengine - INFO - Epoch(val) [21][300/310] eta: 0:00:11 time: 1.1265 data_time: 0.8311 memory: 7627 2022/09/06 23:17:31 - mmengine - INFO - Epoch(val) [21][310/310] acc/top1: 0.6265 acc/top5: 0.8533 acc/mean1: 0.6266 2022/09/06 23:17:49 - mmengine - INFO - Epoch(train) [22][100/3757] lr: 1.0000e-02 eta: 13:06:14 time: 0.1574 data_time: 0.0110 memory: 7627 grad_norm: 5.1595 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8487 loss: 1.8487 2022/09/06 23:17:49 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:18:04 - mmengine - INFO - Epoch(train) [22][200/3757] lr: 1.0000e-02 eta: 13:05:57 time: 0.1578 data_time: 0.0110 memory: 7124 grad_norm: 5.1931 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7406 loss: 1.7406 2022/09/06 23:18:20 - mmengine - INFO - Epoch(train) [22][300/3757] lr: 1.0000e-02 eta: 13:05:41 time: 0.1581 data_time: 0.0095 memory: 7124 grad_norm: 5.1712 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.9793 loss: 1.9793 2022/09/06 23:18:36 - mmengine - INFO - Epoch(train) [22][400/3757] lr: 1.0000e-02 eta: 13:05:24 time: 0.1546 data_time: 0.0106 memory: 7124 grad_norm: 5.3561 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.9408 loss: 1.9408 2022/09/06 23:18:52 - mmengine - INFO - Epoch(train) [22][500/3757] lr: 1.0000e-02 eta: 13:05:07 time: 0.1592 data_time: 0.0092 memory: 7124 grad_norm: 5.3407 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.7721 loss: 1.7721 2022/09/06 23:19:07 - mmengine - INFO - Epoch(train) [22][600/3757] lr: 1.0000e-02 eta: 13:04:50 time: 0.1622 data_time: 0.0117 memory: 7124 grad_norm: 5.2405 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.5322 loss: 1.5322 2022/09/06 23:19:23 - mmengine - INFO - Epoch(train) [22][700/3757] lr: 1.0000e-02 eta: 13:04:34 time: 0.1591 data_time: 0.0103 memory: 7124 grad_norm: 5.1794 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9119 loss: 1.9119 2022/09/06 23:19:39 - mmengine - INFO - Epoch(train) [22][800/3757] lr: 1.0000e-02 eta: 13:04:17 time: 0.1558 data_time: 0.0113 memory: 7124 grad_norm: 5.2066 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.8000 loss: 1.8000 2022/09/06 23:19:55 - mmengine - INFO - Epoch(train) [22][900/3757] lr: 1.0000e-02 eta: 13:04:01 time: 0.1682 data_time: 0.0096 memory: 7124 grad_norm: 5.3049 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9351 loss: 1.9351 2022/09/06 23:20:10 - mmengine - INFO - Epoch(train) [22][1000/3757] lr: 1.0000e-02 eta: 13:03:45 time: 0.1571 data_time: 0.0098 memory: 7124 grad_norm: 5.2311 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6789 loss: 1.6789 2022/09/06 23:20:26 - mmengine - INFO - Epoch(train) [22][1100/3757] lr: 1.0000e-02 eta: 13:03:28 time: 0.1537 data_time: 0.0101 memory: 7124 grad_norm: 5.1507 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8477 loss: 1.8477 2022/09/06 23:20:27 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:20:42 - mmengine - INFO - Epoch(train) [22][1200/3757] lr: 1.0000e-02 eta: 13:03:10 time: 0.1544 data_time: 0.0096 memory: 7124 grad_norm: 5.2135 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.1320 loss: 2.1320 2022/09/06 23:20:57 - mmengine - INFO - Epoch(train) [22][1300/3757] lr: 1.0000e-02 eta: 13:02:53 time: 0.1533 data_time: 0.0100 memory: 7124 grad_norm: 5.2516 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9321 loss: 1.9321 2022/09/06 23:21:13 - mmengine - INFO - Epoch(train) [22][1400/3757] lr: 1.0000e-02 eta: 13:02:37 time: 0.1605 data_time: 0.0137 memory: 7124 grad_norm: 5.1203 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.2087 loss: 2.2087 2022/09/06 23:21:29 - mmengine - INFO - Epoch(train) [22][1500/3757] lr: 1.0000e-02 eta: 13:02:20 time: 0.1570 data_time: 0.0101 memory: 7124 grad_norm: 4.9909 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.8503 loss: 1.8503 2022/09/06 23:21:44 - mmengine - INFO - Epoch(train) [22][1600/3757] lr: 1.0000e-02 eta: 13:02:03 time: 0.1601 data_time: 0.0111 memory: 7124 grad_norm: 5.0540 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.9875 loss: 1.9875 2022/09/06 23:22:00 - mmengine - INFO - Epoch(train) [22][1700/3757] lr: 1.0000e-02 eta: 13:01:46 time: 0.1529 data_time: 0.0096 memory: 7124 grad_norm: 5.3658 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.1152 loss: 2.1152 2022/09/06 23:22:15 - mmengine - INFO - Epoch(train) [22][1800/3757] lr: 1.0000e-02 eta: 13:01:29 time: 0.1575 data_time: 0.0102 memory: 7124 grad_norm: 5.1270 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.0838 loss: 2.0838 2022/09/06 23:22:31 - mmengine - INFO - Epoch(train) [22][1900/3757] lr: 1.0000e-02 eta: 13:01:12 time: 0.1520 data_time: 0.0119 memory: 7124 grad_norm: 5.1291 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.7113 loss: 1.7113 2022/09/06 23:22:47 - mmengine - INFO - Epoch(train) [22][2000/3757] lr: 1.0000e-02 eta: 13:00:55 time: 0.1539 data_time: 0.0095 memory: 7124 grad_norm: 5.2171 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.8965 loss: 1.8965 2022/09/06 23:23:03 - mmengine - INFO - Epoch(train) [22][2100/3757] lr: 1.0000e-02 eta: 13:00:38 time: 0.1573 data_time: 0.0102 memory: 7124 grad_norm: 5.2498 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.7474 loss: 1.7474 2022/09/06 23:23:03 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:23:18 - mmengine - INFO - Epoch(train) [22][2200/3757] lr: 1.0000e-02 eta: 13:00:22 time: 0.1579 data_time: 0.0103 memory: 7124 grad_norm: 5.2486 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.8505 loss: 1.8505 2022/09/06 23:23:34 - mmengine - INFO - Epoch(train) [22][2300/3757] lr: 1.0000e-02 eta: 13:00:05 time: 0.1551 data_time: 0.0102 memory: 7124 grad_norm: 5.2480 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9220 loss: 1.9220 2022/09/06 23:23:49 - mmengine - INFO - Epoch(train) [22][2400/3757] lr: 1.0000e-02 eta: 12:59:47 time: 0.1538 data_time: 0.0097 memory: 7124 grad_norm: 5.1618 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8030 loss: 1.8030 2022/09/06 23:24:05 - mmengine - INFO - Epoch(train) [22][2500/3757] lr: 1.0000e-02 eta: 12:59:31 time: 0.1556 data_time: 0.0101 memory: 7124 grad_norm: 5.1988 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.9436 loss: 1.9436 2022/09/06 23:24:21 - mmengine - INFO - Epoch(train) [22][2600/3757] lr: 1.0000e-02 eta: 12:59:14 time: 0.1574 data_time: 0.0099 memory: 7124 grad_norm: 5.2593 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8964 loss: 1.8964 2022/09/06 23:24:37 - mmengine - INFO - Epoch(train) [22][2700/3757] lr: 1.0000e-02 eta: 12:58:57 time: 0.1547 data_time: 0.0099 memory: 7124 grad_norm: 5.3475 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.5582 loss: 1.5582 2022/09/06 23:24:52 - mmengine - INFO - Epoch(train) [22][2800/3757] lr: 1.0000e-02 eta: 12:58:40 time: 0.1541 data_time: 0.0096 memory: 7124 grad_norm: 5.3456 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.0968 loss: 2.0968 2022/09/06 23:25:08 - mmengine - INFO - Epoch(train) [22][2900/3757] lr: 1.0000e-02 eta: 12:58:23 time: 0.1543 data_time: 0.0093 memory: 7124 grad_norm: 5.2271 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.9880 loss: 1.9880 2022/09/06 23:25:23 - mmengine - INFO - Epoch(train) [22][3000/3757] lr: 1.0000e-02 eta: 12:58:06 time: 0.1576 data_time: 0.0108 memory: 7124 grad_norm: 5.1405 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.9310 loss: 1.9310 2022/09/06 23:25:39 - mmengine - INFO - Epoch(train) [22][3100/3757] lr: 1.0000e-02 eta: 12:57:49 time: 0.1557 data_time: 0.0107 memory: 7124 grad_norm: 5.0652 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.0135 loss: 2.0135 2022/09/06 23:25:40 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:25:55 - mmengine - INFO - Epoch(train) [22][3200/3757] lr: 1.0000e-02 eta: 12:57:34 time: 0.1789 data_time: 0.0108 memory: 7124 grad_norm: 5.0869 top1_acc: 0.3750 top5_acc: 1.0000 loss_cls: 2.1396 loss: 2.1396 2022/09/06 23:26:11 - mmengine - INFO - Epoch(train) [22][3300/3757] lr: 1.0000e-02 eta: 12:57:17 time: 0.1595 data_time: 0.0092 memory: 7124 grad_norm: 5.0388 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6416 loss: 1.6416 2022/09/06 23:26:26 - mmengine - INFO - Epoch(train) [22][3400/3757] lr: 1.0000e-02 eta: 12:57:00 time: 0.1543 data_time: 0.0094 memory: 7124 grad_norm: 5.5095 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.3810 loss: 2.3810 2022/09/06 23:26:42 - mmengine - INFO - Epoch(train) [22][3500/3757] lr: 1.0000e-02 eta: 12:56:43 time: 0.1549 data_time: 0.0099 memory: 7124 grad_norm: 5.0000 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.6111 loss: 1.6111 2022/09/06 23:26:58 - mmengine - INFO - Epoch(train) [22][3600/3757] lr: 1.0000e-02 eta: 12:56:27 time: 0.1583 data_time: 0.0097 memory: 7124 grad_norm: 5.1706 top1_acc: 0.2500 top5_acc: 0.8750 loss_cls: 1.9595 loss: 1.9595 2022/09/06 23:27:14 - mmengine - INFO - Epoch(train) [22][3700/3757] lr: 1.0000e-02 eta: 12:56:10 time: 0.1578 data_time: 0.0131 memory: 7124 grad_norm: 5.0558 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9659 loss: 1.9659 2022/09/06 23:27:22 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:27:22 - mmengine - INFO - Epoch(train) [22][3757/3757] lr: 1.0000e-02 eta: 12:56:03 time: 0.1354 data_time: 0.0074 memory: 7124 grad_norm: 5.1991 top1_acc: 0.8571 top5_acc: 0.8571 loss_cls: 2.0409 loss: 2.0409 2022/09/06 23:27:22 - mmengine - INFO - Saving checkpoint at 22 epochs 2022/09/06 23:29:42 - mmengine - INFO - Epoch(val) [22][100/310] eta: 0:04:12 time: 1.2035 data_time: 0.8974 memory: 7627 2022/09/06 23:31:58 - mmengine - INFO - Epoch(val) [22][200/310] eta: 0:02:12 time: 1.2059 data_time: 0.9033 memory: 7627 2022/09/06 23:34:03 - mmengine - INFO - Epoch(val) [22][300/310] eta: 0:00:12 time: 1.2291 data_time: 0.9291 memory: 7627 2022/09/06 23:34:19 - mmengine - INFO - Epoch(val) [22][310/310] acc/top1: 0.6224 acc/top5: 0.8528 acc/mean1: 0.6224 2022/09/06 23:34:37 - mmengine - INFO - Epoch(train) [23][100/3757] lr: 1.0000e-02 eta: 12:55:41 time: 0.1568 data_time: 0.0123 memory: 7627 grad_norm: 5.0107 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.5353 loss: 1.5353 2022/09/06 23:34:53 - mmengine - INFO - Epoch(train) [23][200/3757] lr: 1.0000e-02 eta: 12:55:24 time: 0.1555 data_time: 0.0101 memory: 7124 grad_norm: 5.2672 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.8920 loss: 1.8920 2022/09/06 23:35:08 - mmengine - INFO - Epoch(train) [23][300/3757] lr: 1.0000e-02 eta: 12:55:07 time: 0.1559 data_time: 0.0087 memory: 7124 grad_norm: 5.1911 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 2.2784 loss: 2.2784 2022/09/06 23:35:15 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:35:24 - mmengine - INFO - Epoch(train) [23][400/3757] lr: 1.0000e-02 eta: 12:54:50 time: 0.1537 data_time: 0.0103 memory: 7124 grad_norm: 5.2259 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.0840 loss: 2.0840 2022/09/06 23:35:39 - mmengine - INFO - Epoch(train) [23][500/3757] lr: 1.0000e-02 eta: 12:54:33 time: 0.1526 data_time: 0.0098 memory: 7124 grad_norm: 5.2729 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.6397 loss: 1.6397 2022/09/06 23:35:55 - mmengine - INFO - Epoch(train) [23][600/3757] lr: 1.0000e-02 eta: 12:54:16 time: 0.1605 data_time: 0.0087 memory: 7124 grad_norm: 5.2006 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9160 loss: 1.9160 2022/09/06 23:36:11 - mmengine - INFO - Epoch(train) [23][700/3757] lr: 1.0000e-02 eta: 12:53:59 time: 0.1553 data_time: 0.0097 memory: 7124 grad_norm: 5.2038 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9800 loss: 1.9800 2022/09/06 23:36:26 - mmengine - INFO - Epoch(train) [23][800/3757] lr: 1.0000e-02 eta: 12:53:42 time: 0.1534 data_time: 0.0099 memory: 7124 grad_norm: 5.1952 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.6842 loss: 1.6842 2022/09/06 23:36:42 - mmengine - INFO - Epoch(train) [23][900/3757] lr: 1.0000e-02 eta: 12:53:26 time: 0.1543 data_time: 0.0102 memory: 7124 grad_norm: 5.1235 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.1459 loss: 2.1459 2022/09/06 23:36:58 - mmengine - INFO - Epoch(train) [23][1000/3757] lr: 1.0000e-02 eta: 12:53:09 time: 0.1527 data_time: 0.0104 memory: 7124 grad_norm: 5.1455 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8022 loss: 1.8022 2022/09/06 23:37:13 - mmengine - INFO - Epoch(train) [23][1100/3757] lr: 1.0000e-02 eta: 12:52:52 time: 0.1521 data_time: 0.0098 memory: 7124 grad_norm: 5.3253 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.9359 loss: 1.9359 2022/09/06 23:37:29 - mmengine - INFO - Epoch(train) [23][1200/3757] lr: 1.0000e-02 eta: 12:52:35 time: 0.1513 data_time: 0.0097 memory: 7124 grad_norm: 5.1366 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.7215 loss: 1.7215 2022/09/06 23:37:45 - mmengine - INFO - Epoch(train) [23][1300/3757] lr: 1.0000e-02 eta: 12:52:19 time: 0.1603 data_time: 0.0100 memory: 7124 grad_norm: 5.3033 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.2225 loss: 2.2225 2022/09/06 23:37:52 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:38:01 - mmengine - INFO - Epoch(train) [23][1400/3757] lr: 1.0000e-02 eta: 12:52:03 time: 0.1612 data_time: 0.0100 memory: 7124 grad_norm: 5.0260 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.6188 loss: 1.6188 2022/09/06 23:38:16 - mmengine - INFO - Epoch(train) [23][1500/3757] lr: 1.0000e-02 eta: 12:51:46 time: 0.1580 data_time: 0.0080 memory: 7124 grad_norm: 5.1077 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 1.8352 loss: 1.8352 2022/09/06 23:38:32 - mmengine - INFO - Epoch(train) [23][1600/3757] lr: 1.0000e-02 eta: 12:51:30 time: 0.1502 data_time: 0.0094 memory: 7124 grad_norm: 5.3797 top1_acc: 0.1250 top5_acc: 0.2500 loss_cls: 1.9270 loss: 1.9270 2022/09/06 23:38:48 - mmengine - INFO - Epoch(train) [23][1700/3757] lr: 1.0000e-02 eta: 12:51:13 time: 0.1559 data_time: 0.0115 memory: 7124 grad_norm: 5.2121 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 1.6548 loss: 1.6548 2022/09/06 23:39:04 - mmengine - INFO - Epoch(train) [23][1800/3757] lr: 1.0000e-02 eta: 12:50:56 time: 0.1542 data_time: 0.0094 memory: 7124 grad_norm: 5.4038 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.0976 loss: 2.0976 2022/09/06 23:39:19 - mmengine - INFO - Epoch(train) [23][1900/3757] lr: 1.0000e-02 eta: 12:50:40 time: 0.1544 data_time: 0.0082 memory: 7124 grad_norm: 5.2056 top1_acc: 0.1250 top5_acc: 0.7500 loss_cls: 1.8959 loss: 1.8959 2022/09/06 23:39:35 - mmengine - INFO - Epoch(train) [23][2000/3757] lr: 1.0000e-02 eta: 12:50:24 time: 0.1718 data_time: 0.0101 memory: 7124 grad_norm: 5.2798 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8871 loss: 1.8871 2022/09/06 23:39:51 - mmengine - INFO - Epoch(train) [23][2100/3757] lr: 1.0000e-02 eta: 12:50:08 time: 0.1550 data_time: 0.0111 memory: 7124 grad_norm: 5.4359 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9862 loss: 1.9862 2022/09/06 23:40:07 - mmengine - INFO - Epoch(train) [23][2200/3757] lr: 1.0000e-02 eta: 12:49:52 time: 0.1577 data_time: 0.0094 memory: 7124 grad_norm: 5.2422 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 2.0352 loss: 2.0352 2022/09/06 23:40:23 - mmengine - INFO - Epoch(train) [23][2300/3757] lr: 1.0000e-02 eta: 12:49:35 time: 0.1563 data_time: 0.0098 memory: 7124 grad_norm: 5.1553 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.4326 loss: 1.4326 2022/09/06 23:40:30 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:40:38 - mmengine - INFO - Epoch(train) [23][2400/3757] lr: 1.0000e-02 eta: 12:49:18 time: 0.1526 data_time: 0.0093 memory: 7124 grad_norm: 5.0873 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.5865 loss: 1.5865 2022/09/06 23:40:54 - mmengine - INFO - Epoch(train) [23][2500/3757] lr: 1.0000e-02 eta: 12:49:01 time: 0.1567 data_time: 0.0119 memory: 7124 grad_norm: 5.3836 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8978 loss: 1.8978 2022/09/06 23:41:10 - mmengine - INFO - Epoch(train) [23][2600/3757] lr: 1.0000e-02 eta: 12:48:45 time: 0.1560 data_time: 0.0098 memory: 7124 grad_norm: 5.3584 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.9770 loss: 1.9770 2022/09/06 23:41:26 - mmengine - INFO - Epoch(train) [23][2700/3757] lr: 1.0000e-02 eta: 12:48:29 time: 0.1671 data_time: 0.0094 memory: 7124 grad_norm: 5.3596 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9250 loss: 1.9250 2022/09/06 23:41:41 - mmengine - INFO - Epoch(train) [23][2800/3757] lr: 1.0000e-02 eta: 12:48:12 time: 0.1525 data_time: 0.0090 memory: 7124 grad_norm: 5.3198 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.8042 loss: 1.8042 2022/09/06 23:41:57 - mmengine - INFO - Epoch(train) [23][2900/3757] lr: 1.0000e-02 eta: 12:47:56 time: 0.1619 data_time: 0.0099 memory: 7124 grad_norm: 5.1055 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9953 loss: 1.9953 2022/09/06 23:42:13 - mmengine - INFO - Epoch(train) [23][3000/3757] lr: 1.0000e-02 eta: 12:47:39 time: 0.1536 data_time: 0.0110 memory: 7124 grad_norm: 5.2132 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8348 loss: 1.8348 2022/09/06 23:42:28 - mmengine - INFO - Epoch(train) [23][3100/3757] lr: 1.0000e-02 eta: 12:47:22 time: 0.1502 data_time: 0.0103 memory: 7124 grad_norm: 5.3488 top1_acc: 0.1250 top5_acc: 0.3750 loss_cls: 2.1231 loss: 2.1231 2022/09/06 23:42:44 - mmengine - INFO - Epoch(train) [23][3200/3757] lr: 1.0000e-02 eta: 12:47:06 time: 0.1565 data_time: 0.0095 memory: 7124 grad_norm: 5.3177 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9712 loss: 1.9712 2022/09/06 23:43:00 - mmengine - INFO - Epoch(train) [23][3300/3757] lr: 1.0000e-02 eta: 12:46:49 time: 0.1520 data_time: 0.0104 memory: 7124 grad_norm: 5.3419 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.1057 loss: 2.1057 2022/09/06 23:43:07 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:43:15 - mmengine - INFO - Epoch(train) [23][3400/3757] lr: 1.0000e-02 eta: 12:46:32 time: 0.1617 data_time: 0.0100 memory: 7124 grad_norm: 5.0290 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.8914 loss: 1.8914 2022/09/06 23:43:31 - mmengine - INFO - Epoch(train) [23][3500/3757] lr: 1.0000e-02 eta: 12:46:16 time: 0.1546 data_time: 0.0094 memory: 7124 grad_norm: 5.1913 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.9417 loss: 1.9417 2022/09/06 23:43:47 - mmengine - INFO - Epoch(train) [23][3600/3757] lr: 1.0000e-02 eta: 12:45:59 time: 0.1557 data_time: 0.0094 memory: 7124 grad_norm: 4.9814 top1_acc: 0.1250 top5_acc: 0.7500 loss_cls: 1.6272 loss: 1.6272 2022/09/06 23:44:02 - mmengine - INFO - Epoch(train) [23][3700/3757] lr: 1.0000e-02 eta: 12:45:42 time: 0.1528 data_time: 0.0093 memory: 7124 grad_norm: 5.2481 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9413 loss: 1.9413 2022/09/06 23:44:11 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:44:11 - mmengine - INFO - Epoch(train) [23][3757/3757] lr: 1.0000e-02 eta: 12:45:36 time: 0.1359 data_time: 0.0073 memory: 7124 grad_norm: 5.4738 top1_acc: 0.4286 top5_acc: 0.7143 loss_cls: 1.6480 loss: 1.6480 2022/09/06 23:44:11 - mmengine - INFO - Saving checkpoint at 23 epochs 2022/09/06 23:46:30 - mmengine - INFO - Epoch(val) [23][100/310] eta: 0:04:18 time: 1.2291 data_time: 0.9190 memory: 7627 2022/09/06 23:48:44 - mmengine - INFO - Epoch(val) [23][200/310] eta: 0:02:12 time: 1.2028 data_time: 0.8952 memory: 7627 2022/09/06 23:50:51 - mmengine - INFO - Epoch(val) [23][300/310] eta: 0:00:13 time: 1.3054 data_time: 1.0012 memory: 7627 2022/09/06 23:51:09 - mmengine - INFO - Epoch(val) [23][310/310] acc/top1: 0.6280 acc/top5: 0.8532 acc/mean1: 0.6279 2022/09/06 23:51:27 - mmengine - INFO - Epoch(train) [24][100/3757] lr: 1.0000e-02 eta: 12:45:14 time: 0.1647 data_time: 0.0100 memory: 7627 grad_norm: 5.0256 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.1701 loss: 2.1701 2022/09/06 23:51:43 - mmengine - INFO - Epoch(train) [24][200/3757] lr: 1.0000e-02 eta: 12:44:58 time: 0.1566 data_time: 0.0104 memory: 7124 grad_norm: 5.1693 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6435 loss: 1.6435 2022/09/06 23:51:59 - mmengine - INFO - Epoch(train) [24][300/3757] lr: 1.0000e-02 eta: 12:44:42 time: 0.1553 data_time: 0.0103 memory: 7124 grad_norm: 5.2836 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8968 loss: 1.8968 2022/09/06 23:52:14 - mmengine - INFO - Epoch(train) [24][400/3757] lr: 1.0000e-02 eta: 12:44:26 time: 0.1554 data_time: 0.0091 memory: 7124 grad_norm: 5.0494 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.2145 loss: 2.2145 2022/09/06 23:52:30 - mmengine - INFO - Epoch(train) [24][500/3757] lr: 1.0000e-02 eta: 12:44:10 time: 0.1558 data_time: 0.0074 memory: 7124 grad_norm: 5.1938 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.4670 loss: 2.4670 2022/09/06 23:52:44 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:52:46 - mmengine - INFO - Epoch(train) [24][600/3757] lr: 1.0000e-02 eta: 12:43:54 time: 0.1539 data_time: 0.0106 memory: 7124 grad_norm: 5.1670 top1_acc: 0.3750 top5_acc: 1.0000 loss_cls: 2.1822 loss: 2.1822 2022/09/06 23:53:02 - mmengine - INFO - Epoch(train) [24][700/3757] lr: 1.0000e-02 eta: 12:43:37 time: 0.1557 data_time: 0.0102 memory: 7124 grad_norm: 5.2165 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.6718 loss: 1.6718 2022/09/06 23:53:18 - mmengine - INFO - Epoch(train) [24][800/3757] lr: 1.0000e-02 eta: 12:43:21 time: 0.1643 data_time: 0.0113 memory: 7124 grad_norm: 5.1653 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8458 loss: 1.8458 2022/09/06 23:53:33 - mmengine - INFO - Epoch(train) [24][900/3757] lr: 1.0000e-02 eta: 12:43:04 time: 0.1535 data_time: 0.0107 memory: 7124 grad_norm: 5.3914 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 2.2877 loss: 2.2877 2022/09/06 23:53:50 - mmengine - INFO - Epoch(train) [24][1000/3757] lr: 1.0000e-02 eta: 12:42:50 time: 0.1598 data_time: 0.0107 memory: 7124 grad_norm: 5.1613 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.9605 loss: 1.9605 2022/09/06 23:54:06 - mmengine - INFO - Epoch(train) [24][1100/3757] lr: 1.0000e-02 eta: 12:42:35 time: 0.1570 data_time: 0.0100 memory: 7124 grad_norm: 5.3878 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9741 loss: 1.9741 2022/09/06 23:54:22 - mmengine - INFO - Epoch(train) [24][1200/3757] lr: 1.0000e-02 eta: 12:42:19 time: 0.1547 data_time: 0.0102 memory: 7124 grad_norm: 5.1256 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.9241 loss: 1.9241 2022/09/06 23:54:38 - mmengine - INFO - Epoch(train) [24][1300/3757] lr: 1.0000e-02 eta: 12:42:03 time: 0.1735 data_time: 0.0089 memory: 7124 grad_norm: 5.1645 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9262 loss: 1.9262 2022/09/06 23:54:54 - mmengine - INFO - Epoch(train) [24][1400/3757] lr: 1.0000e-02 eta: 12:41:47 time: 0.1556 data_time: 0.0104 memory: 7124 grad_norm: 5.3003 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.0716 loss: 2.0716 2022/09/06 23:55:10 - mmengine - INFO - Epoch(train) [24][1500/3757] lr: 1.0000e-02 eta: 12:41:32 time: 0.1616 data_time: 0.0114 memory: 7124 grad_norm: 5.2857 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8497 loss: 1.8497 2022/09/06 23:55:24 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:55:25 - mmengine - INFO - Epoch(train) [24][1600/3757] lr: 1.0000e-02 eta: 12:41:15 time: 0.1545 data_time: 0.0091 memory: 7124 grad_norm: 5.2201 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.8691 loss: 1.8691 2022/09/06 23:55:41 - mmengine - INFO - Epoch(train) [24][1700/3757] lr: 1.0000e-02 eta: 12:40:59 time: 0.1523 data_time: 0.0098 memory: 7124 grad_norm: 5.2387 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.2653 loss: 2.2653 2022/09/06 23:55:57 - mmengine - INFO - Epoch(train) [24][1800/3757] lr: 1.0000e-02 eta: 12:40:43 time: 0.1600 data_time: 0.0099 memory: 7124 grad_norm: 5.0996 top1_acc: 0.2500 top5_acc: 0.8750 loss_cls: 1.9557 loss: 1.9557 2022/09/06 23:56:13 - mmengine - INFO - Epoch(train) [24][1900/3757] lr: 1.0000e-02 eta: 12:40:27 time: 0.1583 data_time: 0.0107 memory: 7124 grad_norm: 5.0570 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.9565 loss: 1.9565 2022/09/06 23:56:29 - mmengine - INFO - Epoch(train) [24][2000/3757] lr: 1.0000e-02 eta: 12:40:11 time: 0.1695 data_time: 0.0097 memory: 7124 grad_norm: 5.0917 top1_acc: 0.2500 top5_acc: 1.0000 loss_cls: 1.9665 loss: 1.9665 2022/09/06 23:56:45 - mmengine - INFO - Epoch(train) [24][2100/3757] lr: 1.0000e-02 eta: 12:39:55 time: 0.1566 data_time: 0.0092 memory: 7124 grad_norm: 5.1394 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8909 loss: 1.8909 2022/09/06 23:57:01 - mmengine - INFO - Epoch(train) [24][2200/3757] lr: 1.0000e-02 eta: 12:39:39 time: 0.1542 data_time: 0.0102 memory: 7124 grad_norm: 5.0285 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9843 loss: 1.9843 2022/09/06 23:57:17 - mmengine - INFO - Epoch(train) [24][2300/3757] lr: 1.0000e-02 eta: 12:39:23 time: 0.1528 data_time: 0.0111 memory: 7124 grad_norm: 5.0864 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.0290 loss: 2.0290 2022/09/06 23:57:32 - mmengine - INFO - Epoch(train) [24][2400/3757] lr: 1.0000e-02 eta: 12:39:07 time: 0.1559 data_time: 0.0111 memory: 7124 grad_norm: 5.1114 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.8053 loss: 1.8053 2022/09/06 23:57:48 - mmengine - INFO - Epoch(train) [24][2500/3757] lr: 1.0000e-02 eta: 12:38:51 time: 0.1648 data_time: 0.0087 memory: 7124 grad_norm: 5.2626 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.0867 loss: 2.0867 2022/09/06 23:58:02 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/06 23:58:04 - mmengine - INFO - Epoch(train) [24][2600/3757] lr: 1.0000e-02 eta: 12:38:34 time: 0.1539 data_time: 0.0108 memory: 7124 grad_norm: 5.1875 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.7624 loss: 1.7624 2022/09/06 23:58:20 - mmengine - INFO - Epoch(train) [24][2700/3757] lr: 1.0000e-02 eta: 12:38:19 time: 0.1587 data_time: 0.0112 memory: 7124 grad_norm: 5.2114 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8558 loss: 1.8558 2022/09/06 23:58:36 - mmengine - INFO - Epoch(train) [24][2800/3757] lr: 1.0000e-02 eta: 12:38:02 time: 0.1560 data_time: 0.0091 memory: 7124 grad_norm: 5.1935 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.0306 loss: 2.0306 2022/09/06 23:58:52 - mmengine - INFO - Epoch(train) [24][2900/3757] lr: 1.0000e-02 eta: 12:37:46 time: 0.1519 data_time: 0.0099 memory: 7124 grad_norm: 5.2849 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.0295 loss: 2.0295 2022/09/06 23:59:08 - mmengine - INFO - Epoch(train) [24][3000/3757] lr: 1.0000e-02 eta: 12:37:32 time: 0.1658 data_time: 0.0220 memory: 7124 grad_norm: 5.2518 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.0908 loss: 2.0908 2022/09/06 23:59:24 - mmengine - INFO - Epoch(train) [24][3100/3757] lr: 1.0000e-02 eta: 12:37:17 time: 0.1557 data_time: 0.0105 memory: 7124 grad_norm: 5.2320 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.9898 loss: 1.9898 2022/09/06 23:59:40 - mmengine - INFO - Epoch(train) [24][3200/3757] lr: 1.0000e-02 eta: 12:37:01 time: 0.1531 data_time: 0.0096 memory: 7124 grad_norm: 5.0006 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5568 loss: 1.5568 2022/09/06 23:59:56 - mmengine - INFO - Epoch(train) [24][3300/3757] lr: 1.0000e-02 eta: 12:36:45 time: 0.1588 data_time: 0.0094 memory: 7124 grad_norm: 5.1748 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8522 loss: 1.8522 2022/09/07 00:00:16 - mmengine - INFO - Epoch(train) [24][3400/3757] lr: 1.0000e-02 eta: 12:36:42 time: 0.1544 data_time: 0.0099 memory: 7124 grad_norm: 5.2678 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.9690 loss: 1.9690 2022/09/07 00:00:32 - mmengine - INFO - Epoch(train) [24][3500/3757] lr: 1.0000e-02 eta: 12:36:27 time: 0.1548 data_time: 0.0097 memory: 7124 grad_norm: 5.2198 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8117 loss: 1.8117 2022/09/07 00:00:46 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:00:48 - mmengine - INFO - Epoch(train) [24][3600/3757] lr: 1.0000e-02 eta: 12:36:11 time: 0.1599 data_time: 0.0103 memory: 7124 grad_norm: 5.2110 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9138 loss: 1.9138 2022/09/07 00:01:04 - mmengine - INFO - Epoch(train) [24][3700/3757] lr: 1.0000e-02 eta: 12:35:55 time: 0.1587 data_time: 0.0082 memory: 7124 grad_norm: 5.1846 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7149 loss: 1.7149 2022/09/07 00:01:13 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:01:13 - mmengine - INFO - Epoch(train) [24][3757/3757] lr: 1.0000e-02 eta: 12:35:49 time: 0.1380 data_time: 0.0064 memory: 7124 grad_norm: 5.3776 top1_acc: 0.4286 top5_acc: 0.8571 loss_cls: 1.8553 loss: 1.8553 2022/09/07 00:01:13 - mmengine - INFO - Saving checkpoint at 24 epochs 2022/09/07 00:03:31 - mmengine - INFO - Epoch(val) [24][100/310] eta: 0:04:00 time: 1.1441 data_time: 0.8463 memory: 7627 2022/09/07 00:05:50 - mmengine - INFO - Epoch(val) [24][200/310] eta: 0:02:31 time: 1.3802 data_time: 1.0836 memory: 7627 2022/09/07 00:07:54 - mmengine - INFO - Epoch(val) [24][300/310] eta: 0:00:10 time: 1.0937 data_time: 0.7986 memory: 7627 2022/09/07 00:08:10 - mmengine - INFO - Epoch(val) [24][310/310] acc/top1: 0.6325 acc/top5: 0.8565 acc/mean1: 0.6325 2022/09/07 00:08:10 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_19.pth is removed 2022/09/07 00:08:12 - mmengine - INFO - The best checkpoint with 0.6325 acc/top1 at 24 epoch is saved to best_acc/top1_epoch_24.pth. 2022/09/07 00:08:28 - mmengine - INFO - Epoch(train) [25][100/3757] lr: 1.0000e-02 eta: 12:35:23 time: 0.1590 data_time: 0.0102 memory: 7627 grad_norm: 5.2557 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 2.1937 loss: 2.1937 2022/09/07 00:08:44 - mmengine - INFO - Epoch(train) [25][200/3757] lr: 1.0000e-02 eta: 12:35:07 time: 0.1579 data_time: 0.0101 memory: 7124 grad_norm: 5.4253 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.8911 loss: 1.8911 2022/09/07 00:09:00 - mmengine - INFO - Epoch(train) [25][300/3757] lr: 1.0000e-02 eta: 12:34:51 time: 0.1551 data_time: 0.0091 memory: 7124 grad_norm: 5.1541 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8472 loss: 1.8472 2022/09/07 00:09:16 - mmengine - INFO - Epoch(train) [25][400/3757] lr: 1.0000e-02 eta: 12:34:34 time: 0.1540 data_time: 0.0106 memory: 7124 grad_norm: 5.0644 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9925 loss: 1.9925 2022/09/07 00:09:31 - mmengine - INFO - Epoch(train) [25][500/3757] lr: 1.0000e-02 eta: 12:34:18 time: 0.1551 data_time: 0.0089 memory: 7124 grad_norm: 5.3065 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8715 loss: 1.8715 2022/09/07 00:09:47 - mmengine - INFO - Epoch(train) [25][600/3757] lr: 1.0000e-02 eta: 12:34:01 time: 0.1567 data_time: 0.0117 memory: 7124 grad_norm: 5.2136 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.0906 loss: 2.0906 2022/09/07 00:10:03 - mmengine - INFO - Epoch(train) [25][700/3757] lr: 1.0000e-02 eta: 12:33:45 time: 0.1555 data_time: 0.0103 memory: 7124 grad_norm: 5.3602 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9640 loss: 1.9640 2022/09/07 00:10:19 - mmengine - INFO - Epoch(train) [25][800/3757] lr: 1.0000e-02 eta: 12:33:29 time: 0.1605 data_time: 0.0102 memory: 7124 grad_norm: 5.3341 top1_acc: 0.1250 top5_acc: 0.2500 loss_cls: 1.9411 loss: 1.9411 2022/09/07 00:10:24 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:10:35 - mmengine - INFO - Epoch(train) [25][900/3757] lr: 1.0000e-02 eta: 12:33:13 time: 0.1540 data_time: 0.0089 memory: 7124 grad_norm: 5.2611 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.8441 loss: 1.8441 2022/09/07 00:10:50 - mmengine - INFO - Epoch(train) [25][1000/3757] lr: 1.0000e-02 eta: 12:32:56 time: 0.1564 data_time: 0.0098 memory: 7124 grad_norm: 5.1732 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.7443 loss: 1.7443 2022/09/07 00:11:06 - mmengine - INFO - Epoch(train) [25][1100/3757] lr: 1.0000e-02 eta: 12:32:41 time: 0.1575 data_time: 0.0104 memory: 7124 grad_norm: 5.1866 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7503 loss: 1.7503 2022/09/07 00:11:22 - mmengine - INFO - Epoch(train) [25][1200/3757] lr: 1.0000e-02 eta: 12:32:25 time: 0.1639 data_time: 0.0129 memory: 7124 grad_norm: 5.0005 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.6980 loss: 1.6980 2022/09/07 00:11:38 - mmengine - INFO - Epoch(train) [25][1300/3757] lr: 1.0000e-02 eta: 12:32:08 time: 0.1635 data_time: 0.0105 memory: 7124 grad_norm: 5.1047 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.1027 loss: 2.1027 2022/09/07 00:11:54 - mmengine - INFO - Epoch(train) [25][1400/3757] lr: 1.0000e-02 eta: 12:31:52 time: 0.1595 data_time: 0.0097 memory: 7124 grad_norm: 5.4304 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7629 loss: 1.7629 2022/09/07 00:12:10 - mmengine - INFO - Epoch(train) [25][1500/3757] lr: 1.0000e-02 eta: 12:31:36 time: 0.1567 data_time: 0.0098 memory: 7124 grad_norm: 5.1322 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0558 loss: 2.0558 2022/09/07 00:12:26 - mmengine - INFO - Epoch(train) [25][1600/3757] lr: 1.0000e-02 eta: 12:31:20 time: 0.1547 data_time: 0.0091 memory: 7124 grad_norm: 5.1015 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8906 loss: 1.8906 2022/09/07 00:12:41 - mmengine - INFO - Epoch(train) [25][1700/3757] lr: 1.0000e-02 eta: 12:31:04 time: 0.1501 data_time: 0.0091 memory: 7124 grad_norm: 5.2555 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.5966 loss: 1.5966 2022/09/07 00:12:57 - mmengine - INFO - Epoch(train) [25][1800/3757] lr: 1.0000e-02 eta: 12:30:48 time: 0.1545 data_time: 0.0102 memory: 7124 grad_norm: 5.2152 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.0448 loss: 2.0448 2022/09/07 00:13:02 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:13:13 - mmengine - INFO - Epoch(train) [25][1900/3757] lr: 1.0000e-02 eta: 12:30:32 time: 0.1554 data_time: 0.0104 memory: 7124 grad_norm: 5.2605 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7464 loss: 1.7464 2022/09/07 00:13:29 - mmengine - INFO - Epoch(train) [25][2000/3757] lr: 1.0000e-02 eta: 12:30:16 time: 0.1777 data_time: 0.0103 memory: 7124 grad_norm: 5.2391 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9488 loss: 1.9488 2022/09/07 00:13:45 - mmengine - INFO - Epoch(train) [25][2100/3757] lr: 1.0000e-02 eta: 12:30:00 time: 0.1557 data_time: 0.0100 memory: 7124 grad_norm: 5.1524 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.3599 loss: 2.3599 2022/09/07 00:14:01 - mmengine - INFO - Epoch(train) [25][2200/3757] lr: 1.0000e-02 eta: 12:29:44 time: 0.1517 data_time: 0.0110 memory: 7124 grad_norm: 5.2340 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7970 loss: 1.7970 2022/09/07 00:14:16 - mmengine - INFO - Epoch(train) [25][2300/3757] lr: 1.0000e-02 eta: 12:29:27 time: 0.1555 data_time: 0.0110 memory: 7124 grad_norm: 5.3507 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.0043 loss: 2.0043 2022/09/07 00:14:32 - mmengine - INFO - Epoch(train) [25][2400/3757] lr: 1.0000e-02 eta: 12:29:11 time: 0.1576 data_time: 0.0106 memory: 7124 grad_norm: 5.0938 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.8043 loss: 1.8043 2022/09/07 00:14:48 - mmengine - INFO - Epoch(train) [25][2500/3757] lr: 1.0000e-02 eta: 12:28:55 time: 0.1620 data_time: 0.0095 memory: 7124 grad_norm: 5.3041 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 1.9166 loss: 1.9166 2022/09/07 00:15:04 - mmengine - INFO - Epoch(train) [25][2600/3757] lr: 1.0000e-02 eta: 12:28:39 time: 0.1602 data_time: 0.0116 memory: 7124 grad_norm: 5.1841 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9593 loss: 1.9593 2022/09/07 00:15:20 - mmengine - INFO - Epoch(train) [25][2700/3757] lr: 1.0000e-02 eta: 12:28:23 time: 0.1577 data_time: 0.0105 memory: 7124 grad_norm: 5.3352 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.8191 loss: 1.8191 2022/09/07 00:15:36 - mmengine - INFO - Epoch(train) [25][2800/3757] lr: 1.0000e-02 eta: 12:28:06 time: 0.1564 data_time: 0.0103 memory: 7124 grad_norm: 5.0364 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.8623 loss: 1.8623 2022/09/07 00:15:41 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:15:51 - mmengine - INFO - Epoch(train) [25][2900/3757] lr: 1.0000e-02 eta: 12:27:50 time: 0.1548 data_time: 0.0093 memory: 7124 grad_norm: 5.1854 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.9787 loss: 1.9787 2022/09/07 00:16:07 - mmengine - INFO - Epoch(train) [25][3000/3757] lr: 1.0000e-02 eta: 12:27:34 time: 0.1550 data_time: 0.0097 memory: 7124 grad_norm: 5.2046 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.1109 loss: 2.1109 2022/09/07 00:16:23 - mmengine - INFO - Epoch(train) [25][3100/3757] lr: 1.0000e-02 eta: 12:27:17 time: 0.1523 data_time: 0.0095 memory: 7124 grad_norm: 5.1690 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.1389 loss: 2.1389 2022/09/07 00:16:39 - mmengine - INFO - Epoch(train) [25][3200/3757] lr: 1.0000e-02 eta: 12:27:01 time: 0.1617 data_time: 0.0109 memory: 7124 grad_norm: 5.2632 top1_acc: 0.2500 top5_acc: 0.8750 loss_cls: 1.9532 loss: 1.9532 2022/09/07 00:16:54 - mmengine - INFO - Epoch(train) [25][3300/3757] lr: 1.0000e-02 eta: 12:26:45 time: 0.1567 data_time: 0.0095 memory: 7124 grad_norm: 5.2185 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.0130 loss: 2.0130 2022/09/07 00:17:10 - mmengine - INFO - Epoch(train) [25][3400/3757] lr: 1.0000e-02 eta: 12:26:28 time: 0.1547 data_time: 0.0091 memory: 7124 grad_norm: 5.1523 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8859 loss: 1.8859 2022/09/07 00:17:26 - mmengine - INFO - Epoch(train) [25][3500/3757] lr: 1.0000e-02 eta: 12:26:12 time: 0.1572 data_time: 0.0091 memory: 7124 grad_norm: 5.1204 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9811 loss: 1.9811 2022/09/07 00:17:42 - mmengine - INFO - Epoch(train) [25][3600/3757] lr: 1.0000e-02 eta: 12:25:57 time: 0.1549 data_time: 0.0116 memory: 7124 grad_norm: 5.4091 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.4944 loss: 1.4944 2022/09/07 00:17:58 - mmengine - INFO - Epoch(train) [25][3700/3757] lr: 1.0000e-02 eta: 12:25:40 time: 0.1539 data_time: 0.0103 memory: 7124 grad_norm: 5.3064 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7565 loss: 1.7565 2022/09/07 00:18:06 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:18:06 - mmengine - INFO - Epoch(train) [25][3757/3757] lr: 1.0000e-02 eta: 12:25:34 time: 0.1350 data_time: 0.0070 memory: 7124 grad_norm: 5.2580 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 2.0666 loss: 2.0666 2022/09/07 00:18:06 - mmengine - INFO - Saving checkpoint at 25 epochs 2022/09/07 00:20:23 - mmengine - INFO - Epoch(val) [25][100/310] eta: 0:03:39 time: 1.0473 data_time: 0.7521 memory: 7627 2022/09/07 00:22:43 - mmengine - INFO - Epoch(val) [25][200/310] eta: 0:02:30 time: 1.3721 data_time: 1.0728 memory: 7627 2022/09/07 00:24:46 - mmengine - INFO - Epoch(val) [25][300/310] eta: 0:00:11 time: 1.1271 data_time: 0.8294 memory: 7627 2022/09/07 00:25:03 - mmengine - INFO - Epoch(val) [25][310/310] acc/top1: 0.6425 acc/top5: 0.8627 acc/mean1: 0.6424 2022/09/07 00:25:03 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_24.pth is removed 2022/09/07 00:25:05 - mmengine - INFO - The best checkpoint with 0.6425 acc/top1 at 25 epoch is saved to best_acc/top1_epoch_25.pth. 2022/09/07 00:25:17 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:25:21 - mmengine - INFO - Epoch(train) [26][100/3757] lr: 1.0000e-02 eta: 12:25:09 time: 0.1584 data_time: 0.0092 memory: 7627 grad_norm: 5.4488 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.0833 loss: 2.0833 2022/09/07 00:25:37 - mmengine - INFO - Epoch(train) [26][200/3757] lr: 1.0000e-02 eta: 12:24:53 time: 0.1606 data_time: 0.0114 memory: 7124 grad_norm: 5.2274 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0404 loss: 2.0404 2022/09/07 00:25:53 - mmengine - INFO - Epoch(train) [26][300/3757] lr: 1.0000e-02 eta: 12:24:37 time: 0.1593 data_time: 0.0099 memory: 7124 grad_norm: 5.3810 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.1091 loss: 2.1091 2022/09/07 00:26:09 - mmengine - INFO - Epoch(train) [26][400/3757] lr: 1.0000e-02 eta: 12:24:21 time: 0.1521 data_time: 0.0098 memory: 7124 grad_norm: 5.2652 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8070 loss: 1.8070 2022/09/07 00:26:25 - mmengine - INFO - Epoch(train) [26][500/3757] lr: 1.0000e-02 eta: 12:24:04 time: 0.1535 data_time: 0.0095 memory: 7124 grad_norm: 5.1267 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0971 loss: 2.0971 2022/09/07 00:26:41 - mmengine - INFO - Epoch(train) [26][600/3757] lr: 1.0000e-02 eta: 12:23:49 time: 0.1536 data_time: 0.0106 memory: 7124 grad_norm: 5.1748 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 1.8889 loss: 1.8889 2022/09/07 00:26:57 - mmengine - INFO - Epoch(train) [26][700/3757] lr: 1.0000e-02 eta: 12:23:33 time: 0.1579 data_time: 0.0103 memory: 7124 grad_norm: 5.2410 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8877 loss: 1.8877 2022/09/07 00:27:12 - mmengine - INFO - Epoch(train) [26][800/3757] lr: 1.0000e-02 eta: 12:23:16 time: 0.1541 data_time: 0.0097 memory: 7124 grad_norm: 5.2276 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9637 loss: 1.9637 2022/09/07 00:27:28 - mmengine - INFO - Epoch(train) [26][900/3757] lr: 1.0000e-02 eta: 12:23:00 time: 0.1558 data_time: 0.0100 memory: 7124 grad_norm: 5.1856 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.8039 loss: 1.8039 2022/09/07 00:27:44 - mmengine - INFO - Epoch(train) [26][1000/3757] lr: 1.0000e-02 eta: 12:22:44 time: 0.1617 data_time: 0.0093 memory: 7124 grad_norm: 5.1843 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.1812 loss: 2.1812 2022/09/07 00:27:56 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:28:00 - mmengine - INFO - Epoch(train) [26][1100/3757] lr: 1.0000e-02 eta: 12:22:28 time: 0.1552 data_time: 0.0085 memory: 7124 grad_norm: 5.0289 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6632 loss: 1.6632 2022/09/07 00:28:15 - mmengine - INFO - Epoch(train) [26][1200/3757] lr: 1.0000e-02 eta: 12:22:11 time: 0.1553 data_time: 0.0096 memory: 7124 grad_norm: 5.1360 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7588 loss: 1.7588 2022/09/07 00:28:32 - mmengine - INFO - Epoch(train) [26][1300/3757] lr: 1.0000e-02 eta: 12:21:56 time: 0.1845 data_time: 0.0097 memory: 7124 grad_norm: 5.3203 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.6630 loss: 1.6630 2022/09/07 00:28:47 - mmengine - INFO - Epoch(train) [26][1400/3757] lr: 1.0000e-02 eta: 12:21:40 time: 0.1573 data_time: 0.0098 memory: 7124 grad_norm: 5.3880 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6830 loss: 1.6830 2022/09/07 00:29:03 - mmengine - INFO - Epoch(train) [26][1500/3757] lr: 1.0000e-02 eta: 12:21:24 time: 0.1583 data_time: 0.0107 memory: 7124 grad_norm: 5.1081 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8625 loss: 1.8625 2022/09/07 00:29:19 - mmengine - INFO - Epoch(train) [26][1600/3757] lr: 1.0000e-02 eta: 12:21:08 time: 0.1609 data_time: 0.0096 memory: 7124 grad_norm: 5.2253 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7856 loss: 1.7856 2022/09/07 00:29:35 - mmengine - INFO - Epoch(train) [26][1700/3757] lr: 1.0000e-02 eta: 12:20:51 time: 0.1534 data_time: 0.0099 memory: 7124 grad_norm: 5.1901 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.1712 loss: 2.1712 2022/09/07 00:29:51 - mmengine - INFO - Epoch(train) [26][1800/3757] lr: 1.0000e-02 eta: 12:20:35 time: 0.1647 data_time: 0.0115 memory: 7124 grad_norm: 5.4629 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8254 loss: 1.8254 2022/09/07 00:30:06 - mmengine - INFO - Epoch(train) [26][1900/3757] lr: 1.0000e-02 eta: 12:20:19 time: 0.1545 data_time: 0.0090 memory: 7124 grad_norm: 5.2803 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.6242 loss: 1.6242 2022/09/07 00:30:22 - mmengine - INFO - Epoch(train) [26][2000/3757] lr: 1.0000e-02 eta: 12:20:03 time: 0.1542 data_time: 0.0098 memory: 7124 grad_norm: 4.9709 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8720 loss: 1.8720 2022/09/07 00:30:35 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:30:38 - mmengine - INFO - Epoch(train) [26][2100/3757] lr: 1.0000e-02 eta: 12:19:48 time: 0.1543 data_time: 0.0097 memory: 7124 grad_norm: 4.9921 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6660 loss: 1.6660 2022/09/07 00:30:54 - mmengine - INFO - Epoch(train) [26][2200/3757] lr: 1.0000e-02 eta: 12:19:32 time: 0.1539 data_time: 0.0097 memory: 7124 grad_norm: 5.3230 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8517 loss: 1.8517 2022/09/07 00:31:10 - mmengine - INFO - Epoch(train) [26][2300/3757] lr: 1.0000e-02 eta: 12:19:15 time: 0.1574 data_time: 0.0095 memory: 7124 grad_norm: 5.4474 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9479 loss: 1.9479 2022/09/07 00:31:26 - mmengine - INFO - Epoch(train) [26][2400/3757] lr: 1.0000e-02 eta: 12:18:58 time: 0.1534 data_time: 0.0095 memory: 7124 grad_norm: 4.9988 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.8663 loss: 1.8663 2022/09/07 00:31:41 - mmengine - INFO - Epoch(train) [26][2500/3757] lr: 1.0000e-02 eta: 12:18:42 time: 0.1626 data_time: 0.0092 memory: 7124 grad_norm: 5.2195 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9416 loss: 1.9416 2022/09/07 00:31:57 - mmengine - INFO - Epoch(train) [26][2600/3757] lr: 1.0000e-02 eta: 12:18:26 time: 0.1538 data_time: 0.0104 memory: 7124 grad_norm: 5.2865 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.9200 loss: 1.9200 2022/09/07 00:32:13 - mmengine - INFO - Epoch(train) [26][2700/3757] lr: 1.0000e-02 eta: 12:18:10 time: 0.1561 data_time: 0.0098 memory: 7124 grad_norm: 5.2025 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.6473 loss: 1.6473 2022/09/07 00:32:29 - mmengine - INFO - Epoch(train) [26][2800/3757] lr: 1.0000e-02 eta: 12:17:53 time: 0.1561 data_time: 0.0094 memory: 7124 grad_norm: 5.4571 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.6424 loss: 1.6424 2022/09/07 00:32:45 - mmengine - INFO - Epoch(train) [26][2900/3757] lr: 1.0000e-02 eta: 12:17:37 time: 0.1572 data_time: 0.0098 memory: 7124 grad_norm: 5.1163 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8254 loss: 1.8254 2022/09/07 00:33:01 - mmengine - INFO - Epoch(train) [26][3000/3757] lr: 1.0000e-02 eta: 12:17:22 time: 0.1587 data_time: 0.0102 memory: 7124 grad_norm: 5.2684 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.9527 loss: 1.9527 2022/09/07 00:33:12 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:33:16 - mmengine - INFO - Epoch(train) [26][3100/3757] lr: 1.0000e-02 eta: 12:17:05 time: 0.1555 data_time: 0.0095 memory: 7124 grad_norm: 5.4118 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9159 loss: 1.9159 2022/09/07 00:33:32 - mmengine - INFO - Epoch(train) [26][3200/3757] lr: 1.0000e-02 eta: 12:16:49 time: 0.1578 data_time: 0.0114 memory: 7124 grad_norm: 5.1372 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8160 loss: 1.8160 2022/09/07 00:33:48 - mmengine - INFO - Epoch(train) [26][3300/3757] lr: 1.0000e-02 eta: 12:16:33 time: 0.1587 data_time: 0.0097 memory: 7124 grad_norm: 5.2422 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.0425 loss: 2.0425 2022/09/07 00:34:04 - mmengine - INFO - Epoch(train) [26][3400/3757] lr: 1.0000e-02 eta: 12:16:16 time: 0.1559 data_time: 0.0093 memory: 7124 grad_norm: 5.1871 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9804 loss: 1.9804 2022/09/07 00:34:19 - mmengine - INFO - Epoch(train) [26][3500/3757] lr: 1.0000e-02 eta: 12:16:00 time: 0.1578 data_time: 0.0093 memory: 7124 grad_norm: 5.2493 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.9875 loss: 1.9875 2022/09/07 00:34:35 - mmengine - INFO - Epoch(train) [26][3600/3757] lr: 1.0000e-02 eta: 12:15:44 time: 0.1555 data_time: 0.0107 memory: 7124 grad_norm: 5.2457 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.2446 loss: 2.2446 2022/09/07 00:34:51 - mmengine - INFO - Epoch(train) [26][3700/3757] lr: 1.0000e-02 eta: 12:15:28 time: 0.1593 data_time: 0.0097 memory: 7124 grad_norm: 5.1958 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.1419 loss: 2.1419 2022/09/07 00:35:00 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:35:00 - mmengine - INFO - Epoch(train) [26][3757/3757] lr: 1.0000e-02 eta: 12:15:21 time: 0.1361 data_time: 0.0067 memory: 7124 grad_norm: 5.3128 top1_acc: 0.1429 top5_acc: 0.7143 loss_cls: 2.0480 loss: 2.0480 2022/09/07 00:35:00 - mmengine - INFO - Saving checkpoint at 26 epochs 2022/09/07 00:37:20 - mmengine - INFO - Epoch(val) [26][100/310] eta: 0:04:39 time: 1.3307 data_time: 1.0175 memory: 7627 2022/09/07 00:39:33 - mmengine - INFO - Epoch(val) [26][200/310] eta: 0:02:04 time: 1.1360 data_time: 0.8339 memory: 7627 2022/09/07 00:41:41 - mmengine - INFO - Epoch(val) [26][300/310] eta: 0:00:12 time: 1.2854 data_time: 0.9876 memory: 7627 2022/09/07 00:41:57 - mmengine - INFO - Epoch(val) [26][310/310] acc/top1: 0.6456 acc/top5: 0.8613 acc/mean1: 0.6454 2022/09/07 00:41:57 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_25.pth is removed 2022/09/07 00:41:59 - mmengine - INFO - The best checkpoint with 0.6456 acc/top1 at 26 epoch is saved to best_acc/top1_epoch_26.pth. 2022/09/07 00:42:16 - mmengine - INFO - Epoch(train) [27][100/3757] lr: 1.0000e-02 eta: 12:14:57 time: 0.1539 data_time: 0.0096 memory: 7627 grad_norm: 5.2815 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.9534 loss: 1.9534 2022/09/07 00:42:32 - mmengine - INFO - Epoch(train) [27][200/3757] lr: 1.0000e-02 eta: 12:14:42 time: 0.1671 data_time: 0.0099 memory: 7124 grad_norm: 5.0235 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.5086 loss: 1.5086 2022/09/07 00:42:48 - mmengine - INFO - Epoch(train) [27][300/3757] lr: 1.0000e-02 eta: 12:14:26 time: 0.1569 data_time: 0.0101 memory: 7124 grad_norm: 5.3331 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.6999 loss: 1.6999 2022/09/07 00:42:51 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:43:04 - mmengine - INFO - Epoch(train) [27][400/3757] lr: 1.0000e-02 eta: 12:14:10 time: 0.1559 data_time: 0.0091 memory: 7124 grad_norm: 5.1308 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8144 loss: 1.8144 2022/09/07 00:43:19 - mmengine - INFO - Epoch(train) [27][500/3757] lr: 1.0000e-02 eta: 12:13:53 time: 0.1556 data_time: 0.0099 memory: 7124 grad_norm: 5.2295 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.7277 loss: 1.7277 2022/09/07 00:43:35 - mmengine - INFO - Epoch(train) [27][600/3757] lr: 1.0000e-02 eta: 12:13:37 time: 0.1574 data_time: 0.0099 memory: 7124 grad_norm: 5.2630 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.7504 loss: 1.7504 2022/09/07 00:43:51 - mmengine - INFO - Epoch(train) [27][700/3757] lr: 1.0000e-02 eta: 12:13:21 time: 0.1594 data_time: 0.0099 memory: 7124 grad_norm: 5.2487 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.9785 loss: 1.9785 2022/09/07 00:44:07 - mmengine - INFO - Epoch(train) [27][800/3757] lr: 1.0000e-02 eta: 12:13:05 time: 0.1552 data_time: 0.0108 memory: 7124 grad_norm: 5.4232 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.8988 loss: 1.8988 2022/09/07 00:44:23 - mmengine - INFO - Epoch(train) [27][900/3757] lr: 1.0000e-02 eta: 12:12:49 time: 0.1696 data_time: 0.0087 memory: 7124 grad_norm: 5.5317 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.8857 loss: 1.8857 2022/09/07 00:44:39 - mmengine - INFO - Epoch(train) [27][1000/3757] lr: 1.0000e-02 eta: 12:12:33 time: 0.1566 data_time: 0.0116 memory: 7124 grad_norm: 5.0738 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.0318 loss: 2.0318 2022/09/07 00:44:55 - mmengine - INFO - Epoch(train) [27][1100/3757] lr: 1.0000e-02 eta: 12:12:18 time: 0.1555 data_time: 0.0100 memory: 7124 grad_norm: 5.3937 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7437 loss: 1.7437 2022/09/07 00:45:10 - mmengine - INFO - Epoch(train) [27][1200/3757] lr: 1.0000e-02 eta: 12:12:01 time: 0.1550 data_time: 0.0107 memory: 7124 grad_norm: 5.0827 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7574 loss: 1.7574 2022/09/07 00:45:26 - mmengine - INFO - Epoch(train) [27][1300/3757] lr: 1.0000e-02 eta: 12:11:45 time: 0.1571 data_time: 0.0095 memory: 7124 grad_norm: 5.4127 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.6232 loss: 1.6232 2022/09/07 00:45:29 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:45:42 - mmengine - INFO - Epoch(train) [27][1400/3757] lr: 1.0000e-02 eta: 12:11:29 time: 0.1560 data_time: 0.0095 memory: 7124 grad_norm: 5.1635 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.0166 loss: 2.0166 2022/09/07 00:45:58 - mmengine - INFO - Epoch(train) [27][1500/3757] lr: 1.0000e-02 eta: 12:11:12 time: 0.1511 data_time: 0.0101 memory: 7124 grad_norm: 5.6279 top1_acc: 0.2500 top5_acc: 0.2500 loss_cls: 1.9388 loss: 1.9388 2022/09/07 00:46:13 - mmengine - INFO - Epoch(train) [27][1600/3757] lr: 1.0000e-02 eta: 12:10:56 time: 0.1597 data_time: 0.0100 memory: 7124 grad_norm: 5.2160 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9185 loss: 1.9185 2022/09/07 00:46:29 - mmengine - INFO - Epoch(train) [27][1700/3757] lr: 1.0000e-02 eta: 12:10:40 time: 0.1560 data_time: 0.0090 memory: 7124 grad_norm: 5.0810 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.7454 loss: 1.7454 2022/09/07 00:46:45 - mmengine - INFO - Epoch(train) [27][1800/3757] lr: 1.0000e-02 eta: 12:10:24 time: 0.1589 data_time: 0.0099 memory: 7124 grad_norm: 5.1281 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.0621 loss: 2.0621 2022/09/07 00:47:01 - mmengine - INFO - Epoch(train) [27][1900/3757] lr: 1.0000e-02 eta: 12:10:08 time: 0.1536 data_time: 0.0094 memory: 7124 grad_norm: 5.5453 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.1708 loss: 2.1708 2022/09/07 00:47:17 - mmengine - INFO - Epoch(train) [27][2000/3757] lr: 1.0000e-02 eta: 12:09:51 time: 0.1570 data_time: 0.0106 memory: 7124 grad_norm: 5.1263 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.9346 loss: 1.9346 2022/09/07 00:47:33 - mmengine - INFO - Epoch(train) [27][2100/3757] lr: 1.0000e-02 eta: 12:09:36 time: 0.1587 data_time: 0.0098 memory: 7124 grad_norm: 5.1765 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.0331 loss: 2.0331 2022/09/07 00:47:49 - mmengine - INFO - Epoch(train) [27][2200/3757] lr: 1.0000e-02 eta: 12:09:20 time: 0.1555 data_time: 0.0109 memory: 7124 grad_norm: 5.1549 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.9596 loss: 1.9596 2022/09/07 00:48:05 - mmengine - INFO - Epoch(train) [27][2300/3757] lr: 1.0000e-02 eta: 12:09:04 time: 0.1564 data_time: 0.0102 memory: 7124 grad_norm: 5.3535 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.5995 loss: 1.5995 2022/09/07 00:48:07 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:48:21 - mmengine - INFO - Epoch(train) [27][2400/3757] lr: 1.0000e-02 eta: 12:08:49 time: 0.1577 data_time: 0.0110 memory: 7124 grad_norm: 5.3367 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.4903 loss: 1.4903 2022/09/07 00:48:37 - mmengine - INFO - Epoch(train) [27][2500/3757] lr: 1.0000e-02 eta: 12:08:34 time: 0.1640 data_time: 0.0085 memory: 7124 grad_norm: 5.3496 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.8139 loss: 1.8139 2022/09/07 00:48:53 - mmengine - INFO - Epoch(train) [27][2600/3757] lr: 1.0000e-02 eta: 12:08:17 time: 0.1644 data_time: 0.0104 memory: 7124 grad_norm: 5.1891 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.1080 loss: 2.1080 2022/09/07 00:49:08 - mmengine - INFO - Epoch(train) [27][2700/3757] lr: 1.0000e-02 eta: 12:08:01 time: 0.1569 data_time: 0.0093 memory: 7124 grad_norm: 5.0965 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.9324 loss: 1.9324 2022/09/07 00:49:24 - mmengine - INFO - Epoch(train) [27][2800/3757] lr: 1.0000e-02 eta: 12:07:45 time: 0.1559 data_time: 0.0111 memory: 7124 grad_norm: 5.2891 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8374 loss: 1.8374 2022/09/07 00:49:40 - mmengine - INFO - Epoch(train) [27][2900/3757] lr: 1.0000e-02 eta: 12:07:30 time: 0.1554 data_time: 0.0089 memory: 7124 grad_norm: 5.1961 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8854 loss: 1.8854 2022/09/07 00:49:56 - mmengine - INFO - Epoch(train) [27][3000/3757] lr: 1.0000e-02 eta: 12:07:15 time: 0.1535 data_time: 0.0114 memory: 7124 grad_norm: 5.3581 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.8532 loss: 1.8532 2022/09/07 00:50:12 - mmengine - INFO - Epoch(train) [27][3100/3757] lr: 1.0000e-02 eta: 12:06:59 time: 0.1656 data_time: 0.0121 memory: 7124 grad_norm: 5.2497 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.1102 loss: 2.1102 2022/09/07 00:50:28 - mmengine - INFO - Epoch(train) [27][3200/3757] lr: 1.0000e-02 eta: 12:06:43 time: 0.1573 data_time: 0.0099 memory: 7124 grad_norm: 5.1881 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1534 loss: 2.1534 2022/09/07 00:50:44 - mmengine - INFO - Epoch(train) [27][3300/3757] lr: 1.0000e-02 eta: 12:06:27 time: 0.1601 data_time: 0.0133 memory: 7124 grad_norm: 5.2258 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7554 loss: 1.7554 2022/09/07 00:50:47 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:51:00 - mmengine - INFO - Epoch(train) [27][3400/3757] lr: 1.0000e-02 eta: 12:06:11 time: 0.1574 data_time: 0.0109 memory: 7124 grad_norm: 5.0457 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7544 loss: 1.7544 2022/09/07 00:51:16 - mmengine - INFO - Epoch(train) [27][3500/3757] lr: 1.0000e-02 eta: 12:05:54 time: 0.1552 data_time: 0.0093 memory: 7124 grad_norm: 5.2488 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9640 loss: 1.9640 2022/09/07 00:51:32 - mmengine - INFO - Epoch(train) [27][3600/3757] lr: 1.0000e-02 eta: 12:05:38 time: 0.1660 data_time: 0.0105 memory: 7124 grad_norm: 5.1668 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.0640 loss: 2.0640 2022/09/07 00:51:47 - mmengine - INFO - Epoch(train) [27][3700/3757] lr: 1.0000e-02 eta: 12:05:22 time: 0.1570 data_time: 0.0107 memory: 7124 grad_norm: 5.2342 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.7881 loss: 1.7881 2022/09/07 00:51:56 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 00:51:56 - mmengine - INFO - Epoch(train) [27][3757/3757] lr: 1.0000e-02 eta: 12:05:16 time: 0.1365 data_time: 0.0073 memory: 7124 grad_norm: 5.3041 top1_acc: 0.1429 top5_acc: 0.2857 loss_cls: 2.0809 loss: 2.0809 2022/09/07 00:51:56 - mmengine - INFO - Saving checkpoint at 27 epochs 2022/09/07 00:54:16 - mmengine - INFO - Epoch(val) [27][100/310] eta: 0:04:22 time: 1.2485 data_time: 0.9496 memory: 7627 2022/09/07 00:56:29 - mmengine - INFO - Epoch(val) [27][200/310] eta: 0:02:06 time: 1.1469 data_time: 0.8481 memory: 7627 2022/09/07 00:58:37 - mmengine - INFO - Epoch(val) [27][300/310] eta: 0:00:13 time: 1.3324 data_time: 1.0325 memory: 7627 2022/09/07 00:58:54 - mmengine - INFO - Epoch(val) [27][310/310] acc/top1: 0.6339 acc/top5: 0.8561 acc/mean1: 0.6337 2022/09/07 00:59:12 - mmengine - INFO - Epoch(train) [28][100/3757] lr: 1.0000e-02 eta: 12:04:56 time: 0.1564 data_time: 0.0100 memory: 7627 grad_norm: 5.3462 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9751 loss: 1.9751 2022/09/07 00:59:28 - mmengine - INFO - Epoch(train) [28][200/3757] lr: 1.0000e-02 eta: 12:04:40 time: 0.1540 data_time: 0.0099 memory: 7124 grad_norm: 5.3063 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.7069 loss: 1.7069 2022/09/07 00:59:44 - mmengine - INFO - Epoch(train) [28][300/3757] lr: 1.0000e-02 eta: 12:04:24 time: 0.1552 data_time: 0.0098 memory: 7124 grad_norm: 5.2747 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8113 loss: 1.8113 2022/09/07 01:00:00 - mmengine - INFO - Epoch(train) [28][400/3757] lr: 1.0000e-02 eta: 12:04:09 time: 0.1625 data_time: 0.0082 memory: 7124 grad_norm: 5.3094 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.7088 loss: 1.7088 2022/09/07 01:00:16 - mmengine - INFO - Epoch(train) [28][500/3757] lr: 1.0000e-02 eta: 12:03:55 time: 0.1562 data_time: 0.0089 memory: 7124 grad_norm: 5.3213 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.0423 loss: 2.0423 2022/09/07 01:00:26 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:00:32 - mmengine - INFO - Epoch(train) [28][600/3757] lr: 1.0000e-02 eta: 12:03:39 time: 0.1593 data_time: 0.0090 memory: 7124 grad_norm: 5.2160 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6106 loss: 1.6106 2022/09/07 01:00:48 - mmengine - INFO - Epoch(train) [28][700/3757] lr: 1.0000e-02 eta: 12:03:23 time: 0.1570 data_time: 0.0096 memory: 7124 grad_norm: 5.3584 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8727 loss: 1.8727 2022/09/07 01:01:04 - mmengine - INFO - Epoch(train) [28][800/3757] lr: 1.0000e-02 eta: 12:03:07 time: 0.1581 data_time: 0.0098 memory: 7124 grad_norm: 5.0973 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.9229 loss: 1.9229 2022/09/07 01:01:20 - mmengine - INFO - Epoch(train) [28][900/3757] lr: 1.0000e-02 eta: 12:02:51 time: 0.1589 data_time: 0.0119 memory: 7124 grad_norm: 5.1977 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9711 loss: 1.9711 2022/09/07 01:01:36 - mmengine - INFO - Epoch(train) [28][1000/3757] lr: 1.0000e-02 eta: 12:02:36 time: 0.1543 data_time: 0.0100 memory: 7124 grad_norm: 5.1384 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 1.6409 loss: 1.6409 2022/09/07 01:01:52 - mmengine - INFO - Epoch(train) [28][1100/3757] lr: 1.0000e-02 eta: 12:02:20 time: 0.1591 data_time: 0.0107 memory: 7124 grad_norm: 5.2928 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.7172 loss: 1.7172 2022/09/07 01:02:08 - mmengine - INFO - Epoch(train) [28][1200/3757] lr: 1.0000e-02 eta: 12:02:04 time: 0.1566 data_time: 0.0091 memory: 7124 grad_norm: 5.0621 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.0678 loss: 2.0678 2022/09/07 01:02:24 - mmengine - INFO - Epoch(train) [28][1300/3757] lr: 1.0000e-02 eta: 12:01:48 time: 0.1607 data_time: 0.0111 memory: 7124 grad_norm: 5.2154 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7525 loss: 1.7525 2022/09/07 01:02:40 - mmengine - INFO - Epoch(train) [28][1400/3757] lr: 1.0000e-02 eta: 12:01:32 time: 0.1579 data_time: 0.0102 memory: 7124 grad_norm: 5.2963 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9531 loss: 1.9531 2022/09/07 01:02:56 - mmengine - INFO - Epoch(train) [28][1500/3757] lr: 1.0000e-02 eta: 12:01:16 time: 0.1555 data_time: 0.0095 memory: 7124 grad_norm: 5.1069 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9763 loss: 1.9763 2022/09/07 01:03:05 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:03:11 - mmengine - INFO - Epoch(train) [28][1600/3757] lr: 1.0000e-02 eta: 12:01:00 time: 0.1586 data_time: 0.0103 memory: 7124 grad_norm: 5.2800 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.6392 loss: 1.6392 2022/09/07 01:03:27 - mmengine - INFO - Epoch(train) [28][1700/3757] lr: 1.0000e-02 eta: 12:00:44 time: 0.1561 data_time: 0.0113 memory: 7124 grad_norm: 5.2929 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6777 loss: 1.6777 2022/09/07 01:03:43 - mmengine - INFO - Epoch(train) [28][1800/3757] lr: 1.0000e-02 eta: 12:00:29 time: 0.1621 data_time: 0.0112 memory: 7124 grad_norm: 5.3542 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.0531 loss: 2.0531 2022/09/07 01:03:59 - mmengine - INFO - Epoch(train) [28][1900/3757] lr: 1.0000e-02 eta: 12:00:13 time: 0.1578 data_time: 0.0102 memory: 7124 grad_norm: 5.2450 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.7769 loss: 1.7769 2022/09/07 01:04:15 - mmengine - INFO - Epoch(train) [28][2000/3757] lr: 1.0000e-02 eta: 11:59:58 time: 0.1691 data_time: 0.0127 memory: 7124 grad_norm: 5.2845 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9032 loss: 1.9032 2022/09/07 01:04:31 - mmengine - INFO - Epoch(train) [28][2100/3757] lr: 1.0000e-02 eta: 11:59:41 time: 0.1598 data_time: 0.0099 memory: 7124 grad_norm: 5.3466 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9621 loss: 1.9621 2022/09/07 01:04:47 - mmengine - INFO - Epoch(train) [28][2200/3757] lr: 1.0000e-02 eta: 11:59:25 time: 0.1558 data_time: 0.0092 memory: 7124 grad_norm: 5.4634 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.2183 loss: 2.2183 2022/09/07 01:05:03 - mmengine - INFO - Epoch(train) [28][2300/3757] lr: 1.0000e-02 eta: 11:59:10 time: 0.1639 data_time: 0.0118 memory: 7124 grad_norm: 5.2324 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7084 loss: 1.7084 2022/09/07 01:05:19 - mmengine - INFO - Epoch(train) [28][2400/3757] lr: 1.0000e-02 eta: 11:58:53 time: 0.1547 data_time: 0.0112 memory: 7124 grad_norm: 5.2644 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.7102 loss: 1.7102 2022/09/07 01:05:35 - mmengine - INFO - Epoch(train) [28][2500/3757] lr: 1.0000e-02 eta: 11:58:38 time: 0.1664 data_time: 0.0131 memory: 7124 grad_norm: 5.1575 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8849 loss: 1.8849 2022/09/07 01:05:44 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:05:50 - mmengine - INFO - Epoch(train) [28][2600/3757] lr: 1.0000e-02 eta: 11:58:21 time: 0.1578 data_time: 0.0099 memory: 7124 grad_norm: 5.3115 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6871 loss: 1.6871 2022/09/07 01:06:07 - mmengine - INFO - Epoch(train) [28][2700/3757] lr: 1.0000e-02 eta: 11:58:06 time: 0.1532 data_time: 0.0094 memory: 7124 grad_norm: 5.1027 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.9083 loss: 1.9083 2022/09/07 01:06:23 - mmengine - INFO - Epoch(train) [28][2800/3757] lr: 1.0000e-02 eta: 11:57:50 time: 0.1587 data_time: 0.0093 memory: 7124 grad_norm: 5.3676 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9123 loss: 1.9123 2022/09/07 01:06:39 - mmengine - INFO - Epoch(train) [28][2900/3757] lr: 1.0000e-02 eta: 11:57:35 time: 0.1588 data_time: 0.0108 memory: 7124 grad_norm: 5.1823 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.1339 loss: 2.1339 2022/09/07 01:06:54 - mmengine - INFO - Epoch(train) [28][3000/3757] lr: 1.0000e-02 eta: 11:57:19 time: 0.1634 data_time: 0.0109 memory: 7124 grad_norm: 5.1205 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6368 loss: 1.6368 2022/09/07 01:07:10 - mmengine - INFO - Epoch(train) [28][3100/3757] lr: 1.0000e-02 eta: 11:57:03 time: 0.1572 data_time: 0.0100 memory: 7124 grad_norm: 5.4416 top1_acc: 0.3750 top5_acc: 0.3750 loss_cls: 1.9664 loss: 1.9664 2022/09/07 01:07:26 - mmengine - INFO - Epoch(train) [28][3200/3757] lr: 1.0000e-02 eta: 11:56:47 time: 0.1541 data_time: 0.0101 memory: 7124 grad_norm: 5.1229 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 2.0953 loss: 2.0953 2022/09/07 01:07:42 - mmengine - INFO - Epoch(train) [28][3300/3757] lr: 1.0000e-02 eta: 11:56:31 time: 0.1604 data_time: 0.0100 memory: 7124 grad_norm: 5.2890 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.1719 loss: 2.1719 2022/09/07 01:07:58 - mmengine - INFO - Epoch(train) [28][3400/3757] lr: 1.0000e-02 eta: 11:56:15 time: 0.1545 data_time: 0.0107 memory: 7124 grad_norm: 5.1346 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8117 loss: 1.8117 2022/09/07 01:08:14 - mmengine - INFO - Epoch(train) [28][3500/3757] lr: 1.0000e-02 eta: 11:56:00 time: 0.1725 data_time: 0.0250 memory: 7124 grad_norm: 5.2267 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6581 loss: 1.6581 2022/09/07 01:08:24 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:08:30 - mmengine - INFO - Epoch(train) [28][3600/3757] lr: 1.0000e-02 eta: 11:55:44 time: 0.1592 data_time: 0.0114 memory: 7124 grad_norm: 5.4468 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.0404 loss: 2.0404 2022/09/07 01:08:46 - mmengine - INFO - Epoch(train) [28][3700/3757] lr: 1.0000e-02 eta: 11:55:28 time: 0.1572 data_time: 0.0092 memory: 7124 grad_norm: 5.5135 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 2.1244 loss: 2.1244 2022/09/07 01:08:55 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:08:55 - mmengine - INFO - Epoch(train) [28][3757/3757] lr: 1.0000e-02 eta: 11:55:21 time: 0.1414 data_time: 0.0088 memory: 7124 grad_norm: 5.2488 top1_acc: 0.5714 top5_acc: 1.0000 loss_cls: 2.1454 loss: 2.1454 2022/09/07 01:08:55 - mmengine - INFO - Saving checkpoint at 28 epochs 2022/09/07 01:11:14 - mmengine - INFO - Epoch(val) [28][100/310] eta: 0:04:09 time: 1.1858 data_time: 0.8862 memory: 7627 2022/09/07 01:13:31 - mmengine - INFO - Epoch(val) [28][200/310] eta: 0:02:28 time: 1.3518 data_time: 1.0484 memory: 7627 2022/09/07 01:15:36 - mmengine - INFO - Epoch(val) [28][300/310] eta: 0:00:11 time: 1.1061 data_time: 0.8051 memory: 7627 2022/09/07 01:15:52 - mmengine - INFO - Epoch(val) [28][310/310] acc/top1: 0.6403 acc/top5: 0.8596 acc/mean1: 0.6401 2022/09/07 01:16:10 - mmengine - INFO - Epoch(train) [29][100/3757] lr: 1.0000e-02 eta: 11:55:01 time: 0.1577 data_time: 0.0111 memory: 7627 grad_norm: 5.2221 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8680 loss: 1.8680 2022/09/07 01:16:26 - mmengine - INFO - Epoch(train) [29][200/3757] lr: 1.0000e-02 eta: 11:54:45 time: 0.1571 data_time: 0.0107 memory: 7124 grad_norm: 5.2004 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8415 loss: 1.8415 2022/09/07 01:16:42 - mmengine - INFO - Epoch(train) [29][300/3757] lr: 1.0000e-02 eta: 11:54:29 time: 0.1517 data_time: 0.0090 memory: 7124 grad_norm: 5.3477 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5984 loss: 1.5984 2022/09/07 01:16:58 - mmengine - INFO - Epoch(train) [29][400/3757] lr: 1.0000e-02 eta: 11:54:13 time: 0.1570 data_time: 0.0102 memory: 7124 grad_norm: 5.1311 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8818 loss: 1.8818 2022/09/07 01:17:14 - mmengine - INFO - Epoch(train) [29][500/3757] lr: 1.0000e-02 eta: 11:53:57 time: 0.1592 data_time: 0.0116 memory: 7124 grad_norm: 5.2169 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.6786 loss: 1.6786 2022/09/07 01:17:30 - mmengine - INFO - Epoch(train) [29][600/3757] lr: 1.0000e-02 eta: 11:53:41 time: 0.1579 data_time: 0.0106 memory: 7124 grad_norm: 5.3501 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6783 loss: 1.6783 2022/09/07 01:17:46 - mmengine - INFO - Epoch(train) [29][700/3757] lr: 1.0000e-02 eta: 11:53:25 time: 0.1593 data_time: 0.0099 memory: 7124 grad_norm: 5.4728 top1_acc: 0.2500 top5_acc: 0.8750 loss_cls: 1.8044 loss: 1.8044 2022/09/07 01:18:01 - mmengine - INFO - Epoch(train) [29][800/3757] lr: 1.0000e-02 eta: 11:53:09 time: 0.1580 data_time: 0.0088 memory: 7124 grad_norm: 5.2563 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.7845 loss: 1.7845 2022/09/07 01:18:02 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:18:17 - mmengine - INFO - Epoch(train) [29][900/3757] lr: 1.0000e-02 eta: 11:52:54 time: 0.1573 data_time: 0.0103 memory: 7124 grad_norm: 5.2397 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.6673 loss: 1.6673 2022/09/07 01:18:34 - mmengine - INFO - Epoch(train) [29][1000/3757] lr: 1.0000e-02 eta: 11:52:38 time: 0.1542 data_time: 0.0107 memory: 7124 grad_norm: 5.3862 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9806 loss: 1.9806 2022/09/07 01:18:49 - mmengine - INFO - Epoch(train) [29][1100/3757] lr: 1.0000e-02 eta: 11:52:22 time: 0.1553 data_time: 0.0102 memory: 7124 grad_norm: 5.4763 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9605 loss: 1.9605 2022/09/07 01:19:06 - mmengine - INFO - Epoch(train) [29][1200/3757] lr: 1.0000e-02 eta: 11:52:08 time: 0.1813 data_time: 0.0079 memory: 7124 grad_norm: 5.3257 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6764 loss: 1.6764 2022/09/07 01:19:22 - mmengine - INFO - Epoch(train) [29][1300/3757] lr: 1.0000e-02 eta: 11:51:52 time: 0.1570 data_time: 0.0098 memory: 7124 grad_norm: 5.2576 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8935 loss: 1.8935 2022/09/07 01:19:38 - mmengine - INFO - Epoch(train) [29][1400/3757] lr: 1.0000e-02 eta: 11:51:37 time: 0.1562 data_time: 0.0091 memory: 7124 grad_norm: 5.3850 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.7338 loss: 1.7338 2022/09/07 01:19:54 - mmengine - INFO - Epoch(train) [29][1500/3757] lr: 1.0000e-02 eta: 11:51:21 time: 0.1530 data_time: 0.0097 memory: 7124 grad_norm: 5.2486 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.6989 loss: 1.6989 2022/09/07 01:20:10 - mmengine - INFO - Epoch(train) [29][1600/3757] lr: 1.0000e-02 eta: 11:51:05 time: 0.1585 data_time: 0.0095 memory: 7124 grad_norm: 5.1886 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.6329 loss: 1.6329 2022/09/07 01:20:26 - mmengine - INFO - Epoch(train) [29][1700/3757] lr: 1.0000e-02 eta: 11:50:49 time: 0.1586 data_time: 0.0092 memory: 7124 grad_norm: 5.2133 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8933 loss: 1.8933 2022/09/07 01:20:41 - mmengine - INFO - Epoch(train) [29][1800/3757] lr: 1.0000e-02 eta: 11:50:32 time: 0.1548 data_time: 0.0094 memory: 7124 grad_norm: 5.3314 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8187 loss: 1.8187 2022/09/07 01:20:42 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:20:57 - mmengine - INFO - Epoch(train) [29][1900/3757] lr: 1.0000e-02 eta: 11:50:16 time: 0.1567 data_time: 0.0109 memory: 7124 grad_norm: 5.2743 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7314 loss: 1.7314 2022/09/07 01:21:13 - mmengine - INFO - Epoch(train) [29][2000/3757] lr: 1.0000e-02 eta: 11:50:00 time: 0.1555 data_time: 0.0098 memory: 7124 grad_norm: 5.2042 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9389 loss: 1.9389 2022/09/07 01:21:29 - mmengine - INFO - Epoch(train) [29][2100/3757] lr: 1.0000e-02 eta: 11:49:44 time: 0.1563 data_time: 0.0092 memory: 7124 grad_norm: 5.3160 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.7831 loss: 1.7831 2022/09/07 01:21:45 - mmengine - INFO - Epoch(train) [29][2200/3757] lr: 1.0000e-02 eta: 11:49:29 time: 0.1713 data_time: 0.0112 memory: 7124 grad_norm: 5.1586 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 1.6576 loss: 1.6576 2022/09/07 01:22:01 - mmengine - INFO - Epoch(train) [29][2300/3757] lr: 1.0000e-02 eta: 11:49:12 time: 0.1561 data_time: 0.0090 memory: 7124 grad_norm: 5.2312 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.4566 loss: 1.4566 2022/09/07 01:22:17 - mmengine - INFO - Epoch(train) [29][2400/3757] lr: 1.0000e-02 eta: 11:48:57 time: 0.1577 data_time: 0.0097 memory: 7124 grad_norm: 5.2469 top1_acc: 0.2500 top5_acc: 0.8750 loss_cls: 2.2298 loss: 2.2298 2022/09/07 01:22:32 - mmengine - INFO - Epoch(train) [29][2500/3757] lr: 1.0000e-02 eta: 11:48:41 time: 0.1563 data_time: 0.0104 memory: 7124 grad_norm: 5.2758 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8590 loss: 1.8590 2022/09/07 01:22:48 - mmengine - INFO - Epoch(train) [29][2600/3757] lr: 1.0000e-02 eta: 11:48:25 time: 0.1563 data_time: 0.0119 memory: 7124 grad_norm: 5.1050 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9520 loss: 1.9520 2022/09/07 01:23:04 - mmengine - INFO - Epoch(train) [29][2700/3757] lr: 1.0000e-02 eta: 11:48:08 time: 0.1539 data_time: 0.0110 memory: 7124 grad_norm: 5.2379 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9333 loss: 1.9333 2022/09/07 01:23:20 - mmengine - INFO - Epoch(train) [29][2800/3757] lr: 1.0000e-02 eta: 11:47:53 time: 0.1567 data_time: 0.0100 memory: 7124 grad_norm: 5.3266 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7424 loss: 1.7424 2022/09/07 01:23:21 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:23:36 - mmengine - INFO - Epoch(train) [29][2900/3757] lr: 1.0000e-02 eta: 11:47:37 time: 0.1596 data_time: 0.0107 memory: 7124 grad_norm: 5.2916 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8331 loss: 1.8331 2022/09/07 01:23:52 - mmengine - INFO - Epoch(train) [29][3000/3757] lr: 1.0000e-02 eta: 11:47:21 time: 0.1584 data_time: 0.0089 memory: 7124 grad_norm: 5.3881 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9310 loss: 1.9310 2022/09/07 01:24:08 - mmengine - INFO - Epoch(train) [29][3100/3757] lr: 1.0000e-02 eta: 11:47:05 time: 0.1546 data_time: 0.0090 memory: 7124 grad_norm: 5.1029 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.7271 loss: 1.7271 2022/09/07 01:24:23 - mmengine - INFO - Epoch(train) [29][3200/3757] lr: 1.0000e-02 eta: 11:46:49 time: 0.1527 data_time: 0.0091 memory: 7124 grad_norm: 5.2647 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 2.0679 loss: 2.0679 2022/09/07 01:24:40 - mmengine - INFO - Epoch(train) [29][3300/3757] lr: 1.0000e-02 eta: 11:46:35 time: 0.1557 data_time: 0.0095 memory: 7124 grad_norm: 5.2903 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9659 loss: 1.9659 2022/09/07 01:24:56 - mmengine - INFO - Epoch(train) [29][3400/3757] lr: 1.0000e-02 eta: 11:46:19 time: 0.1563 data_time: 0.0100 memory: 7124 grad_norm: 5.3165 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1406 loss: 2.1406 2022/09/07 01:25:12 - mmengine - INFO - Epoch(train) [29][3500/3757] lr: 1.0000e-02 eta: 11:46:03 time: 0.1558 data_time: 0.0093 memory: 7124 grad_norm: 5.0807 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.8724 loss: 1.8724 2022/09/07 01:25:28 - mmengine - INFO - Epoch(train) [29][3600/3757] lr: 1.0000e-02 eta: 11:45:48 time: 0.1541 data_time: 0.0095 memory: 7124 grad_norm: 5.1848 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9253 loss: 1.9253 2022/09/07 01:25:44 - mmengine - INFO - Epoch(train) [29][3700/3757] lr: 1.0000e-02 eta: 11:45:31 time: 0.1560 data_time: 0.0112 memory: 7124 grad_norm: 5.2706 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8620 loss: 1.8620 2022/09/07 01:25:53 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:25:53 - mmengine - INFO - Epoch(train) [29][3757/3757] lr: 1.0000e-02 eta: 11:45:26 time: 0.1340 data_time: 0.0072 memory: 7124 grad_norm: 5.0925 top1_acc: 0.7143 top5_acc: 0.8571 loss_cls: 1.7067 loss: 1.7067 2022/09/07 01:25:53 - mmengine - INFO - Saving checkpoint at 29 epochs 2022/09/07 01:28:10 - mmengine - INFO - Epoch(val) [29][100/310] eta: 0:03:58 time: 1.1346 data_time: 0.8345 memory: 7627 2022/09/07 01:30:29 - mmengine - INFO - Epoch(val) [29][200/310] eta: 0:02:32 time: 1.3850 data_time: 1.0790 memory: 7627 2022/09/07 01:32:33 - mmengine - INFO - Epoch(val) [29][300/310] eta: 0:00:11 time: 1.1230 data_time: 0.8217 memory: 7627 2022/09/07 01:32:51 - mmengine - INFO - Epoch(val) [29][310/310] acc/top1: 0.6471 acc/top5: 0.8657 acc/mean1: 0.6470 2022/09/07 01:32:51 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_26.pth is removed 2022/09/07 01:32:53 - mmengine - INFO - The best checkpoint with 0.6471 acc/top1 at 29 epoch is saved to best_acc/top1_epoch_29.pth. 2022/09/07 01:33:01 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:33:10 - mmengine - INFO - Epoch(train) [30][100/3757] lr: 1.0000e-02 eta: 11:45:03 time: 0.1529 data_time: 0.0107 memory: 7627 grad_norm: 5.2604 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.0786 loss: 2.0786 2022/09/07 01:33:25 - mmengine - INFO - Epoch(train) [30][200/3757] lr: 1.0000e-02 eta: 11:44:47 time: 0.1647 data_time: 0.0153 memory: 7124 grad_norm: 4.9015 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8062 loss: 1.8062 2022/09/07 01:33:41 - mmengine - INFO - Epoch(train) [30][300/3757] lr: 1.0000e-02 eta: 11:44:31 time: 0.1586 data_time: 0.0102 memory: 7124 grad_norm: 5.5577 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.9546 loss: 1.9546 2022/09/07 01:33:57 - mmengine - INFO - Epoch(train) [30][400/3757] lr: 1.0000e-02 eta: 11:44:15 time: 0.1557 data_time: 0.0101 memory: 7124 grad_norm: 5.1400 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8371 loss: 1.8371 2022/09/07 01:34:13 - mmengine - INFO - Epoch(train) [30][500/3757] lr: 1.0000e-02 eta: 11:43:59 time: 0.1570 data_time: 0.0098 memory: 7124 grad_norm: 5.2783 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.6934 loss: 1.6934 2022/09/07 01:34:29 - mmengine - INFO - Epoch(train) [30][600/3757] lr: 1.0000e-02 eta: 11:43:44 time: 0.1579 data_time: 0.0100 memory: 7124 grad_norm: 5.3297 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9088 loss: 1.9088 2022/09/07 01:34:45 - mmengine - INFO - Epoch(train) [30][700/3757] lr: 1.0000e-02 eta: 11:43:27 time: 0.1568 data_time: 0.0102 memory: 7124 grad_norm: 5.3465 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.6974 loss: 1.6974 2022/09/07 01:35:01 - mmengine - INFO - Epoch(train) [30][800/3757] lr: 1.0000e-02 eta: 11:43:11 time: 0.1566 data_time: 0.0112 memory: 7124 grad_norm: 5.6603 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9625 loss: 1.9625 2022/09/07 01:35:17 - mmengine - INFO - Epoch(train) [30][900/3757] lr: 1.0000e-02 eta: 11:42:55 time: 0.1555 data_time: 0.0112 memory: 7124 grad_norm: 5.3535 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7282 loss: 1.7282 2022/09/07 01:35:32 - mmengine - INFO - Epoch(train) [30][1000/3757] lr: 1.0000e-02 eta: 11:42:39 time: 0.1540 data_time: 0.0108 memory: 7124 grad_norm: 5.2985 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.5457 loss: 1.5457 2022/09/07 01:35:40 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:35:48 - mmengine - INFO - Epoch(train) [30][1100/3757] lr: 1.0000e-02 eta: 11:42:23 time: 0.1590 data_time: 0.0105 memory: 7124 grad_norm: 5.2833 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8062 loss: 1.8062 2022/09/07 01:36:04 - mmengine - INFO - Epoch(train) [30][1200/3757] lr: 1.0000e-02 eta: 11:42:07 time: 0.1575 data_time: 0.0099 memory: 7124 grad_norm: 5.3323 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.0593 loss: 2.0593 2022/09/07 01:36:20 - mmengine - INFO - Epoch(train) [30][1300/3757] lr: 1.0000e-02 eta: 11:41:51 time: 0.1593 data_time: 0.0094 memory: 7124 grad_norm: 5.3737 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.0086 loss: 2.0086 2022/09/07 01:36:36 - mmengine - INFO - Epoch(train) [30][1400/3757] lr: 1.0000e-02 eta: 11:41:35 time: 0.1565 data_time: 0.0103 memory: 7124 grad_norm: 5.2327 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 2.0089 loss: 2.0089 2022/09/07 01:36:52 - mmengine - INFO - Epoch(train) [30][1500/3757] lr: 1.0000e-02 eta: 11:41:19 time: 0.1581 data_time: 0.0103 memory: 7124 grad_norm: 5.1470 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.9591 loss: 1.9591 2022/09/07 01:37:08 - mmengine - INFO - Epoch(train) [30][1600/3757] lr: 1.0000e-02 eta: 11:41:04 time: 0.1673 data_time: 0.0113 memory: 7124 grad_norm: 5.1739 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8391 loss: 1.8391 2022/09/07 01:37:23 - mmengine - INFO - Epoch(train) [30][1700/3757] lr: 1.0000e-02 eta: 11:40:47 time: 0.1558 data_time: 0.0101 memory: 7124 grad_norm: 5.2161 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.7630 loss: 1.7630 2022/09/07 01:37:39 - mmengine - INFO - Epoch(train) [30][1800/3757] lr: 1.0000e-02 eta: 11:40:31 time: 0.1564 data_time: 0.0093 memory: 7124 grad_norm: 5.2353 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 1.9375 loss: 1.9375 2022/09/07 01:37:55 - mmengine - INFO - Epoch(train) [30][1900/3757] lr: 1.0000e-02 eta: 11:40:16 time: 0.1573 data_time: 0.0118 memory: 7124 grad_norm: 5.3297 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7789 loss: 1.7789 2022/09/07 01:38:11 - mmengine - INFO - Epoch(train) [30][2000/3757] lr: 1.0000e-02 eta: 11:40:00 time: 0.1555 data_time: 0.0103 memory: 7124 grad_norm: 5.1314 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.0769 loss: 2.0769 2022/09/07 01:38:19 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:38:27 - mmengine - INFO - Epoch(train) [30][2100/3757] lr: 1.0000e-02 eta: 11:39:44 time: 0.1649 data_time: 0.0108 memory: 7124 grad_norm: 5.4470 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.6353 loss: 1.6353 2022/09/07 01:38:43 - mmengine - INFO - Epoch(train) [30][2200/3757] lr: 1.0000e-02 eta: 11:39:28 time: 0.1593 data_time: 0.0092 memory: 7124 grad_norm: 5.4005 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.9177 loss: 1.9177 2022/09/07 01:38:59 - mmengine - INFO - Epoch(train) [30][2300/3757] lr: 1.0000e-02 eta: 11:39:12 time: 0.1561 data_time: 0.0097 memory: 7124 grad_norm: 5.3082 top1_acc: 0.3750 top5_acc: 0.3750 loss_cls: 1.9800 loss: 1.9800 2022/09/07 01:39:15 - mmengine - INFO - Epoch(train) [30][2400/3757] lr: 1.0000e-02 eta: 11:38:56 time: 0.1565 data_time: 0.0097 memory: 7124 grad_norm: 5.1767 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8779 loss: 1.8779 2022/09/07 01:39:31 - mmengine - INFO - Epoch(train) [30][2500/3757] lr: 1.0000e-02 eta: 11:38:40 time: 0.1546 data_time: 0.0093 memory: 7124 grad_norm: 5.4069 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.6783 loss: 1.6783 2022/09/07 01:39:47 - mmengine - INFO - Epoch(train) [30][2600/3757] lr: 1.0000e-02 eta: 11:38:24 time: 0.1612 data_time: 0.0099 memory: 7124 grad_norm: 4.9985 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.9308 loss: 1.9308 2022/09/07 01:40:02 - mmengine - INFO - Epoch(train) [30][2700/3757] lr: 1.0000e-02 eta: 11:38:08 time: 0.1545 data_time: 0.0098 memory: 7124 grad_norm: 5.3875 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8690 loss: 1.8690 2022/09/07 01:40:18 - mmengine - INFO - Epoch(train) [30][2800/3757] lr: 1.0000e-02 eta: 11:37:52 time: 0.1610 data_time: 0.0105 memory: 7124 grad_norm: 5.1177 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6991 loss: 1.6991 2022/09/07 01:40:34 - mmengine - INFO - Epoch(train) [30][2900/3757] lr: 1.0000e-02 eta: 11:37:36 time: 0.1568 data_time: 0.0104 memory: 7124 grad_norm: 5.3421 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.1569 loss: 2.1569 2022/09/07 01:40:50 - mmengine - INFO - Epoch(train) [30][3000/3757] lr: 1.0000e-02 eta: 11:37:20 time: 0.1560 data_time: 0.0126 memory: 7124 grad_norm: 5.2841 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 1.9759 loss: 1.9759 2022/09/07 01:40:57 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:41:06 - mmengine - INFO - Epoch(train) [30][3100/3757] lr: 1.0000e-02 eta: 11:37:04 time: 0.1563 data_time: 0.0091 memory: 7124 grad_norm: 5.3044 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8705 loss: 1.8705 2022/09/07 01:41:21 - mmengine - INFO - Epoch(train) [30][3200/3757] lr: 1.0000e-02 eta: 11:36:47 time: 0.1547 data_time: 0.0097 memory: 7124 grad_norm: 5.3595 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6661 loss: 1.6661 2022/09/07 01:41:37 - mmengine - INFO - Epoch(train) [30][3300/3757] lr: 1.0000e-02 eta: 11:36:32 time: 0.1632 data_time: 0.0110 memory: 7124 grad_norm: 5.1717 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.0472 loss: 2.0472 2022/09/07 01:41:53 - mmengine - INFO - Epoch(train) [30][3400/3757] lr: 1.0000e-02 eta: 11:36:15 time: 0.1574 data_time: 0.0099 memory: 7124 grad_norm: 5.3209 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6661 loss: 1.6661 2022/09/07 01:42:09 - mmengine - INFO - Epoch(train) [30][3500/3757] lr: 1.0000e-02 eta: 11:35:59 time: 0.1587 data_time: 0.0114 memory: 7124 grad_norm: 5.2718 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6175 loss: 1.6175 2022/09/07 01:42:25 - mmengine - INFO - Epoch(train) [30][3600/3757] lr: 1.0000e-02 eta: 11:35:43 time: 0.1542 data_time: 0.0095 memory: 7124 grad_norm: 5.0971 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.8170 loss: 1.8170 2022/09/07 01:42:41 - mmengine - INFO - Epoch(train) [30][3700/3757] lr: 1.0000e-02 eta: 11:35:28 time: 0.1583 data_time: 0.0096 memory: 7124 grad_norm: 5.3224 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 2.1615 loss: 2.1615 2022/09/07 01:42:49 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:42:49 - mmengine - INFO - Epoch(train) [30][3757/3757] lr: 1.0000e-02 eta: 11:35:21 time: 0.1351 data_time: 0.0075 memory: 7124 grad_norm: 5.2921 top1_acc: 0.8571 top5_acc: 0.8571 loss_cls: 1.6110 loss: 1.6110 2022/09/07 01:42:49 - mmengine - INFO - Saving checkpoint at 30 epochs 2022/09/07 01:45:09 - mmengine - INFO - Epoch(val) [30][100/310] eta: 0:04:12 time: 1.2009 data_time: 0.8972 memory: 7627 2022/09/07 01:47:25 - mmengine - INFO - Epoch(val) [30][200/310] eta: 0:02:18 time: 1.2615 data_time: 0.9574 memory: 7627 2022/09/07 01:49:29 - mmengine - INFO - Epoch(val) [30][300/310] eta: 0:00:11 time: 1.1468 data_time: 0.8424 memory: 7627 2022/09/07 01:49:47 - mmengine - INFO - Epoch(val) [30][310/310] acc/top1: 0.6437 acc/top5: 0.8624 acc/mean1: 0.6435 2022/09/07 01:50:05 - mmengine - INFO - Epoch(train) [31][100/3757] lr: 1.0000e-02 eta: 11:35:00 time: 0.1579 data_time: 0.0098 memory: 7627 grad_norm: 5.3894 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6410 loss: 1.6410 2022/09/07 01:50:21 - mmengine - INFO - Epoch(train) [31][200/3757] lr: 1.0000e-02 eta: 11:34:44 time: 0.1586 data_time: 0.0093 memory: 7124 grad_norm: 5.3682 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7126 loss: 1.7126 2022/09/07 01:50:35 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:50:37 - mmengine - INFO - Epoch(train) [31][300/3757] lr: 1.0000e-02 eta: 11:34:28 time: 0.1566 data_time: 0.0098 memory: 7124 grad_norm: 5.1025 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.8840 loss: 1.8840 2022/09/07 01:50:52 - mmengine - INFO - Epoch(train) [31][400/3757] lr: 1.0000e-02 eta: 11:34:11 time: 0.1566 data_time: 0.0100 memory: 7124 grad_norm: 5.1556 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8395 loss: 1.8395 2022/09/07 01:51:08 - mmengine - INFO - Epoch(train) [31][500/3757] lr: 1.0000e-02 eta: 11:33:55 time: 0.1563 data_time: 0.0116 memory: 7124 grad_norm: 5.2694 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8344 loss: 1.8344 2022/09/07 01:51:24 - mmengine - INFO - Epoch(train) [31][600/3757] lr: 1.0000e-02 eta: 11:33:39 time: 0.1569 data_time: 0.0095 memory: 7124 grad_norm: 5.4834 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.7643 loss: 1.7643 2022/09/07 01:51:40 - mmengine - INFO - Epoch(train) [31][700/3757] lr: 1.0000e-02 eta: 11:33:24 time: 0.1662 data_time: 0.0097 memory: 7124 grad_norm: 5.3235 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.9197 loss: 1.9197 2022/09/07 01:51:56 - mmengine - INFO - Epoch(train) [31][800/3757] lr: 1.0000e-02 eta: 11:33:07 time: 0.1585 data_time: 0.0105 memory: 7124 grad_norm: 5.5553 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7768 loss: 1.7768 2022/09/07 01:52:12 - mmengine - INFO - Epoch(train) [31][900/3757] lr: 1.0000e-02 eta: 11:32:51 time: 0.1529 data_time: 0.0100 memory: 7124 grad_norm: 5.2702 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8080 loss: 1.8080 2022/09/07 01:52:27 - mmengine - INFO - Epoch(train) [31][1000/3757] lr: 1.0000e-02 eta: 11:32:35 time: 0.1581 data_time: 0.0121 memory: 7124 grad_norm: 5.4452 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.7986 loss: 1.7986 2022/09/07 01:52:43 - mmengine - INFO - Epoch(train) [31][1100/3757] lr: 1.0000e-02 eta: 11:32:19 time: 0.1564 data_time: 0.0109 memory: 7124 grad_norm: 5.3241 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.7779 loss: 1.7779 2022/09/07 01:52:59 - mmengine - INFO - Epoch(train) [31][1200/3757] lr: 1.0000e-02 eta: 11:32:02 time: 0.1551 data_time: 0.0094 memory: 7124 grad_norm: 5.2270 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8824 loss: 1.8824 2022/09/07 01:53:13 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:53:14 - mmengine - INFO - Epoch(train) [31][1300/3757] lr: 1.0000e-02 eta: 11:31:46 time: 0.1556 data_time: 0.0096 memory: 7124 grad_norm: 5.3538 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.6377 loss: 1.6377 2022/09/07 01:53:31 - mmengine - INFO - Epoch(train) [31][1400/3757] lr: 1.0000e-02 eta: 11:31:31 time: 0.1481 data_time: 0.0077 memory: 7124 grad_norm: 5.5226 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.6873 loss: 1.6873 2022/09/07 01:53:47 - mmengine - INFO - Epoch(train) [31][1500/3757] lr: 1.0000e-02 eta: 11:31:15 time: 0.1551 data_time: 0.0093 memory: 7124 grad_norm: 5.4354 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8224 loss: 1.8224 2022/09/07 01:54:03 - mmengine - INFO - Epoch(train) [31][1600/3757] lr: 1.0000e-02 eta: 11:30:59 time: 0.1581 data_time: 0.0103 memory: 7124 grad_norm: 5.3152 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1229 loss: 2.1229 2022/09/07 01:54:18 - mmengine - INFO - Epoch(train) [31][1700/3757] lr: 1.0000e-02 eta: 11:30:44 time: 0.1575 data_time: 0.0102 memory: 7124 grad_norm: 5.2318 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.6898 loss: 1.6898 2022/09/07 01:54:35 - mmengine - INFO - Epoch(train) [31][1800/3757] lr: 1.0000e-02 eta: 11:30:28 time: 0.1580 data_time: 0.0101 memory: 7124 grad_norm: 5.5654 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9149 loss: 1.9149 2022/09/07 01:54:51 - mmengine - INFO - Epoch(train) [31][1900/3757] lr: 1.0000e-02 eta: 11:30:13 time: 0.1754 data_time: 0.0092 memory: 7124 grad_norm: 5.1892 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.8706 loss: 1.8706 2022/09/07 01:55:06 - mmengine - INFO - Epoch(train) [31][2000/3757] lr: 1.0000e-02 eta: 11:29:56 time: 0.1538 data_time: 0.0121 memory: 7124 grad_norm: 4.9572 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.7048 loss: 1.7048 2022/09/07 01:55:22 - mmengine - INFO - Epoch(train) [31][2100/3757] lr: 1.0000e-02 eta: 11:29:41 time: 0.1661 data_time: 0.0095 memory: 7124 grad_norm: 5.5779 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0568 loss: 2.0568 2022/09/07 01:55:38 - mmengine - INFO - Epoch(train) [31][2200/3757] lr: 1.0000e-02 eta: 11:29:24 time: 0.1621 data_time: 0.0113 memory: 7124 grad_norm: 5.4657 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6453 loss: 1.6453 2022/09/07 01:55:52 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:55:54 - mmengine - INFO - Epoch(train) [31][2300/3757] lr: 1.0000e-02 eta: 11:29:08 time: 0.1562 data_time: 0.0100 memory: 7124 grad_norm: 4.9372 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.7079 loss: 1.7079 2022/09/07 01:56:10 - mmengine - INFO - Epoch(train) [31][2400/3757] lr: 1.0000e-02 eta: 11:28:52 time: 0.1595 data_time: 0.0099 memory: 7124 grad_norm: 5.1901 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.0352 loss: 2.0352 2022/09/07 01:56:26 - mmengine - INFO - Epoch(train) [31][2500/3757] lr: 1.0000e-02 eta: 11:28:37 time: 0.1555 data_time: 0.0100 memory: 7124 grad_norm: 5.4550 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.1408 loss: 2.1408 2022/09/07 01:56:42 - mmengine - INFO - Epoch(train) [31][2600/3757] lr: 1.0000e-02 eta: 11:28:21 time: 0.1550 data_time: 0.0104 memory: 7124 grad_norm: 5.2659 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8906 loss: 1.8906 2022/09/07 01:56:58 - mmengine - INFO - Epoch(train) [31][2700/3757] lr: 1.0000e-02 eta: 11:28:05 time: 0.1612 data_time: 0.0112 memory: 7124 grad_norm: 5.0677 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.7468 loss: 1.7468 2022/09/07 01:57:14 - mmengine - INFO - Epoch(train) [31][2800/3757] lr: 1.0000e-02 eta: 11:27:49 time: 0.1562 data_time: 0.0119 memory: 7124 grad_norm: 5.1946 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.1050 loss: 2.1050 2022/09/07 01:57:29 - mmengine - INFO - Epoch(train) [31][2900/3757] lr: 1.0000e-02 eta: 11:27:33 time: 0.1576 data_time: 0.0093 memory: 7124 grad_norm: 5.1816 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8983 loss: 1.8983 2022/09/07 01:57:45 - mmengine - INFO - Epoch(train) [31][3000/3757] lr: 1.0000e-02 eta: 11:27:17 time: 0.1553 data_time: 0.0089 memory: 7124 grad_norm: 5.4922 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9233 loss: 1.9233 2022/09/07 01:58:01 - mmengine - INFO - Epoch(train) [31][3100/3757] lr: 1.0000e-02 eta: 11:27:01 time: 0.1556 data_time: 0.0114 memory: 7124 grad_norm: 5.0677 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8295 loss: 1.8295 2022/09/07 01:58:17 - mmengine - INFO - Epoch(train) [31][3200/3757] lr: 1.0000e-02 eta: 11:26:45 time: 0.1513 data_time: 0.0109 memory: 7124 grad_norm: 5.3137 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.0193 loss: 2.0193 2022/09/07 01:58:31 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:58:33 - mmengine - INFO - Epoch(train) [31][3300/3757] lr: 1.0000e-02 eta: 11:26:30 time: 0.1578 data_time: 0.0101 memory: 7124 grad_norm: 5.1994 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.9700 loss: 1.9700 2022/09/07 01:58:49 - mmengine - INFO - Epoch(train) [31][3400/3757] lr: 1.0000e-02 eta: 11:26:14 time: 0.1545 data_time: 0.0085 memory: 7124 grad_norm: 5.3364 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.6501 loss: 1.6501 2022/09/07 01:59:05 - mmengine - INFO - Epoch(train) [31][3500/3757] lr: 1.0000e-02 eta: 11:25:57 time: 0.1574 data_time: 0.0101 memory: 7124 grad_norm: 5.3730 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9460 loss: 1.9460 2022/09/07 01:59:20 - mmengine - INFO - Epoch(train) [31][3600/3757] lr: 1.0000e-02 eta: 11:25:41 time: 0.1632 data_time: 0.0104 memory: 7124 grad_norm: 5.4961 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.8517 loss: 1.8517 2022/09/07 01:59:36 - mmengine - INFO - Epoch(train) [31][3700/3757] lr: 1.0000e-02 eta: 11:25:25 time: 0.1584 data_time: 0.0091 memory: 7124 grad_norm: 5.1387 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.8094 loss: 1.8094 2022/09/07 01:59:45 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 01:59:45 - mmengine - INFO - Epoch(train) [31][3757/3757] lr: 1.0000e-02 eta: 11:25:19 time: 0.1361 data_time: 0.0074 memory: 7124 grad_norm: 5.2879 top1_acc: 0.5714 top5_acc: 0.7143 loss_cls: 1.8179 loss: 1.8179 2022/09/07 01:59:45 - mmengine - INFO - Saving checkpoint at 31 epochs 2022/09/07 02:02:05 - mmengine - INFO - Epoch(val) [31][100/310] eta: 0:04:02 time: 1.1557 data_time: 0.8383 memory: 7627 2022/09/07 02:04:20 - mmengine - INFO - Epoch(val) [31][200/310] eta: 0:02:14 time: 1.2267 data_time: 0.9207 memory: 7627 2022/09/07 02:06:26 - mmengine - INFO - Epoch(val) [31][300/310] eta: 0:00:12 time: 1.2925 data_time: 0.9855 memory: 7627 2022/09/07 02:06:46 - mmengine - INFO - Epoch(val) [31][310/310] acc/top1: 0.6481 acc/top5: 0.8630 acc/mean1: 0.6482 2022/09/07 02:06:46 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_29.pth is removed 2022/09/07 02:06:47 - mmengine - INFO - The best checkpoint with 0.6481 acc/top1 at 31 epoch is saved to best_acc/top1_epoch_31.pth. 2022/09/07 02:07:04 - mmengine - INFO - Epoch(train) [32][100/3757] lr: 1.0000e-02 eta: 11:24:56 time: 0.1576 data_time: 0.0112 memory: 7627 grad_norm: 5.4779 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8654 loss: 1.8654 2022/09/07 02:07:20 - mmengine - INFO - Epoch(train) [32][200/3757] lr: 1.0000e-02 eta: 11:24:40 time: 0.1644 data_time: 0.0111 memory: 7124 grad_norm: 5.1966 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.6081 loss: 1.6081 2022/09/07 02:07:36 - mmengine - INFO - Epoch(train) [32][300/3757] lr: 1.0000e-02 eta: 11:24:24 time: 0.1575 data_time: 0.0104 memory: 7124 grad_norm: 5.1046 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8018 loss: 1.8018 2022/09/07 02:07:52 - mmengine - INFO - Epoch(train) [32][400/3757] lr: 1.0000e-02 eta: 11:24:09 time: 0.1582 data_time: 0.0106 memory: 7124 grad_norm: 5.2651 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.7545 loss: 1.7545 2022/09/07 02:08:07 - mmengine - INFO - Epoch(train) [32][500/3757] lr: 1.0000e-02 eta: 11:23:52 time: 0.1572 data_time: 0.0099 memory: 7124 grad_norm: 5.2694 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.9362 loss: 1.9362 2022/09/07 02:08:13 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 02:08:23 - mmengine - INFO - Epoch(train) [32][600/3757] lr: 1.0000e-02 eta: 11:23:36 time: 0.1573 data_time: 0.0092 memory: 7124 grad_norm: 5.5310 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7278 loss: 1.7278 2022/09/07 02:08:39 - mmengine - INFO - Epoch(train) [32][700/3757] lr: 1.0000e-02 eta: 11:23:21 time: 0.1649 data_time: 0.0105 memory: 7124 grad_norm: 5.0367 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7584 loss: 1.7584 2022/09/07 02:08:55 - mmengine - INFO - Epoch(train) [32][800/3757] lr: 1.0000e-02 eta: 11:23:05 time: 0.1589 data_time: 0.0115 memory: 7124 grad_norm: 5.0874 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1774 loss: 2.1774 2022/09/07 02:09:11 - mmengine - INFO - Epoch(train) [32][900/3757] lr: 1.0000e-02 eta: 11:22:49 time: 0.1559 data_time: 0.0116 memory: 7124 grad_norm: 5.3085 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.7956 loss: 1.7956 2022/09/07 02:09:27 - mmengine - INFO - Epoch(train) [32][1000/3757] lr: 1.0000e-02 eta: 11:22:32 time: 0.1585 data_time: 0.0100 memory: 7124 grad_norm: 5.0762 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7386 loss: 1.7386 2022/09/07 02:09:43 - mmengine - INFO - Epoch(train) [32][1100/3757] lr: 1.0000e-02 eta: 11:22:17 time: 0.1619 data_time: 0.0103 memory: 7124 grad_norm: 5.3390 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.0392 loss: 2.0392 2022/09/07 02:09:58 - mmengine - INFO - Epoch(train) [32][1200/3757] lr: 1.0000e-02 eta: 11:22:00 time: 0.1564 data_time: 0.0096 memory: 7124 grad_norm: 5.4045 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.6848 loss: 1.6848 2022/09/07 02:10:14 - mmengine - INFO - Epoch(train) [32][1300/3757] lr: 1.0000e-02 eta: 11:21:45 time: 0.1552 data_time: 0.0098 memory: 7124 grad_norm: 5.2642 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.5898 loss: 1.5898 2022/09/07 02:10:30 - mmengine - INFO - Epoch(train) [32][1400/3757] lr: 1.0000e-02 eta: 11:21:29 time: 0.1566 data_time: 0.0098 memory: 7124 grad_norm: 5.2278 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9435 loss: 1.9435 2022/09/07 02:10:46 - mmengine - INFO - Epoch(train) [32][1500/3757] lr: 1.0000e-02 eta: 11:21:13 time: 0.1539 data_time: 0.0109 memory: 7124 grad_norm: 5.2494 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.6968 loss: 1.6968 2022/09/07 02:10:51 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 02:11:02 - mmengine - INFO - Epoch(train) [32][1600/3757] lr: 1.0000e-02 eta: 11:20:57 time: 0.1732 data_time: 0.0113 memory: 7124 grad_norm: 5.1319 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.1692 loss: 2.1692 2022/09/07 02:11:18 - mmengine - INFO - Epoch(train) [32][1700/3757] lr: 1.0000e-02 eta: 11:20:41 time: 0.1592 data_time: 0.0111 memory: 7124 grad_norm: 5.1609 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.8301 loss: 1.8301 2022/09/07 02:11:34 - mmengine - INFO - Epoch(train) [32][1800/3757] lr: 1.0000e-02 eta: 11:20:25 time: 0.1553 data_time: 0.0110 memory: 7124 grad_norm: 5.2436 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.7713 loss: 1.7713 2022/09/07 02:11:50 - mmengine - INFO - Epoch(train) [32][1900/3757] lr: 1.0000e-02 eta: 11:20:09 time: 0.1588 data_time: 0.0118 memory: 7124 grad_norm: 5.2896 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 1.9437 loss: 1.9437 2022/09/07 02:12:05 - mmengine - INFO - Epoch(train) [32][2000/3757] lr: 1.0000e-02 eta: 11:19:53 time: 0.1549 data_time: 0.0098 memory: 7124 grad_norm: 5.1705 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.8028 loss: 1.8028 2022/09/07 02:12:21 - mmengine - INFO - Epoch(train) [32][2100/3757] lr: 1.0000e-02 eta: 11:19:36 time: 0.1565 data_time: 0.0105 memory: 7124 grad_norm: 5.3424 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.0739 loss: 2.0739 2022/09/07 02:12:37 - mmengine - INFO - Epoch(train) [32][2200/3757] lr: 1.0000e-02 eta: 11:19:20 time: 0.1522 data_time: 0.0100 memory: 7124 grad_norm: 5.3113 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8627 loss: 1.8627 2022/09/07 02:12:53 - mmengine - INFO - Epoch(train) [32][2300/3757] lr: 1.0000e-02 eta: 11:19:05 time: 0.1573 data_time: 0.0107 memory: 7124 grad_norm: 5.1764 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.7966 loss: 1.7966 2022/09/07 02:13:09 - mmengine - INFO - Epoch(train) [32][2400/3757] lr: 1.0000e-02 eta: 11:18:49 time: 0.1547 data_time: 0.0099 memory: 7124 grad_norm: 5.1957 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8714 loss: 1.8714 2022/09/07 02:13:24 - mmengine - INFO - Epoch(train) [32][2500/3757] lr: 1.0000e-02 eta: 11:18:33 time: 0.1551 data_time: 0.0097 memory: 7124 grad_norm: 5.5346 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.7491 loss: 1.7491 2022/09/07 02:13:30 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 02:13:40 - mmengine - INFO - Epoch(train) [32][2600/3757] lr: 1.0000e-02 eta: 11:18:17 time: 0.1598 data_time: 0.0101 memory: 7124 grad_norm: 5.1925 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.9406 loss: 1.9406 2022/09/07 02:13:56 - mmengine - INFO - Epoch(train) [32][2700/3757] lr: 1.0000e-02 eta: 11:18:00 time: 0.1569 data_time: 0.0109 memory: 7124 grad_norm: 5.1890 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.5788 loss: 1.5788 2022/09/07 02:14:12 - mmengine - INFO - Epoch(train) [32][2800/3757] lr: 1.0000e-02 eta: 11:17:44 time: 0.1580 data_time: 0.0105 memory: 7124 grad_norm: 5.3951 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7267 loss: 1.7267 2022/09/07 02:14:28 - mmengine - INFO - Epoch(train) [32][2900/3757] lr: 1.0000e-02 eta: 11:17:28 time: 0.1554 data_time: 0.0108 memory: 7124 grad_norm: 5.2191 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.7503 loss: 1.7503 2022/09/07 02:14:44 - mmengine - INFO - Epoch(train) [32][3000/3757] lr: 1.0000e-02 eta: 11:17:12 time: 0.1543 data_time: 0.0097 memory: 7124 grad_norm: 5.6023 top1_acc: 0.1250 top5_acc: 0.7500 loss_cls: 2.0630 loss: 2.0630 2022/09/07 02:14:59 - mmengine - INFO - Epoch(train) [32][3100/3757] lr: 1.0000e-02 eta: 11:16:56 time: 0.1539 data_time: 0.0099 memory: 7124 grad_norm: 5.3378 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.9185 loss: 1.9185 2022/09/07 02:15:15 - mmengine - INFO - Epoch(train) [32][3200/3757] lr: 1.0000e-02 eta: 11:16:40 time: 0.1549 data_time: 0.0117 memory: 7124 grad_norm: 5.3202 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.6711 loss: 1.6711 2022/09/07 02:15:31 - mmengine - INFO - Epoch(train) [32][3300/3757] lr: 1.0000e-02 eta: 11:16:25 time: 0.1600 data_time: 0.0104 memory: 7124 grad_norm: 5.2608 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7433 loss: 1.7433 2022/09/07 02:15:47 - mmengine - INFO - Epoch(train) [32][3400/3757] lr: 1.0000e-02 eta: 11:16:09 time: 0.1583 data_time: 0.0108 memory: 7124 grad_norm: 5.4639 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8739 loss: 1.8739 2022/09/07 02:16:03 - mmengine - INFO - Epoch(train) [32][3500/3757] lr: 1.0000e-02 eta: 11:15:53 time: 0.1573 data_time: 0.0110 memory: 7124 grad_norm: 5.4143 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9615 loss: 1.9615 2022/09/07 02:16:09 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 02:16:19 - mmengine - INFO - Epoch(train) [32][3600/3757] lr: 1.0000e-02 eta: 11:15:37 time: 0.1587 data_time: 0.0097 memory: 7124 grad_norm: 5.0485 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.5977 loss: 1.5977 2022/09/07 02:16:35 - mmengine - INFO - Epoch(train) [32][3700/3757] lr: 1.0000e-02 eta: 11:15:22 time: 0.1554 data_time: 0.0095 memory: 7124 grad_norm: 5.4903 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8071 loss: 1.8071 2022/09/07 02:16:44 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 02:16:44 - mmengine - INFO - Epoch(train) [32][3757/3757] lr: 1.0000e-02 eta: 11:15:15 time: 0.1373 data_time: 0.0067 memory: 7124 grad_norm: 5.2527 top1_acc: 0.7143 top5_acc: 0.7143 loss_cls: 2.1244 loss: 2.1244 2022/09/07 02:16:44 - mmengine - INFO - Saving checkpoint at 32 epochs 2022/09/07 02:19:02 - mmengine - INFO - Epoch(val) [32][100/310] eta: 0:03:43 time: 1.0646 data_time: 0.7637 memory: 7627 2022/09/07 02:21:22 - mmengine - INFO - Epoch(val) [32][200/310] eta: 0:02:30 time: 1.3714 data_time: 1.0666 memory: 7627 2022/09/07 02:23:26 - mmengine - INFO - Epoch(val) [32][300/310] eta: 0:00:11 time: 1.1106 data_time: 0.8090 memory: 7627 2022/09/07 02:23:42 - mmengine - INFO - Epoch(val) [32][310/310] acc/top1: 0.6472 acc/top5: 0.8625 acc/mean1: 0.6470 2022/09/07 02:24:00 - mmengine - INFO - Epoch(train) [33][100/3757] lr: 1.0000e-02 eta: 11:14:55 time: 0.1564 data_time: 0.0092 memory: 7627 grad_norm: 5.2170 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8631 loss: 1.8631 2022/09/07 02:24:16 - mmengine - INFO - Epoch(train) [33][200/3757] lr: 1.0000e-02 eta: 11:14:39 time: 0.1565 data_time: 0.0100 memory: 7124 grad_norm: 5.2014 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7667 loss: 1.7667 2022/09/07 02:24:32 - mmengine - INFO - Epoch(train) [33][300/3757] lr: 1.0000e-02 eta: 11:14:23 time: 0.1568 data_time: 0.0104 memory: 7124 grad_norm: 5.3209 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7797 loss: 1.7797 2022/09/07 02:24:47 - mmengine - INFO - Epoch(train) [33][400/3757] lr: 1.0000e-02 eta: 11:14:07 time: 0.1554 data_time: 0.0101 memory: 7124 grad_norm: 5.3377 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.5787 loss: 1.5787 2022/09/07 02:25:03 - mmengine - INFO - Epoch(train) [33][500/3757] lr: 1.0000e-02 eta: 11:13:51 time: 0.1639 data_time: 0.0116 memory: 7124 grad_norm: 5.2981 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.7877 loss: 1.7877 2022/09/07 02:25:19 - mmengine - INFO - Epoch(train) [33][600/3757] lr: 1.0000e-02 eta: 11:13:34 time: 0.1573 data_time: 0.0117 memory: 7124 grad_norm: 5.4053 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.7462 loss: 1.7462 2022/09/07 02:25:35 - mmengine - INFO - Epoch(train) [33][700/3757] lr: 1.0000e-02 eta: 11:13:19 time: 0.1553 data_time: 0.0104 memory: 7124 grad_norm: 5.1361 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6294 loss: 1.6294 2022/09/07 02:25:47 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 02:25:51 - mmengine - INFO - Epoch(train) [33][800/3757] lr: 1.0000e-02 eta: 11:13:03 time: 0.1593 data_time: 0.0121 memory: 7124 grad_norm: 5.1495 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9214 loss: 1.9214 2022/09/07 02:26:07 - mmengine - INFO - Epoch(train) [33][900/3757] lr: 1.0000e-02 eta: 11:12:47 time: 0.1570 data_time: 0.0118 memory: 7124 grad_norm: 5.1049 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.9225 loss: 1.9225 2022/09/07 02:26:22 - mmengine - INFO - Epoch(train) [33][1000/3757] lr: 1.0000e-02 eta: 11:12:31 time: 0.1621 data_time: 0.0121 memory: 7124 grad_norm: 5.2194 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.6520 loss: 1.6520 2022/09/07 02:26:38 - mmengine - INFO - Epoch(train) [33][1100/3757] lr: 1.0000e-02 eta: 11:12:15 time: 0.1557 data_time: 0.0115 memory: 7124 grad_norm: 5.2597 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.6660 loss: 1.6660 2022/09/07 02:26:54 - mmengine - INFO - Epoch(train) [33][1200/3757] lr: 1.0000e-02 eta: 11:11:59 time: 0.1604 data_time: 0.0110 memory: 7124 grad_norm: 5.3507 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.7134 loss: 1.7134 2022/09/07 02:27:10 - mmengine - INFO - Epoch(train) [33][1300/3757] lr: 1.0000e-02 eta: 11:11:43 time: 0.1569 data_time: 0.0111 memory: 7124 grad_norm: 5.5816 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8697 loss: 1.8697 2022/09/07 02:27:26 - mmengine - INFO - Epoch(train) [33][1400/3757] lr: 1.0000e-02 eta: 11:11:27 time: 0.1534 data_time: 0.0098 memory: 7124 grad_norm: 5.3267 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.9902 loss: 1.9902 2022/09/07 02:27:42 - mmengine - INFO - Epoch(train) [33][1500/3757] lr: 1.0000e-02 eta: 11:11:11 time: 0.1684 data_time: 0.0147 memory: 7124 grad_norm: 5.1966 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8206 loss: 1.8206 2022/09/07 02:27:57 - mmengine - INFO - Epoch(train) [33][1600/3757] lr: 1.0000e-02 eta: 11:10:55 time: 0.1548 data_time: 0.0118 memory: 7124 grad_norm: 5.2087 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.8542 loss: 1.8542 2022/09/07 02:28:13 - mmengine - INFO - Epoch(train) [33][1700/3757] lr: 1.0000e-02 eta: 11:10:38 time: 0.1566 data_time: 0.0106 memory: 7124 grad_norm: 5.1855 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.6046 loss: 1.6046 2022/09/07 02:28:25 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 02:28:29 - mmengine - INFO - Epoch(train) [33][1800/3757] lr: 1.0000e-02 eta: 11:10:22 time: 0.1554 data_time: 0.0106 memory: 7124 grad_norm: 5.5614 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8339 loss: 1.8339 2022/09/07 02:28:45 - mmengine - INFO - Epoch(train) [33][1900/3757] lr: 1.0000e-02 eta: 11:10:06 time: 0.1558 data_time: 0.0105 memory: 7124 grad_norm: 5.2802 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 2.0898 loss: 2.0898 2022/09/07 02:29:01 - mmengine - INFO - Epoch(train) [33][2000/3757] lr: 1.0000e-02 eta: 11:09:51 time: 0.1672 data_time: 0.0109 memory: 7124 grad_norm: 5.2016 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4664 loss: 1.4664 2022/09/07 02:29:16 - mmengine - INFO - Epoch(train) [33][2100/3757] lr: 1.0000e-02 eta: 11:09:34 time: 0.1528 data_time: 0.0110 memory: 7124 grad_norm: 5.3882 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.9495 loss: 1.9495 2022/09/07 02:29:32 - mmengine - INFO - Epoch(train) [33][2200/3757] lr: 1.0000e-02 eta: 11:09:18 time: 0.1640 data_time: 0.0115 memory: 7124 grad_norm: 5.2148 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.8312 loss: 1.8312 2022/09/07 02:29:48 - mmengine - INFO - Epoch(train) [33][2300/3757] lr: 1.0000e-02 eta: 11:09:02 time: 0.1557 data_time: 0.0119 memory: 7124 grad_norm: 5.4030 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.7950 loss: 1.7950 2022/09/07 02:30:04 - mmengine - INFO - Epoch(train) [33][2400/3757] lr: 1.0000e-02 eta: 11:08:46 time: 0.1580 data_time: 0.0096 memory: 7124 grad_norm: 5.2395 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.4893 loss: 1.4893 2022/09/07 02:30:20 - mmengine - INFO - Epoch(train) [33][2500/3757] lr: 1.0000e-02 eta: 11:08:30 time: 0.1552 data_time: 0.0122 memory: 7124 grad_norm: 5.1795 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9749 loss: 1.9749 2022/09/07 02:30:36 - mmengine - INFO - Epoch(train) [33][2600/3757] lr: 1.0000e-02 eta: 11:08:15 time: 0.1557 data_time: 0.0107 memory: 7124 grad_norm: 5.2163 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.1047 loss: 2.1047 2022/09/07 02:30:52 - mmengine - INFO - Epoch(train) [33][2700/3757] lr: 1.0000e-02 eta: 11:07:59 time: 0.1648 data_time: 0.0117 memory: 7124 grad_norm: 5.1196 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.9022 loss: 1.9022 2022/09/07 02:31:04 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 02:31:08 - mmengine - INFO - Epoch(train) [33][2800/3757] lr: 1.0000e-02 eta: 11:07:43 time: 0.1527 data_time: 0.0111 memory: 7124 grad_norm: 5.3096 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0706 loss: 2.0706 2022/09/07 02:31:23 - mmengine - INFO - Epoch(train) [33][2900/3757] lr: 1.0000e-02 eta: 11:07:27 time: 0.1610 data_time: 0.0101 memory: 7124 grad_norm: 5.2263 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 2.1392 loss: 2.1392 2022/09/07 02:31:39 - mmengine - INFO - Epoch(train) [33][3000/3757] lr: 1.0000e-02 eta: 11:07:11 time: 0.1564 data_time: 0.0099 memory: 7124 grad_norm: 5.3185 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.8074 loss: 1.8074 2022/09/07 02:31:55 - mmengine - INFO - Epoch(train) [33][3100/3757] lr: 1.0000e-02 eta: 11:06:55 time: 0.1551 data_time: 0.0106 memory: 7124 grad_norm: 5.1030 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8980 loss: 1.8980 2022/09/07 02:32:11 - mmengine - INFO - Epoch(train) [33][3200/3757] lr: 1.0000e-02 eta: 11:06:39 time: 0.1574 data_time: 0.0118 memory: 7124 grad_norm: 5.2298 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9353 loss: 1.9353 2022/09/07 02:32:27 - mmengine - INFO - Epoch(train) [33][3300/3757] lr: 1.0000e-02 eta: 11:06:23 time: 0.1579 data_time: 0.0109 memory: 7124 grad_norm: 5.1624 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7375 loss: 1.7375 2022/09/07 02:32:43 - mmengine - INFO - Epoch(train) [33][3400/3757] lr: 1.0000e-02 eta: 11:06:07 time: 0.1617 data_time: 0.0106 memory: 7124 grad_norm: 5.4596 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8501 loss: 1.8501 2022/09/07 02:32:59 - mmengine - INFO - Epoch(train) [33][3500/3757] lr: 1.0000e-02 eta: 11:05:51 time: 0.1568 data_time: 0.0112 memory: 7124 grad_norm: 5.2522 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.0508 loss: 2.0508 2022/09/07 02:33:14 - mmengine - INFO - Epoch(train) [33][3600/3757] lr: 1.0000e-02 eta: 11:05:35 time: 0.1586 data_time: 0.0127 memory: 7124 grad_norm: 5.3616 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.6690 loss: 1.6690 2022/09/07 02:33:30 - mmengine - INFO - Epoch(train) [33][3700/3757] lr: 1.0000e-02 eta: 11:05:19 time: 0.1542 data_time: 0.0098 memory: 7124 grad_norm: 5.2444 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.9833 loss: 1.9833 2022/09/07 02:33:39 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 02:33:39 - mmengine - INFO - Epoch(train) [33][3757/3757] lr: 1.0000e-02 eta: 11:05:12 time: 0.1374 data_time: 0.0086 memory: 7124 grad_norm: 5.1757 top1_acc: 0.4286 top5_acc: 0.5714 loss_cls: 1.9953 loss: 1.9953 2022/09/07 02:33:39 - mmengine - INFO - Saving checkpoint at 33 epochs 2022/09/07 02:35:56 - mmengine - INFO - Epoch(val) [33][100/310] eta: 0:03:44 time: 1.0668 data_time: 0.7651 memory: 7627 2022/09/07 02:38:15 - mmengine - INFO - Epoch(val) [33][200/310] eta: 0:02:30 time: 1.3676 data_time: 1.0659 memory: 7627 2022/09/07 02:40:19 - mmengine - INFO - Epoch(val) [33][300/310] eta: 0:00:11 time: 1.1439 data_time: 0.8459 memory: 7627 2022/09/07 02:40:37 - mmengine - INFO - Epoch(val) [33][310/310] acc/top1: 0.6547 acc/top5: 0.8660 acc/mean1: 0.6546 2022/09/07 02:40:37 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_31.pth is removed 2022/09/07 02:40:40 - mmengine - INFO - The best checkpoint with 0.6547 acc/top1 at 33 epoch is saved to best_acc/top1_epoch_33.pth. 2022/09/07 02:40:44 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 02:40:57 - mmengine - INFO - Epoch(train) [34][100/3757] lr: 1.0000e-02 eta: 11:04:50 time: 0.1583 data_time: 0.0129 memory: 7627 grad_norm: 5.1932 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.5748 loss: 1.5748 2022/09/07 02:41:13 - mmengine - INFO - Epoch(train) [34][200/3757] lr: 1.0000e-02 eta: 11:04:35 time: 0.1694 data_time: 0.0106 memory: 7124 grad_norm: 5.3795 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.6380 loss: 1.6380 2022/09/07 02:41:29 - mmengine - INFO - Epoch(train) [34][300/3757] lr: 1.0000e-02 eta: 11:04:19 time: 0.1599 data_time: 0.0124 memory: 7124 grad_norm: 5.3830 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.4386 loss: 1.4386 2022/09/07 02:41:44 - mmengine - INFO - Epoch(train) [34][400/3757] lr: 1.0000e-02 eta: 11:04:03 time: 0.1596 data_time: 0.0107 memory: 7124 grad_norm: 5.2439 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.6522 loss: 1.6522 2022/09/07 02:42:00 - mmengine - INFO - Epoch(train) [34][500/3757] lr: 1.0000e-02 eta: 11:03:47 time: 0.1567 data_time: 0.0100 memory: 7124 grad_norm: 5.2646 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.9518 loss: 1.9518 2022/09/07 02:42:16 - mmengine - INFO - Epoch(train) [34][600/3757] lr: 1.0000e-02 eta: 11:03:31 time: 0.1544 data_time: 0.0111 memory: 7124 grad_norm: 5.4763 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9345 loss: 1.9345 2022/09/07 02:42:33 - mmengine - INFO - Epoch(train) [34][700/3757] lr: 1.0000e-02 eta: 11:03:16 time: 0.1773 data_time: 0.0113 memory: 7124 grad_norm: 5.0773 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.6966 loss: 1.6966 2022/09/07 02:42:48 - mmengine - INFO - Epoch(train) [34][800/3757] lr: 1.0000e-02 eta: 11:03:00 time: 0.1552 data_time: 0.0105 memory: 7124 grad_norm: 5.3857 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.7481 loss: 1.7481 2022/09/07 02:43:04 - mmengine - INFO - Epoch(train) [34][900/3757] lr: 1.0000e-02 eta: 11:02:44 time: 0.1621 data_time: 0.0120 memory: 7124 grad_norm: 5.3527 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.6174 loss: 1.6174 2022/09/07 02:43:20 - mmengine - INFO - Epoch(train) [34][1000/3757] lr: 1.0000e-02 eta: 11:02:28 time: 0.1594 data_time: 0.0121 memory: 7124 grad_norm: 5.2009 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7794 loss: 1.7794 2022/09/07 02:43:23 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 02:43:36 - mmengine - INFO - Epoch(train) [34][1100/3757] lr: 1.0000e-02 eta: 11:02:13 time: 0.1598 data_time: 0.0109 memory: 7124 grad_norm: 5.3663 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6135 loss: 1.6135 2022/09/07 02:43:52 - mmengine - INFO - Epoch(train) [34][1200/3757] lr: 1.0000e-02 eta: 11:01:57 time: 0.1632 data_time: 0.0125 memory: 7124 grad_norm: 5.3902 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5293 loss: 1.5293 2022/09/07 02:44:08 - mmengine - INFO - Epoch(train) [34][1300/3757] lr: 1.0000e-02 eta: 11:01:41 time: 0.1587 data_time: 0.0131 memory: 7124 grad_norm: 5.2696 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7843 loss: 1.7843 2022/09/07 02:44:24 - mmengine - INFO - Epoch(train) [34][1400/3757] lr: 1.0000e-02 eta: 11:01:25 time: 0.1551 data_time: 0.0101 memory: 7124 grad_norm: 5.3994 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7621 loss: 1.7621 2022/09/07 02:44:40 - mmengine - INFO - Epoch(train) [34][1500/3757] lr: 1.0000e-02 eta: 11:01:09 time: 0.1585 data_time: 0.0094 memory: 7124 grad_norm: 5.2613 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.8303 loss: 1.8303 2022/09/07 02:44:56 - mmengine - INFO - Epoch(train) [34][1600/3757] lr: 1.0000e-02 eta: 11:00:54 time: 0.1562 data_time: 0.0105 memory: 7124 grad_norm: 5.1767 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6780 loss: 1.6780 2022/09/07 02:45:12 - mmengine - INFO - Epoch(train) [34][1700/3757] lr: 1.0000e-02 eta: 11:00:38 time: 0.1549 data_time: 0.0103 memory: 7124 grad_norm: 5.2923 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.0900 loss: 2.0900 2022/09/07 02:45:28 - mmengine - INFO - Epoch(train) [34][1800/3757] lr: 1.0000e-02 eta: 11:00:22 time: 0.1573 data_time: 0.0107 memory: 7124 grad_norm: 5.3085 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9046 loss: 1.9046 2022/09/07 02:45:44 - mmengine - INFO - Epoch(train) [34][1900/3757] lr: 1.0000e-02 eta: 11:00:06 time: 0.1563 data_time: 0.0109 memory: 7124 grad_norm: 5.4348 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8408 loss: 1.8408 2022/09/07 02:46:00 - mmengine - INFO - Epoch(train) [34][2000/3757] lr: 1.0000e-02 eta: 10:59:50 time: 0.1554 data_time: 0.0108 memory: 7124 grad_norm: 5.4771 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8169 loss: 1.8169 2022/09/07 02:46:03 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 02:46:16 - mmengine - INFO - Epoch(train) [34][2100/3757] lr: 1.0000e-02 eta: 10:59:34 time: 0.1566 data_time: 0.0111 memory: 7124 grad_norm: 5.3837 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 2.0261 loss: 2.0261 2022/09/07 02:46:32 - mmengine - INFO - Epoch(train) [34][2200/3757] lr: 1.0000e-02 eta: 10:59:19 time: 0.1556 data_time: 0.0103 memory: 7124 grad_norm: 5.3046 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.7658 loss: 1.7658 2022/09/07 02:46:48 - mmengine - INFO - Epoch(train) [34][2300/3757] lr: 1.0000e-02 eta: 10:59:03 time: 0.1570 data_time: 0.0099 memory: 7124 grad_norm: 5.3510 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8243 loss: 1.8243 2022/09/07 02:47:04 - mmengine - INFO - Epoch(train) [34][2400/3757] lr: 1.0000e-02 eta: 10:58:47 time: 0.1586 data_time: 0.0112 memory: 7124 grad_norm: 5.3647 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8710 loss: 1.8710 2022/09/07 02:47:20 - mmengine - INFO - Epoch(train) [34][2500/3757] lr: 1.0000e-02 eta: 10:58:32 time: 0.1589 data_time: 0.0111 memory: 7124 grad_norm: 5.4718 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.1762 loss: 2.1762 2022/09/07 02:47:35 - mmengine - INFO - Epoch(train) [34][2600/3757] lr: 1.0000e-02 eta: 10:58:16 time: 0.1583 data_time: 0.0111 memory: 7124 grad_norm: 5.2438 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 1.8447 loss: 1.8447 2022/09/07 02:47:51 - mmengine - INFO - Epoch(train) [34][2700/3757] lr: 1.0000e-02 eta: 10:58:00 time: 0.1551 data_time: 0.0115 memory: 7124 grad_norm: 5.5091 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.3243 loss: 2.3243 2022/09/07 02:48:07 - mmengine - INFO - Epoch(train) [34][2800/3757] lr: 1.0000e-02 eta: 10:57:44 time: 0.1578 data_time: 0.0100 memory: 7124 grad_norm: 5.1070 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.4899 loss: 1.4899 2022/09/07 02:48:23 - mmengine - INFO - Epoch(train) [34][2900/3757] lr: 1.0000e-02 eta: 10:57:28 time: 0.1573 data_time: 0.0110 memory: 7124 grad_norm: 5.0577 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 2.1436 loss: 2.1436 2022/09/07 02:48:39 - mmengine - INFO - Epoch(train) [34][3000/3757] lr: 1.0000e-02 eta: 10:57:12 time: 0.1572 data_time: 0.0093 memory: 7124 grad_norm: 5.1513 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7706 loss: 1.7706 2022/09/07 02:48:42 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 02:48:55 - mmengine - INFO - Epoch(train) [34][3100/3757] lr: 1.0000e-02 eta: 10:56:56 time: 0.1664 data_time: 0.0121 memory: 7124 grad_norm: 5.4078 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.0618 loss: 2.0618 2022/09/07 02:49:11 - mmengine - INFO - Epoch(train) [34][3200/3757] lr: 1.0000e-02 eta: 10:56:40 time: 0.1581 data_time: 0.0109 memory: 7124 grad_norm: 5.2781 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 2.1301 loss: 2.1301 2022/09/07 02:49:27 - mmengine - INFO - Epoch(train) [34][3300/3757] lr: 1.0000e-02 eta: 10:56:25 time: 0.1578 data_time: 0.0100 memory: 7124 grad_norm: 5.4107 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.7187 loss: 1.7187 2022/09/07 02:49:43 - mmengine - INFO - Epoch(train) [34][3400/3757] lr: 1.0000e-02 eta: 10:56:09 time: 0.1583 data_time: 0.0116 memory: 7124 grad_norm: 5.2853 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.8659 loss: 1.8659 2022/09/07 02:49:59 - mmengine - INFO - Epoch(train) [34][3500/3757] lr: 1.0000e-02 eta: 10:55:53 time: 0.1570 data_time: 0.0115 memory: 7124 grad_norm: 5.3368 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.8946 loss: 1.8946 2022/09/07 02:50:15 - mmengine - INFO - Epoch(train) [34][3600/3757] lr: 1.0000e-02 eta: 10:55:37 time: 0.1555 data_time: 0.0100 memory: 7124 grad_norm: 5.2877 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.7816 loss: 1.7816 2022/09/07 02:50:31 - mmengine - INFO - Epoch(train) [34][3700/3757] lr: 1.0000e-02 eta: 10:55:21 time: 0.1585 data_time: 0.0115 memory: 7124 grad_norm: 5.2992 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.7049 loss: 1.7049 2022/09/07 02:50:39 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 02:50:39 - mmengine - INFO - Epoch(train) [34][3757/3757] lr: 1.0000e-02 eta: 10:55:15 time: 0.1413 data_time: 0.0066 memory: 7124 grad_norm: 5.5444 top1_acc: 0.5714 top5_acc: 0.8571 loss_cls: 1.8663 loss: 1.8663 2022/09/07 02:50:39 - mmengine - INFO - Saving checkpoint at 34 epochs 2022/09/07 02:52:57 - mmengine - INFO - Epoch(val) [34][100/310] eta: 0:04:02 time: 1.1549 data_time: 0.8533 memory: 7627 2022/09/07 02:55:14 - mmengine - INFO - Epoch(val) [34][200/310] eta: 0:02:22 time: 1.2989 data_time: 0.9913 memory: 7627 2022/09/07 02:57:18 - mmengine - INFO - Epoch(val) [34][300/310] eta: 0:00:11 time: 1.1687 data_time: 0.8710 memory: 7627 2022/09/07 02:57:37 - mmengine - INFO - Epoch(val) [34][310/310] acc/top1: 0.6515 acc/top5: 0.8646 acc/mean1: 0.6514 2022/09/07 02:57:55 - mmengine - INFO - Epoch(train) [35][100/3757] lr: 1.0000e-02 eta: 10:54:55 time: 0.1545 data_time: 0.0099 memory: 7627 grad_norm: 5.3124 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.5834 loss: 1.5834 2022/09/07 02:58:11 - mmengine - INFO - Epoch(train) [35][200/3757] lr: 1.0000e-02 eta: 10:54:39 time: 0.1573 data_time: 0.0119 memory: 7124 grad_norm: 5.4106 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7056 loss: 1.7056 2022/09/07 02:58:21 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 02:58:27 - mmengine - INFO - Epoch(train) [35][300/3757] lr: 1.0000e-02 eta: 10:54:24 time: 0.1654 data_time: 0.0119 memory: 7124 grad_norm: 5.3482 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7179 loss: 1.7179 2022/09/07 02:58:43 - mmengine - INFO - Epoch(train) [35][400/3757] lr: 1.0000e-02 eta: 10:54:08 time: 0.1585 data_time: 0.0099 memory: 7124 grad_norm: 5.2893 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.9514 loss: 1.9514 2022/09/07 02:58:59 - mmengine - INFO - Epoch(train) [35][500/3757] lr: 1.0000e-02 eta: 10:53:52 time: 0.1581 data_time: 0.0106 memory: 7124 grad_norm: 5.2642 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8936 loss: 1.8936 2022/09/07 02:59:15 - mmengine - INFO - Epoch(train) [35][600/3757] lr: 1.0000e-02 eta: 10:53:36 time: 0.1603 data_time: 0.0120 memory: 7124 grad_norm: 5.3116 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.6801 loss: 1.6801 2022/09/07 02:59:31 - mmengine - INFO - Epoch(train) [35][700/3757] lr: 1.0000e-02 eta: 10:53:20 time: 0.1579 data_time: 0.0103 memory: 7124 grad_norm: 5.3418 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.5794 loss: 1.5794 2022/09/07 02:59:46 - mmengine - INFO - Epoch(train) [35][800/3757] lr: 1.0000e-02 eta: 10:53:04 time: 0.1581 data_time: 0.0115 memory: 7124 grad_norm: 5.3722 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9748 loss: 1.9748 2022/09/07 03:00:02 - mmengine - INFO - Epoch(train) [35][900/3757] lr: 1.0000e-02 eta: 10:52:48 time: 0.1584 data_time: 0.0101 memory: 7124 grad_norm: 5.1014 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.7049 loss: 1.7049 2022/09/07 03:00:18 - mmengine - INFO - Epoch(train) [35][1000/3757] lr: 1.0000e-02 eta: 10:52:32 time: 0.1611 data_time: 0.0124 memory: 7124 grad_norm: 5.2421 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.6666 loss: 1.6666 2022/09/07 03:00:34 - mmengine - INFO - Epoch(train) [35][1100/3757] lr: 1.0000e-02 eta: 10:52:17 time: 0.1613 data_time: 0.0128 memory: 7124 grad_norm: 5.1625 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.6592 loss: 1.6592 2022/09/07 03:00:50 - mmengine - INFO - Epoch(train) [35][1200/3757] lr: 1.0000e-02 eta: 10:52:01 time: 0.1576 data_time: 0.0113 memory: 7124 grad_norm: 5.4432 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9934 loss: 1.9934 2022/09/07 03:01:00 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:01:06 - mmengine - INFO - Epoch(train) [35][1300/3757] lr: 1.0000e-02 eta: 10:51:45 time: 0.1619 data_time: 0.0110 memory: 7124 grad_norm: 5.1797 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 2.0165 loss: 2.0165 2022/09/07 03:01:22 - mmengine - INFO - Epoch(train) [35][1400/3757] lr: 1.0000e-02 eta: 10:51:29 time: 0.1559 data_time: 0.0100 memory: 7124 grad_norm: 5.4202 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.5596 loss: 1.5596 2022/09/07 03:01:38 - mmengine - INFO - Epoch(train) [35][1500/3757] lr: 1.0000e-02 eta: 10:51:13 time: 0.1569 data_time: 0.0115 memory: 7124 grad_norm: 5.3584 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.7237 loss: 1.7237 2022/09/07 03:01:54 - mmengine - INFO - Epoch(train) [35][1600/3757] lr: 1.0000e-02 eta: 10:50:57 time: 0.1590 data_time: 0.0105 memory: 7124 grad_norm: 5.2525 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.5698 loss: 1.5698 2022/09/07 03:02:10 - mmengine - INFO - Epoch(train) [35][1700/3757] lr: 1.0000e-02 eta: 10:50:41 time: 0.1556 data_time: 0.0116 memory: 7124 grad_norm: 5.4533 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 2.0363 loss: 2.0363 2022/09/07 03:02:26 - mmengine - INFO - Epoch(train) [35][1800/3757] lr: 1.0000e-02 eta: 10:50:26 time: 0.1614 data_time: 0.0107 memory: 7124 grad_norm: 5.3657 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.9057 loss: 1.9057 2022/09/07 03:02:42 - mmengine - INFO - Epoch(train) [35][1900/3757] lr: 1.0000e-02 eta: 10:50:10 time: 0.1605 data_time: 0.0125 memory: 7124 grad_norm: 5.3021 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9909 loss: 1.9909 2022/09/07 03:02:57 - mmengine - INFO - Epoch(train) [35][2000/3757] lr: 1.0000e-02 eta: 10:49:54 time: 0.1582 data_time: 0.0107 memory: 7124 grad_norm: 5.3610 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7792 loss: 1.7792 2022/09/07 03:03:13 - mmengine - INFO - Epoch(train) [35][2100/3757] lr: 1.0000e-02 eta: 10:49:38 time: 0.1562 data_time: 0.0123 memory: 7124 grad_norm: 5.4775 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 1.8366 loss: 1.8366 2022/09/07 03:03:29 - mmengine - INFO - Epoch(train) [35][2200/3757] lr: 1.0000e-02 eta: 10:49:22 time: 0.1548 data_time: 0.0102 memory: 7124 grad_norm: 5.2357 top1_acc: 0.3750 top5_acc: 1.0000 loss_cls: 1.9114 loss: 1.9114 2022/09/07 03:03:39 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:03:45 - mmengine - INFO - Epoch(train) [35][2300/3757] lr: 1.0000e-02 eta: 10:49:07 time: 0.1592 data_time: 0.0105 memory: 7124 grad_norm: 5.3091 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8815 loss: 1.8815 2022/09/07 03:04:01 - mmengine - INFO - Epoch(train) [35][2400/3757] lr: 1.0000e-02 eta: 10:48:51 time: 0.1575 data_time: 0.0114 memory: 7124 grad_norm: 5.3213 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.6535 loss: 1.6535 2022/09/07 03:04:17 - mmengine - INFO - Epoch(train) [35][2500/3757] lr: 1.0000e-02 eta: 10:48:35 time: 0.1586 data_time: 0.0105 memory: 7124 grad_norm: 5.2303 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.6263 loss: 1.6263 2022/09/07 03:04:33 - mmengine - INFO - Epoch(train) [35][2600/3757] lr: 1.0000e-02 eta: 10:48:19 time: 0.1580 data_time: 0.0104 memory: 7124 grad_norm: 5.3778 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8457 loss: 1.8457 2022/09/07 03:04:49 - mmengine - INFO - Epoch(train) [35][2700/3757] lr: 1.0000e-02 eta: 10:48:04 time: 0.1604 data_time: 0.0101 memory: 7124 grad_norm: 5.3471 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8388 loss: 1.8388 2022/09/07 03:05:05 - mmengine - INFO - Epoch(train) [35][2800/3757] lr: 1.0000e-02 eta: 10:47:48 time: 0.1664 data_time: 0.0132 memory: 7124 grad_norm: 5.3557 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7616 loss: 1.7616 2022/09/07 03:05:21 - mmengine - INFO - Epoch(train) [35][2900/3757] lr: 1.0000e-02 eta: 10:47:32 time: 0.1557 data_time: 0.0101 memory: 7124 grad_norm: 5.2490 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.6978 loss: 1.6978 2022/09/07 03:05:37 - mmengine - INFO - Epoch(train) [35][3000/3757] lr: 1.0000e-02 eta: 10:47:16 time: 0.1596 data_time: 0.0103 memory: 7124 grad_norm: 5.4610 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9742 loss: 1.9742 2022/09/07 03:05:53 - mmengine - INFO - Epoch(train) [35][3100/3757] lr: 1.0000e-02 eta: 10:47:00 time: 0.1565 data_time: 0.0110 memory: 7124 grad_norm: 5.2800 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.9182 loss: 1.9182 2022/09/07 03:06:09 - mmengine - INFO - Epoch(train) [35][3200/3757] lr: 1.0000e-02 eta: 10:46:44 time: 0.1575 data_time: 0.0106 memory: 7124 grad_norm: 5.3143 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.1561 loss: 2.1561 2022/09/07 03:06:18 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:06:24 - mmengine - INFO - Epoch(train) [35][3300/3757] lr: 1.0000e-02 eta: 10:46:28 time: 0.1578 data_time: 0.0119 memory: 7124 grad_norm: 5.6534 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.0613 loss: 2.0613 2022/09/07 03:06:40 - mmengine - INFO - Epoch(train) [35][3400/3757] lr: 1.0000e-02 eta: 10:46:12 time: 0.1591 data_time: 0.0109 memory: 7124 grad_norm: 5.4373 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.5468 loss: 1.5468 2022/09/07 03:06:57 - mmengine - INFO - Epoch(train) [35][3500/3757] lr: 1.0000e-02 eta: 10:45:57 time: 0.1552 data_time: 0.0106 memory: 7124 grad_norm: 5.3642 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.0583 loss: 2.0583 2022/09/07 03:07:12 - mmengine - INFO - Epoch(train) [35][3600/3757] lr: 1.0000e-02 eta: 10:45:41 time: 0.1548 data_time: 0.0102 memory: 7124 grad_norm: 5.4478 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8060 loss: 1.8060 2022/09/07 03:07:28 - mmengine - INFO - Epoch(train) [35][3700/3757] lr: 1.0000e-02 eta: 10:45:25 time: 0.1562 data_time: 0.0109 memory: 7124 grad_norm: 5.1090 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.7819 loss: 1.7819 2022/09/07 03:07:37 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:07:37 - mmengine - INFO - Epoch(train) [35][3757/3757] lr: 1.0000e-02 eta: 10:45:19 time: 0.1394 data_time: 0.0073 memory: 7124 grad_norm: 5.3286 top1_acc: 0.8571 top5_acc: 1.0000 loss_cls: 1.8790 loss: 1.8790 2022/09/07 03:07:37 - mmengine - INFO - Saving checkpoint at 35 epochs 2022/09/07 03:09:57 - mmengine - INFO - Epoch(val) [35][100/310] eta: 0:04:13 time: 1.2058 data_time: 0.8993 memory: 7627 2022/09/07 03:12:11 - mmengine - INFO - Epoch(val) [35][200/310] eta: 0:02:06 time: 1.1464 data_time: 0.8407 memory: 7627 2022/09/07 03:14:18 - mmengine - INFO - Epoch(val) [35][300/310] eta: 0:00:12 time: 1.2417 data_time: 0.9375 memory: 7627 2022/09/07 03:14:36 - mmengine - INFO - Epoch(val) [35][310/310] acc/top1: 0.6474 acc/top5: 0.8653 acc/mean1: 0.6474 2022/09/07 03:14:53 - mmengine - INFO - Epoch(train) [36][100/3757] lr: 1.0000e-02 eta: 10:44:58 time: 0.1589 data_time: 0.0101 memory: 7627 grad_norm: 5.1869 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.8845 loss: 1.8845 2022/09/07 03:15:09 - mmengine - INFO - Epoch(train) [36][200/3757] lr: 1.0000e-02 eta: 10:44:43 time: 0.1560 data_time: 0.0105 memory: 7124 grad_norm: 5.3023 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.7412 loss: 1.7412 2022/09/07 03:15:25 - mmengine - INFO - Epoch(train) [36][300/3757] lr: 1.0000e-02 eta: 10:44:27 time: 0.1575 data_time: 0.0111 memory: 7124 grad_norm: 5.2803 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7352 loss: 1.7352 2022/09/07 03:15:42 - mmengine - INFO - Epoch(train) [36][400/3757] lr: 1.0000e-02 eta: 10:44:12 time: 0.1681 data_time: 0.0100 memory: 7124 grad_norm: 5.1685 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.0787 loss: 2.0787 2022/09/07 03:15:58 - mmengine - INFO - Epoch(train) [36][500/3757] lr: 1.0000e-02 eta: 10:43:56 time: 0.1621 data_time: 0.0119 memory: 7124 grad_norm: 5.3372 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7579 loss: 1.7579 2022/09/07 03:15:58 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:16:13 - mmengine - INFO - Epoch(train) [36][600/3757] lr: 1.0000e-02 eta: 10:43:40 time: 0.1577 data_time: 0.0112 memory: 7124 grad_norm: 5.3995 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6297 loss: 1.6297 2022/09/07 03:16:29 - mmengine - INFO - Epoch(train) [36][700/3757] lr: 1.0000e-02 eta: 10:43:24 time: 0.1566 data_time: 0.0106 memory: 7124 grad_norm: 5.6195 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8889 loss: 1.8889 2022/09/07 03:16:45 - mmengine - INFO - Epoch(train) [36][800/3757] lr: 1.0000e-02 eta: 10:43:08 time: 0.1583 data_time: 0.0114 memory: 7124 grad_norm: 5.5999 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.0404 loss: 2.0404 2022/09/07 03:17:01 - mmengine - INFO - Epoch(train) [36][900/3757] lr: 1.0000e-02 eta: 10:42:52 time: 0.1601 data_time: 0.0106 memory: 7124 grad_norm: 5.4130 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.9105 loss: 1.9105 2022/09/07 03:17:17 - mmengine - INFO - Epoch(train) [36][1000/3757] lr: 1.0000e-02 eta: 10:42:37 time: 0.1624 data_time: 0.0110 memory: 7124 grad_norm: 5.4972 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7276 loss: 1.7276 2022/09/07 03:17:33 - mmengine - INFO - Epoch(train) [36][1100/3757] lr: 1.0000e-02 eta: 10:42:20 time: 0.1586 data_time: 0.0095 memory: 7124 grad_norm: 5.2691 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8618 loss: 1.8618 2022/09/07 03:17:49 - mmengine - INFO - Epoch(train) [36][1200/3757] lr: 1.0000e-02 eta: 10:42:05 time: 0.1587 data_time: 0.0109 memory: 7124 grad_norm: 5.5493 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 2.0341 loss: 2.0341 2022/09/07 03:18:05 - mmengine - INFO - Epoch(train) [36][1300/3757] lr: 1.0000e-02 eta: 10:41:49 time: 0.1581 data_time: 0.0104 memory: 7124 grad_norm: 5.4401 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8387 loss: 1.8387 2022/09/07 03:18:21 - mmengine - INFO - Epoch(train) [36][1400/3757] lr: 1.0000e-02 eta: 10:41:34 time: 0.1582 data_time: 0.0111 memory: 7124 grad_norm: 5.4764 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.5732 loss: 1.5732 2022/09/07 03:18:37 - mmengine - INFO - Epoch(train) [36][1500/3757] lr: 1.0000e-02 eta: 10:41:18 time: 0.1742 data_time: 0.0089 memory: 7124 grad_norm: 5.4751 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.9915 loss: 1.9915 2022/09/07 03:18:38 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:18:53 - mmengine - INFO - Epoch(train) [36][1600/3757] lr: 1.0000e-02 eta: 10:41:02 time: 0.1549 data_time: 0.0098 memory: 7124 grad_norm: 5.4363 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.6152 loss: 1.6152 2022/09/07 03:19:09 - mmengine - INFO - Epoch(train) [36][1700/3757] lr: 1.0000e-02 eta: 10:40:46 time: 0.1568 data_time: 0.0107 memory: 7124 grad_norm: 5.3462 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.7342 loss: 1.7342 2022/09/07 03:19:25 - mmengine - INFO - Epoch(train) [36][1800/3757] lr: 1.0000e-02 eta: 10:40:30 time: 0.1562 data_time: 0.0114 memory: 7124 grad_norm: 5.1756 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.0460 loss: 2.0460 2022/09/07 03:19:41 - mmengine - INFO - Epoch(train) [36][1900/3757] lr: 1.0000e-02 eta: 10:40:14 time: 0.1568 data_time: 0.0106 memory: 7124 grad_norm: 5.1910 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.1398 loss: 2.1398 2022/09/07 03:19:57 - mmengine - INFO - Epoch(train) [36][2000/3757] lr: 1.0000e-02 eta: 10:39:59 time: 0.1622 data_time: 0.0117 memory: 7124 grad_norm: 5.5059 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6515 loss: 1.6515 2022/09/07 03:20:12 - mmengine - INFO - Epoch(train) [36][2100/3757] lr: 1.0000e-02 eta: 10:39:43 time: 0.1560 data_time: 0.0117 memory: 7124 grad_norm: 5.3734 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.6276 loss: 1.6276 2022/09/07 03:20:28 - mmengine - INFO - Epoch(train) [36][2200/3757] lr: 1.0000e-02 eta: 10:39:27 time: 0.1584 data_time: 0.0121 memory: 7124 grad_norm: 5.2250 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.5176 loss: 1.5176 2022/09/07 03:20:44 - mmengine - INFO - Epoch(train) [36][2300/3757] lr: 1.0000e-02 eta: 10:39:11 time: 0.1565 data_time: 0.0111 memory: 7124 grad_norm: 5.5603 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.6500 loss: 1.6500 2022/09/07 03:21:00 - mmengine - INFO - Epoch(train) [36][2400/3757] lr: 1.0000e-02 eta: 10:38:55 time: 0.1550 data_time: 0.0105 memory: 7124 grad_norm: 5.4966 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.0138 loss: 2.0138 2022/09/07 03:21:16 - mmengine - INFO - Epoch(train) [36][2500/3757] lr: 1.0000e-02 eta: 10:38:39 time: 0.1555 data_time: 0.0106 memory: 7124 grad_norm: 5.4032 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9623 loss: 1.9623 2022/09/07 03:21:17 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:21:32 - mmengine - INFO - Epoch(train) [36][2600/3757] lr: 1.0000e-02 eta: 10:38:23 time: 0.1558 data_time: 0.0107 memory: 7124 grad_norm: 5.5511 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.8499 loss: 1.8499 2022/09/07 03:21:48 - mmengine - INFO - Epoch(train) [36][2700/3757] lr: 1.0000e-02 eta: 10:38:07 time: 0.1582 data_time: 0.0114 memory: 7124 grad_norm: 5.2818 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.7103 loss: 1.7103 2022/09/07 03:22:04 - mmengine - INFO - Epoch(train) [36][2800/3757] lr: 1.0000e-02 eta: 10:37:52 time: 0.1580 data_time: 0.0111 memory: 7124 grad_norm: 5.2278 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.6540 loss: 1.6540 2022/09/07 03:22:20 - mmengine - INFO - Epoch(train) [36][2900/3757] lr: 1.0000e-02 eta: 10:37:36 time: 0.1602 data_time: 0.0123 memory: 7124 grad_norm: 5.2476 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.8198 loss: 1.8198 2022/09/07 03:22:36 - mmengine - INFO - Epoch(train) [36][3000/3757] lr: 1.0000e-02 eta: 10:37:20 time: 0.1638 data_time: 0.0104 memory: 7124 grad_norm: 5.5489 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8912 loss: 1.8912 2022/09/07 03:22:52 - mmengine - INFO - Epoch(train) [36][3100/3757] lr: 1.0000e-02 eta: 10:37:04 time: 0.1573 data_time: 0.0106 memory: 7124 grad_norm: 5.3050 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0352 loss: 2.0352 2022/09/07 03:23:08 - mmengine - INFO - Epoch(train) [36][3200/3757] lr: 1.0000e-02 eta: 10:36:49 time: 0.1567 data_time: 0.0113 memory: 7124 grad_norm: 5.0832 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9134 loss: 1.9134 2022/09/07 03:23:24 - mmengine - INFO - Epoch(train) [36][3300/3757] lr: 1.0000e-02 eta: 10:36:33 time: 0.1556 data_time: 0.0103 memory: 7124 grad_norm: 5.4254 top1_acc: 0.2500 top5_acc: 0.8750 loss_cls: 1.8197 loss: 1.8197 2022/09/07 03:23:40 - mmengine - INFO - Epoch(train) [36][3400/3757] lr: 1.0000e-02 eta: 10:36:17 time: 0.1576 data_time: 0.0101 memory: 7124 grad_norm: 5.2613 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.9294 loss: 1.9294 2022/09/07 03:23:55 - mmengine - INFO - Epoch(train) [36][3500/3757] lr: 1.0000e-02 eta: 10:36:01 time: 0.1579 data_time: 0.0124 memory: 7124 grad_norm: 5.3781 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9428 loss: 1.9428 2022/09/07 03:23:57 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:24:11 - mmengine - INFO - Epoch(train) [36][3600/3757] lr: 1.0000e-02 eta: 10:35:45 time: 0.1563 data_time: 0.0106 memory: 7124 grad_norm: 5.3000 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7847 loss: 1.7847 2022/09/07 03:24:28 - mmengine - INFO - Epoch(train) [36][3700/3757] lr: 1.0000e-02 eta: 10:35:30 time: 0.1597 data_time: 0.0110 memory: 7124 grad_norm: 5.5303 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8872 loss: 1.8872 2022/09/07 03:24:36 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:24:36 - mmengine - INFO - Epoch(train) [36][3757/3757] lr: 1.0000e-02 eta: 10:35:23 time: 0.1522 data_time: 0.0068 memory: 7124 grad_norm: 5.3488 top1_acc: 0.5714 top5_acc: 0.7143 loss_cls: 1.9142 loss: 1.9142 2022/09/07 03:24:36 - mmengine - INFO - Saving checkpoint at 36 epochs 2022/09/07 03:26:55 - mmengine - INFO - Epoch(val) [36][100/310] eta: 0:03:59 time: 1.1408 data_time: 0.8308 memory: 7627 2022/09/07 03:29:13 - mmengine - INFO - Epoch(val) [36][200/310] eta: 0:02:30 time: 1.3702 data_time: 1.0637 memory: 7627 2022/09/07 03:31:17 - mmengine - INFO - Epoch(val) [36][300/310] eta: 0:00:11 time: 1.1325 data_time: 0.8252 memory: 7627 2022/09/07 03:31:35 - mmengine - INFO - Epoch(val) [36][310/310] acc/top1: 0.6518 acc/top5: 0.8663 acc/mean1: 0.6516 2022/09/07 03:31:52 - mmengine - INFO - Epoch(train) [37][100/3757] lr: 1.0000e-02 eta: 10:35:03 time: 0.1563 data_time: 0.0101 memory: 7627 grad_norm: 5.2808 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.7722 loss: 1.7722 2022/09/07 03:32:08 - mmengine - INFO - Epoch(train) [37][200/3757] lr: 1.0000e-02 eta: 10:34:47 time: 0.1584 data_time: 0.0124 memory: 7124 grad_norm: 5.3978 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.5565 loss: 1.5565 2022/09/07 03:32:24 - mmengine - INFO - Epoch(train) [37][300/3757] lr: 1.0000e-02 eta: 10:34:31 time: 0.1547 data_time: 0.0105 memory: 7124 grad_norm: 5.0853 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.6735 loss: 1.6735 2022/09/07 03:32:40 - mmengine - INFO - Epoch(train) [37][400/3757] lr: 1.0000e-02 eta: 10:34:16 time: 0.1710 data_time: 0.0116 memory: 7124 grad_norm: 5.2619 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8034 loss: 1.8034 2022/09/07 03:32:56 - mmengine - INFO - Epoch(train) [37][500/3757] lr: 1.0000e-02 eta: 10:34:00 time: 0.1562 data_time: 0.0115 memory: 7124 grad_norm: 5.3441 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.0107 loss: 2.0107 2022/09/07 03:33:12 - mmengine - INFO - Epoch(train) [37][600/3757] lr: 1.0000e-02 eta: 10:33:44 time: 0.1584 data_time: 0.0113 memory: 7124 grad_norm: 5.2783 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7684 loss: 1.7684 2022/09/07 03:33:28 - mmengine - INFO - Epoch(train) [37][700/3757] lr: 1.0000e-02 eta: 10:33:29 time: 0.1579 data_time: 0.0115 memory: 7124 grad_norm: 5.3308 top1_acc: 0.3750 top5_acc: 1.0000 loss_cls: 1.7071 loss: 1.7071 2022/09/07 03:33:36 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:33:44 - mmengine - INFO - Epoch(train) [37][800/3757] lr: 1.0000e-02 eta: 10:33:13 time: 0.1551 data_time: 0.0095 memory: 7124 grad_norm: 5.3611 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 2.0310 loss: 2.0310 2022/09/07 03:34:00 - mmengine - INFO - Epoch(train) [37][900/3757] lr: 1.0000e-02 eta: 10:32:57 time: 0.1607 data_time: 0.0107 memory: 7124 grad_norm: 5.4615 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.5398 loss: 1.5398 2022/09/07 03:34:16 - mmengine - INFO - Epoch(train) [37][1000/3757] lr: 1.0000e-02 eta: 10:32:41 time: 0.1570 data_time: 0.0086 memory: 7124 grad_norm: 5.3076 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7364 loss: 1.7364 2022/09/07 03:34:32 - mmengine - INFO - Epoch(train) [37][1100/3757] lr: 1.0000e-02 eta: 10:32:25 time: 0.1579 data_time: 0.0102 memory: 7124 grad_norm: 5.5952 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.0494 loss: 2.0494 2022/09/07 03:34:48 - mmengine - INFO - Epoch(train) [37][1200/3757] lr: 1.0000e-02 eta: 10:32:10 time: 0.1608 data_time: 0.0113 memory: 7124 grad_norm: 5.3844 top1_acc: 0.3750 top5_acc: 1.0000 loss_cls: 1.8478 loss: 1.8478 2022/09/07 03:35:04 - mmengine - INFO - Epoch(train) [37][1300/3757] lr: 1.0000e-02 eta: 10:31:54 time: 0.1583 data_time: 0.0113 memory: 7124 grad_norm: 5.4067 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9548 loss: 1.9548 2022/09/07 03:35:20 - mmengine - INFO - Epoch(train) [37][1400/3757] lr: 1.0000e-02 eta: 10:31:38 time: 0.1601 data_time: 0.0104 memory: 7124 grad_norm: 5.4950 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 2.1020 loss: 2.1020 2022/09/07 03:35:36 - mmengine - INFO - Epoch(train) [37][1500/3757] lr: 1.0000e-02 eta: 10:31:22 time: 0.1563 data_time: 0.0107 memory: 7124 grad_norm: 5.4834 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8787 loss: 1.8787 2022/09/07 03:35:52 - mmengine - INFO - Epoch(train) [37][1600/3757] lr: 1.0000e-02 eta: 10:31:07 time: 0.1588 data_time: 0.0106 memory: 7124 grad_norm: 5.3226 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.7306 loss: 1.7306 2022/09/07 03:36:08 - mmengine - INFO - Epoch(train) [37][1700/3757] lr: 1.0000e-02 eta: 10:30:51 time: 0.1609 data_time: 0.0105 memory: 7124 grad_norm: 5.4593 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9357 loss: 1.9357 2022/09/07 03:36:15 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:36:24 - mmengine - INFO - Epoch(train) [37][1800/3757] lr: 1.0000e-02 eta: 10:30:35 time: 0.1593 data_time: 0.0108 memory: 7124 grad_norm: 5.5948 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7830 loss: 1.7830 2022/09/07 03:36:40 - mmengine - INFO - Epoch(train) [37][1900/3757] lr: 1.0000e-02 eta: 10:30:19 time: 0.1645 data_time: 0.0104 memory: 7124 grad_norm: 5.4436 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.1898 loss: 2.1898 2022/09/07 03:36:55 - mmengine - INFO - Epoch(train) [37][2000/3757] lr: 1.0000e-02 eta: 10:30:03 time: 0.1579 data_time: 0.0109 memory: 7124 grad_norm: 5.3357 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8815 loss: 1.8815 2022/09/07 03:37:12 - mmengine - INFO - Epoch(train) [37][2100/3757] lr: 1.0000e-02 eta: 10:29:48 time: 0.1588 data_time: 0.0103 memory: 7124 grad_norm: 5.2949 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8009 loss: 1.8009 2022/09/07 03:37:28 - mmengine - INFO - Epoch(train) [37][2200/3757] lr: 1.0000e-02 eta: 10:29:33 time: 0.1830 data_time: 0.0101 memory: 7124 grad_norm: 5.3016 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4181 loss: 1.4181 2022/09/07 03:37:44 - mmengine - INFO - Epoch(train) [37][2300/3757] lr: 1.0000e-02 eta: 10:29:17 time: 0.1596 data_time: 0.0113 memory: 7124 grad_norm: 5.5094 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.0225 loss: 2.0225 2022/09/07 03:38:00 - mmengine - INFO - Epoch(train) [37][2400/3757] lr: 1.0000e-02 eta: 10:29:01 time: 0.1642 data_time: 0.0118 memory: 7124 grad_norm: 5.4775 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.6361 loss: 1.6361 2022/09/07 03:38:16 - mmengine - INFO - Epoch(train) [37][2500/3757] lr: 1.0000e-02 eta: 10:28:45 time: 0.1576 data_time: 0.0104 memory: 7124 grad_norm: 5.2679 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.6416 loss: 1.6416 2022/09/07 03:38:32 - mmengine - INFO - Epoch(train) [37][2600/3757] lr: 1.0000e-02 eta: 10:28:29 time: 0.1557 data_time: 0.0105 memory: 7124 grad_norm: 5.3637 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.6694 loss: 1.6694 2022/09/07 03:38:48 - mmengine - INFO - Epoch(train) [37][2700/3757] lr: 1.0000e-02 eta: 10:28:14 time: 0.1614 data_time: 0.0101 memory: 7124 grad_norm: 5.1599 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5901 loss: 1.5901 2022/09/07 03:38:55 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:39:04 - mmengine - INFO - Epoch(train) [37][2800/3757] lr: 1.0000e-02 eta: 10:27:58 time: 0.1556 data_time: 0.0108 memory: 7124 grad_norm: 5.3723 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.8139 loss: 1.8139 2022/09/07 03:39:19 - mmengine - INFO - Epoch(train) [37][2900/3757] lr: 1.0000e-02 eta: 10:27:42 time: 0.1595 data_time: 0.0115 memory: 7124 grad_norm: 5.2325 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.9576 loss: 1.9576 2022/09/07 03:39:35 - mmengine - INFO - Epoch(train) [37][3000/3757] lr: 1.0000e-02 eta: 10:27:26 time: 0.1571 data_time: 0.0108 memory: 7124 grad_norm: 5.3193 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.9165 loss: 1.9165 2022/09/07 03:39:51 - mmengine - INFO - Epoch(train) [37][3100/3757] lr: 1.0000e-02 eta: 10:27:10 time: 0.1571 data_time: 0.0108 memory: 7124 grad_norm: 5.1821 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8878 loss: 1.8878 2022/09/07 03:40:07 - mmengine - INFO - Epoch(train) [37][3200/3757] lr: 1.0000e-02 eta: 10:26:54 time: 0.1596 data_time: 0.0106 memory: 7124 grad_norm: 5.2574 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5564 loss: 1.5564 2022/09/07 03:40:23 - mmengine - INFO - Epoch(train) [37][3300/3757] lr: 1.0000e-02 eta: 10:26:39 time: 0.1569 data_time: 0.0108 memory: 7124 grad_norm: 5.4233 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8997 loss: 1.8997 2022/09/07 03:40:39 - mmengine - INFO - Epoch(train) [37][3400/3757] lr: 1.0000e-02 eta: 10:26:23 time: 0.1612 data_time: 0.0108 memory: 7124 grad_norm: 5.3383 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9442 loss: 1.9442 2022/09/07 03:40:55 - mmengine - INFO - Epoch(train) [37][3500/3757] lr: 1.0000e-02 eta: 10:26:07 time: 0.1566 data_time: 0.0104 memory: 7124 grad_norm: 5.3231 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8832 loss: 1.8832 2022/09/07 03:41:11 - mmengine - INFO - Epoch(train) [37][3600/3757] lr: 1.0000e-02 eta: 10:25:51 time: 0.1566 data_time: 0.0109 memory: 7124 grad_norm: 5.6079 top1_acc: 0.1250 top5_acc: 0.5000 loss_cls: 2.1853 loss: 2.1853 2022/09/07 03:41:27 - mmengine - INFO - Epoch(train) [37][3700/3757] lr: 1.0000e-02 eta: 10:25:35 time: 0.1549 data_time: 0.0118 memory: 7124 grad_norm: 5.3618 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9611 loss: 1.9611 2022/09/07 03:41:34 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:41:35 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:41:35 - mmengine - INFO - Epoch(train) [37][3757/3757] lr: 1.0000e-02 eta: 10:25:28 time: 0.1372 data_time: 0.0073 memory: 7124 grad_norm: 5.2418 top1_acc: 0.7143 top5_acc: 1.0000 loss_cls: 1.9369 loss: 1.9369 2022/09/07 03:41:35 - mmengine - INFO - Saving checkpoint at 37 epochs 2022/09/07 03:43:52 - mmengine - INFO - Epoch(val) [37][100/310] eta: 0:03:52 time: 1.1084 data_time: 0.8048 memory: 7627 2022/09/07 03:46:11 - mmengine - INFO - Epoch(val) [37][200/310] eta: 0:02:29 time: 1.3561 data_time: 1.0501 memory: 7627 2022/09/07 03:48:15 - mmengine - INFO - Epoch(val) [37][300/310] eta: 0:00:11 time: 1.1243 data_time: 0.8214 memory: 7627 2022/09/07 03:48:33 - mmengine - INFO - Epoch(val) [37][310/310] acc/top1: 0.6596 acc/top5: 0.8705 acc/mean1: 0.6595 2022/09/07 03:48:33 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_33.pth is removed 2022/09/07 03:48:35 - mmengine - INFO - The best checkpoint with 0.6596 acc/top1 at 37 epoch is saved to best_acc/top1_epoch_37.pth. 2022/09/07 03:48:52 - mmengine - INFO - Epoch(train) [38][100/3757] lr: 1.0000e-02 eta: 10:25:07 time: 0.1566 data_time: 0.0106 memory: 7627 grad_norm: 5.2023 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8171 loss: 1.8171 2022/09/07 03:49:08 - mmengine - INFO - Epoch(train) [38][200/3757] lr: 1.0000e-02 eta: 10:24:51 time: 0.1591 data_time: 0.0102 memory: 7124 grad_norm: 5.1219 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.7761 loss: 1.7761 2022/09/07 03:49:24 - mmengine - INFO - Epoch(train) [38][300/3757] lr: 1.0000e-02 eta: 10:24:36 time: 0.1586 data_time: 0.0109 memory: 7124 grad_norm: 5.0911 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.6672 loss: 1.6672 2022/09/07 03:49:40 - mmengine - INFO - Epoch(train) [38][400/3757] lr: 1.0000e-02 eta: 10:24:20 time: 0.1583 data_time: 0.0101 memory: 7124 grad_norm: 5.4185 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.9128 loss: 1.9128 2022/09/07 03:49:56 - mmengine - INFO - Epoch(train) [38][500/3757] lr: 1.0000e-02 eta: 10:24:04 time: 0.1595 data_time: 0.0108 memory: 7124 grad_norm: 5.2485 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7012 loss: 1.7012 2022/09/07 03:50:12 - mmengine - INFO - Epoch(train) [38][600/3757] lr: 1.0000e-02 eta: 10:23:49 time: 0.1590 data_time: 0.0106 memory: 7124 grad_norm: 5.2438 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7517 loss: 1.7517 2022/09/07 03:50:28 - mmengine - INFO - Epoch(train) [38][700/3757] lr: 1.0000e-02 eta: 10:23:33 time: 0.1604 data_time: 0.0110 memory: 7124 grad_norm: 5.3259 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.8034 loss: 1.8034 2022/09/07 03:50:44 - mmengine - INFO - Epoch(train) [38][800/3757] lr: 1.0000e-02 eta: 10:23:17 time: 0.1565 data_time: 0.0114 memory: 7124 grad_norm: 5.3649 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.8848 loss: 1.8848 2022/09/07 03:51:00 - mmengine - INFO - Epoch(train) [38][900/3757] lr: 1.0000e-02 eta: 10:23:01 time: 0.1584 data_time: 0.0111 memory: 7124 grad_norm: 5.4885 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9978 loss: 1.9978 2022/09/07 03:51:15 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:51:16 - mmengine - INFO - Epoch(train) [38][1000/3757] lr: 1.0000e-02 eta: 10:22:45 time: 0.1591 data_time: 0.0117 memory: 7124 grad_norm: 5.3301 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.9979 loss: 1.9979 2022/09/07 03:51:32 - mmengine - INFO - Epoch(train) [38][1100/3757] lr: 1.0000e-02 eta: 10:22:30 time: 0.1547 data_time: 0.0103 memory: 7124 grad_norm: 5.5597 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.5823 loss: 1.5823 2022/09/07 03:51:48 - mmengine - INFO - Epoch(train) [38][1200/3757] lr: 1.0000e-02 eta: 10:22:14 time: 0.1547 data_time: 0.0104 memory: 7124 grad_norm: 5.2575 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0115 loss: 2.0115 2022/09/07 03:52:04 - mmengine - INFO - Epoch(train) [38][1300/3757] lr: 1.0000e-02 eta: 10:21:58 time: 0.1586 data_time: 0.0101 memory: 7124 grad_norm: 5.4399 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8075 loss: 1.8075 2022/09/07 03:52:20 - mmengine - INFO - Epoch(train) [38][1400/3757] lr: 1.0000e-02 eta: 10:21:42 time: 0.1626 data_time: 0.0101 memory: 7124 grad_norm: 5.4534 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7664 loss: 1.7664 2022/09/07 03:52:36 - mmengine - INFO - Epoch(train) [38][1500/3757] lr: 1.0000e-02 eta: 10:21:26 time: 0.1586 data_time: 0.0110 memory: 7124 grad_norm: 5.3433 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.7692 loss: 1.7692 2022/09/07 03:52:52 - mmengine - INFO - Epoch(train) [38][1600/3757] lr: 1.0000e-02 eta: 10:21:11 time: 0.1593 data_time: 0.0099 memory: 7124 grad_norm: 5.4829 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.8446 loss: 1.8446 2022/09/07 03:53:08 - mmengine - INFO - Epoch(train) [38][1700/3757] lr: 1.0000e-02 eta: 10:20:55 time: 0.1584 data_time: 0.0118 memory: 7124 grad_norm: 5.2050 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.6753 loss: 1.6753 2022/09/07 03:53:24 - mmengine - INFO - Epoch(train) [38][1800/3757] lr: 1.0000e-02 eta: 10:20:39 time: 0.1580 data_time: 0.0102 memory: 7124 grad_norm: 5.2863 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9473 loss: 1.9473 2022/09/07 03:53:39 - mmengine - INFO - Epoch(train) [38][1900/3757] lr: 1.0000e-02 eta: 10:20:23 time: 0.1634 data_time: 0.0110 memory: 7124 grad_norm: 5.1116 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8072 loss: 1.8072 2022/09/07 03:53:54 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:53:55 - mmengine - INFO - Epoch(train) [38][2000/3757] lr: 1.0000e-02 eta: 10:20:07 time: 0.1566 data_time: 0.0120 memory: 7124 grad_norm: 5.2236 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7690 loss: 1.7690 2022/09/07 03:54:11 - mmengine - INFO - Epoch(train) [38][2100/3757] lr: 1.0000e-02 eta: 10:19:51 time: 0.1570 data_time: 0.0114 memory: 7124 grad_norm: 5.2056 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.6640 loss: 1.6640 2022/09/07 03:54:27 - mmengine - INFO - Epoch(train) [38][2200/3757] lr: 1.0000e-02 eta: 10:19:35 time: 0.1575 data_time: 0.0113 memory: 7124 grad_norm: 5.4300 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.9765 loss: 1.9765 2022/09/07 03:54:43 - mmengine - INFO - Epoch(train) [38][2300/3757] lr: 1.0000e-02 eta: 10:19:19 time: 0.1571 data_time: 0.0110 memory: 7124 grad_norm: 5.3956 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9430 loss: 1.9430 2022/09/07 03:54:59 - mmengine - INFO - Epoch(train) [38][2400/3757] lr: 1.0000e-02 eta: 10:19:04 time: 0.1636 data_time: 0.0103 memory: 7124 grad_norm: 5.5809 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.8054 loss: 1.8054 2022/09/07 03:55:15 - mmengine - INFO - Epoch(train) [38][2500/3757] lr: 1.0000e-02 eta: 10:18:48 time: 0.1587 data_time: 0.0117 memory: 7124 grad_norm: 5.2499 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.7017 loss: 1.7017 2022/09/07 03:55:31 - mmengine - INFO - Epoch(train) [38][2600/3757] lr: 1.0000e-02 eta: 10:18:32 time: 0.1558 data_time: 0.0111 memory: 7124 grad_norm: 5.3873 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.0293 loss: 2.0293 2022/09/07 03:55:47 - mmengine - INFO - Epoch(train) [38][2700/3757] lr: 1.0000e-02 eta: 10:18:17 time: 0.1607 data_time: 0.0106 memory: 7124 grad_norm: 5.3834 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.9066 loss: 1.9066 2022/09/07 03:56:03 - mmengine - INFO - Epoch(train) [38][2800/3757] lr: 1.0000e-02 eta: 10:18:01 time: 0.1582 data_time: 0.0100 memory: 7124 grad_norm: 5.1183 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8486 loss: 1.8486 2022/09/07 03:56:19 - mmengine - INFO - Epoch(train) [38][2900/3757] lr: 1.0000e-02 eta: 10:17:45 time: 0.1596 data_time: 0.0123 memory: 7124 grad_norm: 5.0866 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.9301 loss: 1.9301 2022/09/07 03:56:33 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:56:35 - mmengine - INFO - Epoch(train) [38][3000/3757] lr: 1.0000e-02 eta: 10:17:29 time: 0.1582 data_time: 0.0100 memory: 7124 grad_norm: 5.4055 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.5841 loss: 1.5841 2022/09/07 03:56:51 - mmengine - INFO - Epoch(train) [38][3100/3757] lr: 1.0000e-02 eta: 10:17:13 time: 0.1627 data_time: 0.0104 memory: 7124 grad_norm: 5.3498 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.9936 loss: 1.9936 2022/09/07 03:57:06 - mmengine - INFO - Epoch(train) [38][3200/3757] lr: 1.0000e-02 eta: 10:16:57 time: 0.1559 data_time: 0.0102 memory: 7124 grad_norm: 5.3395 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.5308 loss: 1.5308 2022/09/07 03:57:22 - mmengine - INFO - Epoch(train) [38][3300/3757] lr: 1.0000e-02 eta: 10:16:41 time: 0.1583 data_time: 0.0104 memory: 7124 grad_norm: 5.3112 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7843 loss: 1.7843 2022/09/07 03:57:38 - mmengine - INFO - Epoch(train) [38][3400/3757] lr: 1.0000e-02 eta: 10:16:25 time: 0.1592 data_time: 0.0108 memory: 7124 grad_norm: 5.3839 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.5799 loss: 1.5799 2022/09/07 03:57:54 - mmengine - INFO - Epoch(train) [38][3500/3757] lr: 1.0000e-02 eta: 10:16:10 time: 0.1619 data_time: 0.0097 memory: 7124 grad_norm: 5.3977 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.8930 loss: 1.8930 2022/09/07 03:58:10 - mmengine - INFO - Epoch(train) [38][3600/3757] lr: 1.0000e-02 eta: 10:15:54 time: 0.1610 data_time: 0.0120 memory: 7124 grad_norm: 5.4375 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.5432 loss: 1.5432 2022/09/07 03:58:26 - mmengine - INFO - Epoch(train) [38][3700/3757] lr: 1.0000e-02 eta: 10:15:38 time: 0.1583 data_time: 0.0111 memory: 7124 grad_norm: 5.3196 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.8180 loss: 1.8180 2022/09/07 03:58:35 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 03:58:35 - mmengine - INFO - Epoch(train) [38][3757/3757] lr: 1.0000e-02 eta: 10:15:32 time: 0.1360 data_time: 0.0075 memory: 7124 grad_norm: 5.3276 top1_acc: 0.4286 top5_acc: 0.7143 loss_cls: 1.7774 loss: 1.7774 2022/09/07 03:58:35 - mmengine - INFO - Saving checkpoint at 38 epochs 2022/09/07 04:00:55 - mmengine - INFO - Epoch(val) [38][100/310] eta: 0:04:24 time: 1.2588 data_time: 0.9576 memory: 7627 2022/09/07 04:03:10 - mmengine - INFO - Epoch(val) [38][200/310] eta: 0:02:12 time: 1.2016 data_time: 0.8924 memory: 7627 2022/09/07 04:05:15 - mmengine - INFO - Epoch(val) [38][300/310] eta: 0:00:11 time: 1.1601 data_time: 0.8614 memory: 7627 2022/09/07 04:05:34 - mmengine - INFO - Epoch(val) [38][310/310] acc/top1: 0.6505 acc/top5: 0.8647 acc/mean1: 0.6504 2022/09/07 04:05:52 - mmengine - INFO - Epoch(train) [39][100/3757] lr: 1.0000e-02 eta: 10:15:12 time: 0.1660 data_time: 0.0131 memory: 7627 grad_norm: 5.3662 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7240 loss: 1.7240 2022/09/07 04:06:08 - mmengine - INFO - Epoch(train) [39][200/3757] lr: 1.0000e-02 eta: 10:14:56 time: 0.1562 data_time: 0.0101 memory: 7124 grad_norm: 5.3664 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9672 loss: 1.9672 2022/09/07 04:06:13 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:06:24 - mmengine - INFO - Epoch(train) [39][300/3757] lr: 1.0000e-02 eta: 10:14:40 time: 0.1564 data_time: 0.0096 memory: 7124 grad_norm: 5.3223 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8269 loss: 1.8269 2022/09/07 04:06:40 - mmengine - INFO - Epoch(train) [39][400/3757] lr: 1.0000e-02 eta: 10:14:25 time: 0.1575 data_time: 0.0112 memory: 7124 grad_norm: 5.4348 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.6221 loss: 1.6221 2022/09/07 04:06:56 - mmengine - INFO - Epoch(train) [39][500/3757] lr: 1.0000e-02 eta: 10:14:10 time: 0.1583 data_time: 0.0104 memory: 7124 grad_norm: 5.3781 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.6670 loss: 1.6670 2022/09/07 04:07:12 - mmengine - INFO - Epoch(train) [39][600/3757] lr: 1.0000e-02 eta: 10:13:54 time: 0.1635 data_time: 0.0107 memory: 7124 grad_norm: 5.5667 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6255 loss: 1.6255 2022/09/07 04:07:28 - mmengine - INFO - Epoch(train) [39][700/3757] lr: 1.0000e-02 eta: 10:13:38 time: 0.1573 data_time: 0.0102 memory: 7124 grad_norm: 5.3594 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.9561 loss: 1.9561 2022/09/07 04:07:44 - mmengine - INFO - Epoch(train) [39][800/3757] lr: 1.0000e-02 eta: 10:13:22 time: 0.1557 data_time: 0.0098 memory: 7124 grad_norm: 5.3932 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.0574 loss: 2.0574 2022/09/07 04:08:00 - mmengine - INFO - Epoch(train) [39][900/3757] lr: 1.0000e-02 eta: 10:13:06 time: 0.1549 data_time: 0.0113 memory: 7124 grad_norm: 5.3293 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9331 loss: 1.9331 2022/09/07 04:08:16 - mmengine - INFO - Epoch(train) [39][1000/3757] lr: 1.0000e-02 eta: 10:12:51 time: 0.1561 data_time: 0.0094 memory: 7124 grad_norm: 5.3637 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.8599 loss: 1.8599 2022/09/07 04:08:32 - mmengine - INFO - Epoch(train) [39][1100/3757] lr: 1.0000e-02 eta: 10:12:35 time: 0.1665 data_time: 0.0157 memory: 7124 grad_norm: 5.2868 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9151 loss: 1.9151 2022/09/07 04:08:48 - mmengine - INFO - Epoch(train) [39][1200/3757] lr: 1.0000e-02 eta: 10:12:19 time: 0.1579 data_time: 0.0100 memory: 7124 grad_norm: 5.4683 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.6372 loss: 1.6372 2022/09/07 04:08:53 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:09:04 - mmengine - INFO - Epoch(train) [39][1300/3757] lr: 1.0000e-02 eta: 10:12:03 time: 0.1569 data_time: 0.0115 memory: 7124 grad_norm: 5.2548 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.6861 loss: 1.6861 2022/09/07 04:09:20 - mmengine - INFO - Epoch(train) [39][1400/3757] lr: 1.0000e-02 eta: 10:11:47 time: 0.1595 data_time: 0.0109 memory: 7124 grad_norm: 5.4939 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.7866 loss: 1.7866 2022/09/07 04:09:35 - mmengine - INFO - Epoch(train) [39][1500/3757] lr: 1.0000e-02 eta: 10:11:31 time: 0.1584 data_time: 0.0102 memory: 7124 grad_norm: 5.3933 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.8633 loss: 1.8633 2022/09/07 04:09:52 - mmengine - INFO - Epoch(train) [39][1600/3757] lr: 1.0000e-02 eta: 10:11:16 time: 0.1696 data_time: 0.0090 memory: 7124 grad_norm: 5.4397 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7501 loss: 1.7501 2022/09/07 04:10:08 - mmengine - INFO - Epoch(train) [39][1700/3757] lr: 1.0000e-02 eta: 10:11:00 time: 0.1551 data_time: 0.0118 memory: 7124 grad_norm: 5.3269 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.4567 loss: 1.4567 2022/09/07 04:10:24 - mmengine - INFO - Epoch(train) [39][1800/3757] lr: 1.0000e-02 eta: 10:10:45 time: 0.1605 data_time: 0.0100 memory: 7124 grad_norm: 5.2215 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.5501 loss: 1.5501 2022/09/07 04:10:40 - mmengine - INFO - Epoch(train) [39][1900/3757] lr: 1.0000e-02 eta: 10:10:29 time: 0.1553 data_time: 0.0100 memory: 7124 grad_norm: 5.2966 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.0599 loss: 2.0599 2022/09/07 04:10:56 - mmengine - INFO - Epoch(train) [39][2000/3757] lr: 1.0000e-02 eta: 10:10:13 time: 0.1559 data_time: 0.0105 memory: 7124 grad_norm: 5.1248 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.7807 loss: 1.7807 2022/09/07 04:11:12 - mmengine - INFO - Epoch(train) [39][2100/3757] lr: 1.0000e-02 eta: 10:09:57 time: 0.1607 data_time: 0.0106 memory: 7124 grad_norm: 5.3111 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0142 loss: 2.0142 2022/09/07 04:11:28 - mmengine - INFO - Epoch(train) [39][2200/3757] lr: 1.0000e-02 eta: 10:09:41 time: 0.1573 data_time: 0.0112 memory: 7124 grad_norm: 5.2568 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6230 loss: 1.6230 2022/09/07 04:11:33 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:11:43 - mmengine - INFO - Epoch(train) [39][2300/3757] lr: 1.0000e-02 eta: 10:09:26 time: 0.1568 data_time: 0.0096 memory: 7124 grad_norm: 5.2082 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7585 loss: 1.7585 2022/09/07 04:11:59 - mmengine - INFO - Epoch(train) [39][2400/3757] lr: 1.0000e-02 eta: 10:09:10 time: 0.1585 data_time: 0.0102 memory: 7124 grad_norm: 5.4787 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.7890 loss: 1.7890 2022/09/07 04:12:15 - mmengine - INFO - Epoch(train) [39][2500/3757] lr: 1.0000e-02 eta: 10:08:54 time: 0.1577 data_time: 0.0106 memory: 7124 grad_norm: 5.4983 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9468 loss: 1.9468 2022/09/07 04:12:31 - mmengine - INFO - Epoch(train) [39][2600/3757] lr: 1.0000e-02 eta: 10:08:38 time: 0.1544 data_time: 0.0093 memory: 7124 grad_norm: 5.2166 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.1473 loss: 2.1473 2022/09/07 04:12:47 - mmengine - INFO - Epoch(train) [39][2700/3757] lr: 1.0000e-02 eta: 10:08:22 time: 0.1590 data_time: 0.0103 memory: 7124 grad_norm: 5.4555 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7503 loss: 1.7503 2022/09/07 04:13:03 - mmengine - INFO - Epoch(train) [39][2800/3757] lr: 1.0000e-02 eta: 10:08:06 time: 0.1586 data_time: 0.0103 memory: 7124 grad_norm: 5.4710 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 1.9640 loss: 1.9640 2022/09/07 04:13:19 - mmengine - INFO - Epoch(train) [39][2900/3757] lr: 1.0000e-02 eta: 10:07:50 time: 0.1578 data_time: 0.0115 memory: 7124 grad_norm: 5.6560 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 1.7220 loss: 1.7220 2022/09/07 04:13:35 - mmengine - INFO - Epoch(train) [39][3000/3757] lr: 1.0000e-02 eta: 10:07:34 time: 0.1589 data_time: 0.0101 memory: 7124 grad_norm: 5.1119 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.6039 loss: 1.6039 2022/09/07 04:13:51 - mmengine - INFO - Epoch(train) [39][3100/3757] lr: 1.0000e-02 eta: 10:07:18 time: 0.1608 data_time: 0.0109 memory: 7124 grad_norm: 5.4062 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.9149 loss: 1.9149 2022/09/07 04:14:06 - mmengine - INFO - Epoch(train) [39][3200/3757] lr: 1.0000e-02 eta: 10:07:02 time: 0.1550 data_time: 0.0125 memory: 7124 grad_norm: 5.2362 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9200 loss: 1.9200 2022/09/07 04:14:12 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:14:22 - mmengine - INFO - Epoch(train) [39][3300/3757] lr: 1.0000e-02 eta: 10:06:46 time: 0.1581 data_time: 0.0108 memory: 7124 grad_norm: 5.2537 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7032 loss: 1.7032 2022/09/07 04:14:38 - mmengine - INFO - Epoch(train) [39][3400/3757] lr: 1.0000e-02 eta: 10:06:30 time: 0.1569 data_time: 0.0100 memory: 7124 grad_norm: 5.3263 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8212 loss: 1.8212 2022/09/07 04:14:54 - mmengine - INFO - Epoch(train) [39][3500/3757] lr: 1.0000e-02 eta: 10:06:15 time: 0.1557 data_time: 0.0097 memory: 7124 grad_norm: 5.6000 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9490 loss: 1.9490 2022/09/07 04:15:10 - mmengine - INFO - Epoch(train) [39][3600/3757] lr: 1.0000e-02 eta: 10:05:59 time: 0.1551 data_time: 0.0102 memory: 7124 grad_norm: 5.5017 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9098 loss: 1.9098 2022/09/07 04:15:26 - mmengine - INFO - Epoch(train) [39][3700/3757] lr: 1.0000e-02 eta: 10:05:43 time: 0.1581 data_time: 0.0095 memory: 7124 grad_norm: 5.5747 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.1315 loss: 2.1315 2022/09/07 04:15:34 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:15:34 - mmengine - INFO - Epoch(train) [39][3757/3757] lr: 1.0000e-02 eta: 10:05:36 time: 0.1367 data_time: 0.0074 memory: 7124 grad_norm: 5.4683 top1_acc: 0.4286 top5_acc: 0.5714 loss_cls: 1.7541 loss: 1.7541 2022/09/07 04:15:34 - mmengine - INFO - Saving checkpoint at 39 epochs 2022/09/07 04:17:53 - mmengine - INFO - Epoch(val) [39][100/310] eta: 0:04:19 time: 1.2376 data_time: 0.9325 memory: 7627 2022/09/07 04:20:08 - mmengine - INFO - Epoch(val) [39][200/310] eta: 0:02:13 time: 1.2116 data_time: 0.9037 memory: 7627 2022/09/07 04:22:14 - mmengine - INFO - Epoch(val) [39][300/310] eta: 0:00:12 time: 1.2885 data_time: 0.9862 memory: 7627 2022/09/07 04:22:32 - mmengine - INFO - Epoch(val) [39][310/310] acc/top1: 0.6566 acc/top5: 0.8684 acc/mean1: 0.6566 2022/09/07 04:22:50 - mmengine - INFO - Epoch(train) [40][100/3757] lr: 1.0000e-02 eta: 10:05:17 time: 0.1600 data_time: 0.0099 memory: 7627 grad_norm: 5.5027 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8562 loss: 1.8562 2022/09/07 04:23:06 - mmengine - INFO - Epoch(train) [40][200/3757] lr: 1.0000e-02 eta: 10:05:01 time: 0.1574 data_time: 0.0090 memory: 7124 grad_norm: 5.4594 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6627 loss: 1.6627 2022/09/07 04:23:22 - mmengine - INFO - Epoch(train) [40][300/3757] lr: 1.0000e-02 eta: 10:04:45 time: 0.1578 data_time: 0.0098 memory: 7124 grad_norm: 5.3838 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6532 loss: 1.6532 2022/09/07 04:23:38 - mmengine - INFO - Epoch(train) [40][400/3757] lr: 1.0000e-02 eta: 10:04:30 time: 0.1598 data_time: 0.0103 memory: 7124 grad_norm: 5.6448 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8436 loss: 1.8436 2022/09/07 04:23:50 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:23:54 - mmengine - INFO - Epoch(train) [40][500/3757] lr: 1.0000e-02 eta: 10:04:14 time: 0.1634 data_time: 0.0102 memory: 7124 grad_norm: 5.3516 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.6987 loss: 1.6987 2022/09/07 04:24:10 - mmengine - INFO - Epoch(train) [40][600/3757] lr: 1.0000e-02 eta: 10:03:58 time: 0.1590 data_time: 0.0103 memory: 7124 grad_norm: 5.3602 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.9621 loss: 1.9621 2022/09/07 04:24:26 - mmengine - INFO - Epoch(train) [40][700/3757] lr: 1.0000e-02 eta: 10:03:42 time: 0.1562 data_time: 0.0114 memory: 7124 grad_norm: 5.4097 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6962 loss: 1.6962 2022/09/07 04:24:42 - mmengine - INFO - Epoch(train) [40][800/3757] lr: 1.0000e-02 eta: 10:03:26 time: 0.1640 data_time: 0.0105 memory: 7124 grad_norm: 5.2790 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8199 loss: 1.8199 2022/09/07 04:24:57 - mmengine - INFO - Epoch(train) [40][900/3757] lr: 1.0000e-02 eta: 10:03:10 time: 0.1556 data_time: 0.0099 memory: 7124 grad_norm: 5.2652 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7682 loss: 1.7682 2022/09/07 04:25:13 - mmengine - INFO - Epoch(train) [40][1000/3757] lr: 1.0000e-02 eta: 10:02:54 time: 0.1596 data_time: 0.0092 memory: 7124 grad_norm: 5.1883 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.5470 loss: 1.5470 2022/09/07 04:25:29 - mmengine - INFO - Epoch(train) [40][1100/3757] lr: 1.0000e-02 eta: 10:02:38 time: 0.1566 data_time: 0.0102 memory: 7124 grad_norm: 5.4007 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.5434 loss: 1.5434 2022/09/07 04:25:45 - mmengine - INFO - Epoch(train) [40][1200/3757] lr: 1.0000e-02 eta: 10:02:23 time: 0.1589 data_time: 0.0097 memory: 7124 grad_norm: 5.5504 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8235 loss: 1.8235 2022/09/07 04:26:01 - mmengine - INFO - Epoch(train) [40][1300/3757] lr: 1.0000e-02 eta: 10:02:07 time: 0.1594 data_time: 0.0110 memory: 7124 grad_norm: 5.5140 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6942 loss: 1.6942 2022/09/07 04:26:17 - mmengine - INFO - Epoch(train) [40][1400/3757] lr: 1.0000e-02 eta: 10:01:51 time: 0.1560 data_time: 0.0101 memory: 7124 grad_norm: 5.3867 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9903 loss: 1.9903 2022/09/07 04:26:29 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:26:33 - mmengine - INFO - Epoch(train) [40][1500/3757] lr: 1.0000e-02 eta: 10:01:35 time: 0.1643 data_time: 0.0110 memory: 7124 grad_norm: 5.5394 top1_acc: 0.3750 top5_acc: 0.3750 loss_cls: 1.9255 loss: 1.9255 2022/09/07 04:26:49 - mmengine - INFO - Epoch(train) [40][1600/3757] lr: 1.0000e-02 eta: 10:01:19 time: 0.1572 data_time: 0.0106 memory: 7124 grad_norm: 5.5413 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.5185 loss: 1.5185 2022/09/07 04:27:05 - mmengine - INFO - Epoch(train) [40][1700/3757] lr: 1.0000e-02 eta: 10:01:03 time: 0.1579 data_time: 0.0115 memory: 7124 grad_norm: 5.2785 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6491 loss: 1.6491 2022/09/07 04:27:21 - mmengine - INFO - Epoch(train) [40][1800/3757] lr: 1.0000e-02 eta: 10:00:48 time: 0.1606 data_time: 0.0096 memory: 7124 grad_norm: 5.5366 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.3688 loss: 1.3688 2022/09/07 04:27:37 - mmengine - INFO - Epoch(train) [40][1900/3757] lr: 1.0000e-02 eta: 10:00:32 time: 0.1537 data_time: 0.0090 memory: 7124 grad_norm: 5.5718 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.0158 loss: 2.0158 2022/09/07 04:27:53 - mmengine - INFO - Epoch(train) [40][2000/3757] lr: 1.0000e-02 eta: 10:00:16 time: 0.1652 data_time: 0.0093 memory: 7124 grad_norm: 5.2512 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.7332 loss: 1.7332 2022/09/07 04:28:09 - mmengine - INFO - Epoch(train) [40][2100/3757] lr: 1.0000e-02 eta: 10:00:00 time: 0.1561 data_time: 0.0107 memory: 7124 grad_norm: 5.4874 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.7314 loss: 1.7314 2022/09/07 04:28:24 - mmengine - INFO - Epoch(train) [40][2200/3757] lr: 1.0000e-02 eta: 9:59:44 time: 0.1570 data_time: 0.0100 memory: 7124 grad_norm: 5.4979 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7249 loss: 1.7249 2022/09/07 04:28:40 - mmengine - INFO - Epoch(train) [40][2300/3757] lr: 1.0000e-02 eta: 9:59:28 time: 0.1560 data_time: 0.0088 memory: 7124 grad_norm: 5.2386 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7968 loss: 1.7968 2022/09/07 04:28:56 - mmengine - INFO - Epoch(train) [40][2400/3757] lr: 1.0000e-02 eta: 9:59:12 time: 0.1568 data_time: 0.0107 memory: 7124 grad_norm: 5.1690 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.7793 loss: 1.7793 2022/09/07 04:29:08 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:29:12 - mmengine - INFO - Epoch(train) [40][2500/3757] lr: 1.0000e-02 eta: 9:58:57 time: 0.1628 data_time: 0.0101 memory: 7124 grad_norm: 5.1519 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.4312 loss: 1.4312 2022/09/07 04:29:28 - mmengine - INFO - Epoch(train) [40][2600/3757] lr: 1.0000e-02 eta: 9:58:41 time: 0.1559 data_time: 0.0100 memory: 7124 grad_norm: 5.4430 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8391 loss: 1.8391 2022/09/07 04:29:44 - mmengine - INFO - Epoch(train) [40][2700/3757] lr: 1.0000e-02 eta: 9:58:25 time: 0.1576 data_time: 0.0096 memory: 7124 grad_norm: 5.3362 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 2.0828 loss: 2.0828 2022/09/07 04:30:00 - mmengine - INFO - Epoch(train) [40][2800/3757] lr: 1.0000e-02 eta: 9:58:09 time: 0.1553 data_time: 0.0100 memory: 7124 grad_norm: 5.1863 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5408 loss: 1.5408 2022/09/07 04:30:16 - mmengine - INFO - Epoch(train) [40][2900/3757] lr: 1.0000e-02 eta: 9:57:53 time: 0.1564 data_time: 0.0119 memory: 7124 grad_norm: 5.2810 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.6344 loss: 1.6344 2022/09/07 04:30:32 - mmengine - INFO - Epoch(train) [40][3000/3757] lr: 1.0000e-02 eta: 9:57:37 time: 0.1560 data_time: 0.0104 memory: 7124 grad_norm: 5.2585 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7992 loss: 1.7992 2022/09/07 04:30:48 - mmengine - INFO - Epoch(train) [40][3100/3757] lr: 1.0000e-02 eta: 9:57:21 time: 0.1572 data_time: 0.0102 memory: 7124 grad_norm: 5.4310 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.6944 loss: 1.6944 2022/09/07 04:31:04 - mmengine - INFO - Epoch(train) [40][3200/3757] lr: 1.0000e-02 eta: 9:57:06 time: 0.1588 data_time: 0.0114 memory: 7124 grad_norm: 5.4158 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.0632 loss: 2.0632 2022/09/07 04:31:19 - mmengine - INFO - Epoch(train) [40][3300/3757] lr: 1.0000e-02 eta: 9:56:50 time: 0.1564 data_time: 0.0085 memory: 7124 grad_norm: 5.3394 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.7597 loss: 1.7597 2022/09/07 04:31:35 - mmengine - INFO - Epoch(train) [40][3400/3757] lr: 1.0000e-02 eta: 9:56:34 time: 0.1668 data_time: 0.0104 memory: 7124 grad_norm: 5.4370 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.7902 loss: 1.7902 2022/09/07 04:31:48 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:31:51 - mmengine - INFO - Epoch(train) [40][3500/3757] lr: 1.0000e-02 eta: 9:56:18 time: 0.1581 data_time: 0.0095 memory: 7124 grad_norm: 5.5773 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.4936 loss: 1.4936 2022/09/07 04:32:07 - mmengine - INFO - Epoch(train) [40][3600/3757] lr: 1.0000e-02 eta: 9:56:02 time: 0.1580 data_time: 0.0090 memory: 7124 grad_norm: 5.2612 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.8349 loss: 1.8349 2022/09/07 04:32:23 - mmengine - INFO - Epoch(train) [40][3700/3757] lr: 1.0000e-02 eta: 9:55:46 time: 0.1563 data_time: 0.0109 memory: 7124 grad_norm: 5.3719 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.7401 loss: 1.7401 2022/09/07 04:32:32 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:32:32 - mmengine - INFO - Epoch(train) [40][3757/3757] lr: 1.0000e-02 eta: 9:55:40 time: 0.1377 data_time: 0.0075 memory: 7124 grad_norm: 5.4390 top1_acc: 0.5714 top5_acc: 0.8571 loss_cls: 1.5787 loss: 1.5787 2022/09/07 04:32:32 - mmengine - INFO - Saving checkpoint at 40 epochs 2022/09/07 04:34:50 - mmengine - INFO - Epoch(val) [40][100/310] eta: 0:03:50 time: 1.0973 data_time: 0.7923 memory: 7627 2022/09/07 04:37:07 - mmengine - INFO - Epoch(val) [40][200/310] eta: 0:02:19 time: 1.2699 data_time: 0.9695 memory: 7627 2022/09/07 04:39:10 - mmengine - INFO - Epoch(val) [40][300/310] eta: 0:00:11 time: 1.1260 data_time: 0.8267 memory: 7627 2022/09/07 04:39:28 - mmengine - INFO - Epoch(val) [40][310/310] acc/top1: 0.6580 acc/top5: 0.8693 acc/mean1: 0.6580 2022/09/07 04:39:45 - mmengine - INFO - Epoch(train) [41][100/3757] lr: 1.0000e-02 eta: 9:55:19 time: 0.1609 data_time: 0.0098 memory: 7627 grad_norm: 5.4747 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8382 loss: 1.8382 2022/09/07 04:40:01 - mmengine - INFO - Epoch(train) [41][200/3757] lr: 1.0000e-02 eta: 9:55:03 time: 0.1573 data_time: 0.0099 memory: 7124 grad_norm: 5.3354 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0176 loss: 2.0176 2022/09/07 04:40:17 - mmengine - INFO - Epoch(train) [41][300/3757] lr: 1.0000e-02 eta: 9:54:48 time: 0.1548 data_time: 0.0104 memory: 7124 grad_norm: 5.3239 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7199 loss: 1.7199 2022/09/07 04:40:33 - mmengine - INFO - Epoch(train) [41][400/3757] lr: 1.0000e-02 eta: 9:54:32 time: 0.1584 data_time: 0.0098 memory: 7124 grad_norm: 5.5345 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.6257 loss: 1.6257 2022/09/07 04:40:49 - mmengine - INFO - Epoch(train) [41][500/3757] lr: 1.0000e-02 eta: 9:54:16 time: 0.1584 data_time: 0.0107 memory: 7124 grad_norm: 5.4113 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.6470 loss: 1.6470 2022/09/07 04:41:04 - mmengine - INFO - Epoch(train) [41][600/3757] lr: 1.0000e-02 eta: 9:54:00 time: 0.1627 data_time: 0.0101 memory: 7124 grad_norm: 5.3969 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.7949 loss: 1.7949 2022/09/07 04:41:20 - mmengine - INFO - Epoch(train) [41][700/3757] lr: 1.0000e-02 eta: 9:53:44 time: 0.1561 data_time: 0.0110 memory: 7124 grad_norm: 5.3981 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.8789 loss: 1.8789 2022/09/07 04:41:23 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:41:36 - mmengine - INFO - Epoch(train) [41][800/3757] lr: 1.0000e-02 eta: 9:53:28 time: 0.1648 data_time: 0.0089 memory: 7124 grad_norm: 5.2212 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.6077 loss: 1.6077 2022/09/07 04:41:52 - mmengine - INFO - Epoch(train) [41][900/3757] lr: 1.0000e-02 eta: 9:53:12 time: 0.1551 data_time: 0.0100 memory: 7124 grad_norm: 5.0247 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8153 loss: 1.8153 2022/09/07 04:42:08 - mmengine - INFO - Epoch(train) [41][1000/3757] lr: 1.0000e-02 eta: 9:52:56 time: 0.1555 data_time: 0.0100 memory: 7124 grad_norm: 5.4081 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8006 loss: 1.8006 2022/09/07 04:42:24 - mmengine - INFO - Epoch(train) [41][1100/3757] lr: 1.0000e-02 eta: 9:52:40 time: 0.1578 data_time: 0.0101 memory: 7124 grad_norm: 5.3434 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0103 loss: 2.0103 2022/09/07 04:42:40 - mmengine - INFO - Epoch(train) [41][1200/3757] lr: 1.0000e-02 eta: 9:52:24 time: 0.1588 data_time: 0.0098 memory: 7124 grad_norm: 5.3113 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.6137 loss: 1.6137 2022/09/07 04:42:56 - mmengine - INFO - Epoch(train) [41][1300/3757] lr: 1.0000e-02 eta: 9:52:09 time: 0.1621 data_time: 0.0113 memory: 7124 grad_norm: 5.3596 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8194 loss: 1.8194 2022/09/07 04:43:11 - mmengine - INFO - Epoch(train) [41][1400/3757] lr: 1.0000e-02 eta: 9:51:53 time: 0.1585 data_time: 0.0094 memory: 7124 grad_norm: 5.4535 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.7226 loss: 1.7226 2022/09/07 04:43:27 - mmengine - INFO - Epoch(train) [41][1500/3757] lr: 1.0000e-02 eta: 9:51:37 time: 0.1561 data_time: 0.0099 memory: 7124 grad_norm: 5.5942 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5676 loss: 1.5676 2022/09/07 04:43:43 - mmengine - INFO - Epoch(train) [41][1600/3757] lr: 1.0000e-02 eta: 9:51:21 time: 0.1567 data_time: 0.0116 memory: 7124 grad_norm: 5.5631 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.4333 loss: 1.4333 2022/09/07 04:43:59 - mmengine - INFO - Epoch(train) [41][1700/3757] lr: 1.0000e-02 eta: 9:51:05 time: 0.1591 data_time: 0.0115 memory: 7124 grad_norm: 5.3038 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.5398 loss: 1.5398 2022/09/07 04:44:02 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:44:15 - mmengine - INFO - Epoch(train) [41][1800/3757] lr: 1.0000e-02 eta: 9:50:49 time: 0.1576 data_time: 0.0107 memory: 7124 grad_norm: 5.3407 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6234 loss: 1.6234 2022/09/07 04:44:31 - mmengine - INFO - Epoch(train) [41][1900/3757] lr: 1.0000e-02 eta: 9:50:33 time: 0.1564 data_time: 0.0099 memory: 7124 grad_norm: 5.2174 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.9430 loss: 1.9430 2022/09/07 04:44:47 - mmengine - INFO - Epoch(train) [41][2000/3757] lr: 1.0000e-02 eta: 9:50:17 time: 0.1586 data_time: 0.0106 memory: 7124 grad_norm: 5.3077 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9477 loss: 1.9477 2022/09/07 04:45:02 - mmengine - INFO - Epoch(train) [41][2100/3757] lr: 1.0000e-02 eta: 9:50:01 time: 0.1582 data_time: 0.0106 memory: 7124 grad_norm: 5.4506 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.8893 loss: 1.8893 2022/09/07 04:45:18 - mmengine - INFO - Epoch(train) [41][2200/3757] lr: 1.0000e-02 eta: 9:49:45 time: 0.1610 data_time: 0.0104 memory: 7124 grad_norm: 5.2954 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.8678 loss: 1.8678 2022/09/07 04:45:34 - mmengine - INFO - Epoch(train) [41][2300/3757] lr: 1.0000e-02 eta: 9:49:30 time: 0.1613 data_time: 0.0107 memory: 7124 grad_norm: 5.4771 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.6632 loss: 1.6632 2022/09/07 04:45:50 - mmengine - INFO - Epoch(train) [41][2400/3757] lr: 1.0000e-02 eta: 9:49:13 time: 0.1572 data_time: 0.0102 memory: 7124 grad_norm: 5.5156 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6575 loss: 1.6575 2022/09/07 04:46:06 - mmengine - INFO - Epoch(train) [41][2500/3757] lr: 1.0000e-02 eta: 9:48:58 time: 0.1613 data_time: 0.0115 memory: 7124 grad_norm: 5.4278 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.9054 loss: 1.9054 2022/09/07 04:46:22 - mmengine - INFO - Epoch(train) [41][2600/3757] lr: 1.0000e-02 eta: 9:48:42 time: 0.1588 data_time: 0.0098 memory: 7124 grad_norm: 5.5659 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.7901 loss: 1.7901 2022/09/07 04:46:38 - mmengine - INFO - Epoch(train) [41][2700/3757] lr: 1.0000e-02 eta: 9:48:26 time: 0.1565 data_time: 0.0112 memory: 7124 grad_norm: 5.5569 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.8905 loss: 1.8905 2022/09/07 04:46:41 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:46:54 - mmengine - INFO - Epoch(train) [41][2800/3757] lr: 1.0000e-02 eta: 9:48:10 time: 0.1624 data_time: 0.0138 memory: 7124 grad_norm: 5.0453 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8976 loss: 1.8976 2022/09/07 04:47:10 - mmengine - INFO - Epoch(train) [41][2900/3757] lr: 1.0000e-02 eta: 9:47:54 time: 0.1574 data_time: 0.0093 memory: 7124 grad_norm: 5.4654 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9578 loss: 1.9578 2022/09/07 04:47:26 - mmengine - INFO - Epoch(train) [41][3000/3757] lr: 1.0000e-02 eta: 9:47:38 time: 0.1583 data_time: 0.0116 memory: 7124 grad_norm: 5.2336 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.6399 loss: 1.6399 2022/09/07 04:47:41 - mmengine - INFO - Epoch(train) [41][3100/3757] lr: 1.0000e-02 eta: 9:47:22 time: 0.1573 data_time: 0.0116 memory: 7124 grad_norm: 5.1671 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.6683 loss: 1.6683 2022/09/07 04:47:57 - mmengine - INFO - Epoch(train) [41][3200/3757] lr: 1.0000e-02 eta: 9:47:07 time: 0.1594 data_time: 0.0104 memory: 7124 grad_norm: 5.4479 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.7882 loss: 1.7882 2022/09/07 04:48:13 - mmengine - INFO - Epoch(train) [41][3300/3757] lr: 1.0000e-02 eta: 9:46:51 time: 0.1576 data_time: 0.0109 memory: 7124 grad_norm: 5.3097 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7616 loss: 1.7616 2022/09/07 04:48:29 - mmengine - INFO - Epoch(train) [41][3400/3757] lr: 1.0000e-02 eta: 9:46:35 time: 0.1578 data_time: 0.0099 memory: 7124 grad_norm: 5.4078 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.9070 loss: 1.9070 2022/09/07 04:48:45 - mmengine - INFO - Epoch(train) [41][3500/3757] lr: 1.0000e-02 eta: 9:46:19 time: 0.1588 data_time: 0.0094 memory: 7124 grad_norm: 5.5765 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9502 loss: 1.9502 2022/09/07 04:49:01 - mmengine - INFO - Epoch(train) [41][3600/3757] lr: 1.0000e-02 eta: 9:46:03 time: 0.1602 data_time: 0.0099 memory: 7124 grad_norm: 5.2793 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.7132 loss: 1.7132 2022/09/07 04:49:17 - mmengine - INFO - Epoch(train) [41][3700/3757] lr: 1.0000e-02 eta: 9:45:47 time: 0.1618 data_time: 0.0100 memory: 7124 grad_norm: 5.1467 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.9444 loss: 1.9444 2022/09/07 04:49:20 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:49:26 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:49:26 - mmengine - INFO - Epoch(train) [41][3757/3757] lr: 1.0000e-02 eta: 9:45:41 time: 0.1350 data_time: 0.0074 memory: 7124 grad_norm: 5.2675 top1_acc: 0.2857 top5_acc: 0.8571 loss_cls: 1.8198 loss: 1.8198 2022/09/07 04:49:26 - mmengine - INFO - Saving checkpoint at 41 epochs 2022/09/07 04:51:43 - mmengine - INFO - Epoch(val) [41][100/310] eta: 0:03:38 time: 1.0400 data_time: 0.7374 memory: 7627 2022/09/07 04:54:04 - mmengine - INFO - Epoch(val) [41][200/310] eta: 0:02:29 time: 1.3551 data_time: 1.0521 memory: 7627 2022/09/07 04:56:08 - mmengine - INFO - Epoch(val) [41][300/310] eta: 0:00:10 time: 1.0814 data_time: 0.7862 memory: 7627 2022/09/07 04:56:22 - mmengine - INFO - Epoch(val) [41][310/310] acc/top1: 0.6547 acc/top5: 0.8655 acc/mean1: 0.6547 2022/09/07 04:56:40 - mmengine - INFO - Epoch(train) [42][100/3757] lr: 1.0000e-02 eta: 9:45:21 time: 0.1571 data_time: 0.0110 memory: 7627 grad_norm: 5.3550 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.6505 loss: 1.6505 2022/09/07 04:56:56 - mmengine - INFO - Epoch(train) [42][200/3757] lr: 1.0000e-02 eta: 9:45:05 time: 0.1625 data_time: 0.0113 memory: 7124 grad_norm: 5.5367 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7234 loss: 1.7234 2022/09/07 04:57:12 - mmengine - INFO - Epoch(train) [42][300/3757] lr: 1.0000e-02 eta: 9:44:49 time: 0.1565 data_time: 0.0093 memory: 7124 grad_norm: 5.4623 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.5856 loss: 1.5856 2022/09/07 04:57:27 - mmengine - INFO - Epoch(train) [42][400/3757] lr: 1.0000e-02 eta: 9:44:33 time: 0.1548 data_time: 0.0095 memory: 7124 grad_norm: 5.4269 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.7068 loss: 1.7068 2022/09/07 04:57:43 - mmengine - INFO - Epoch(train) [42][500/3757] lr: 1.0000e-02 eta: 9:44:17 time: 0.1574 data_time: 0.0107 memory: 7124 grad_norm: 5.4915 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 1.7341 loss: 1.7341 2022/09/07 04:57:59 - mmengine - INFO - Epoch(train) [42][600/3757] lr: 1.0000e-02 eta: 9:44:01 time: 0.1556 data_time: 0.0106 memory: 7124 grad_norm: 5.3981 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.6396 loss: 1.6396 2022/09/07 04:58:15 - mmengine - INFO - Epoch(train) [42][700/3757] lr: 1.0000e-02 eta: 9:43:46 time: 0.1615 data_time: 0.0098 memory: 7124 grad_norm: 5.3926 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4050 loss: 1.4050 2022/09/07 04:58:31 - mmengine - INFO - Epoch(train) [42][800/3757] lr: 1.0000e-02 eta: 9:43:30 time: 0.1561 data_time: 0.0093 memory: 7124 grad_norm: 5.1611 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7587 loss: 1.7587 2022/09/07 04:58:47 - mmengine - INFO - Epoch(train) [42][900/3757] lr: 1.0000e-02 eta: 9:43:14 time: 0.1592 data_time: 0.0101 memory: 7124 grad_norm: 5.4084 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.7994 loss: 1.7994 2022/09/07 04:58:57 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 04:59:03 - mmengine - INFO - Epoch(train) [42][1000/3757] lr: 1.0000e-02 eta: 9:42:58 time: 0.1576 data_time: 0.0096 memory: 7124 grad_norm: 5.2443 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6408 loss: 1.6408 2022/09/07 04:59:19 - mmengine - INFO - Epoch(train) [42][1100/3757] lr: 1.0000e-02 eta: 9:42:42 time: 0.1571 data_time: 0.0097 memory: 7124 grad_norm: 5.4264 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.6354 loss: 1.6354 2022/09/07 04:59:35 - mmengine - INFO - Epoch(train) [42][1200/3757] lr: 1.0000e-02 eta: 9:42:27 time: 0.1604 data_time: 0.0111 memory: 7124 grad_norm: 5.5381 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8805 loss: 1.8805 2022/09/07 04:59:51 - mmengine - INFO - Epoch(train) [42][1300/3757] lr: 1.0000e-02 eta: 9:42:11 time: 0.1614 data_time: 0.0107 memory: 7124 grad_norm: 5.2599 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.5349 loss: 1.5349 2022/09/07 05:00:07 - mmengine - INFO - Epoch(train) [42][1400/3757] lr: 1.0000e-02 eta: 9:41:55 time: 0.1622 data_time: 0.0099 memory: 7124 grad_norm: 5.4745 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 2.0279 loss: 2.0279 2022/09/07 05:00:23 - mmengine - INFO - Epoch(train) [42][1500/3757] lr: 1.0000e-02 eta: 9:41:39 time: 0.1571 data_time: 0.0091 memory: 7124 grad_norm: 5.1971 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.6899 loss: 1.6899 2022/09/07 05:00:39 - mmengine - INFO - Epoch(train) [42][1600/3757] lr: 1.0000e-02 eta: 9:41:24 time: 0.1563 data_time: 0.0119 memory: 7124 grad_norm: 5.4663 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8842 loss: 1.8842 2022/09/07 05:00:55 - mmengine - INFO - Epoch(train) [42][1700/3757] lr: 1.0000e-02 eta: 9:41:08 time: 0.1642 data_time: 0.0107 memory: 7124 grad_norm: 5.3767 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9007 loss: 1.9007 2022/09/07 05:01:11 - mmengine - INFO - Epoch(train) [42][1800/3757] lr: 1.0000e-02 eta: 9:40:52 time: 0.1571 data_time: 0.0098 memory: 7124 grad_norm: 5.1466 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6783 loss: 1.6783 2022/09/07 05:01:27 - mmengine - INFO - Epoch(train) [42][1900/3757] lr: 1.0000e-02 eta: 9:40:37 time: 0.1581 data_time: 0.0094 memory: 7124 grad_norm: 5.4014 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.7418 loss: 1.7418 2022/09/07 05:01:37 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:01:43 - mmengine - INFO - Epoch(train) [42][2000/3757] lr: 1.0000e-02 eta: 9:40:21 time: 0.1599 data_time: 0.0106 memory: 7124 grad_norm: 5.3081 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.5548 loss: 1.5548 2022/09/07 05:01:59 - mmengine - INFO - Epoch(train) [42][2100/3757] lr: 1.0000e-02 eta: 9:40:05 time: 0.1559 data_time: 0.0101 memory: 7124 grad_norm: 5.3484 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9577 loss: 1.9577 2022/09/07 05:02:15 - mmengine - INFO - Epoch(train) [42][2200/3757] lr: 1.0000e-02 eta: 9:39:49 time: 0.1550 data_time: 0.0105 memory: 7124 grad_norm: 5.2503 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.8213 loss: 1.8213 2022/09/07 05:02:31 - mmengine - INFO - Epoch(train) [42][2300/3757] lr: 1.0000e-02 eta: 9:39:33 time: 0.1555 data_time: 0.0097 memory: 7124 grad_norm: 5.2835 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9491 loss: 1.9491 2022/09/07 05:02:47 - mmengine - INFO - Epoch(train) [42][2400/3757] lr: 1.0000e-02 eta: 9:39:18 time: 0.1609 data_time: 0.0111 memory: 7124 grad_norm: 5.4290 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6598 loss: 1.6598 2022/09/07 05:03:03 - mmengine - INFO - Epoch(train) [42][2500/3757] lr: 1.0000e-02 eta: 9:39:02 time: 0.1558 data_time: 0.0092 memory: 7124 grad_norm: 5.3522 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.6160 loss: 1.6160 2022/09/07 05:03:19 - mmengine - INFO - Epoch(train) [42][2600/3757] lr: 1.0000e-02 eta: 9:38:46 time: 0.1603 data_time: 0.0096 memory: 7124 grad_norm: 5.3262 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.6593 loss: 1.6593 2022/09/07 05:03:35 - mmengine - INFO - Epoch(train) [42][2700/3757] lr: 1.0000e-02 eta: 9:38:30 time: 0.1566 data_time: 0.0102 memory: 7124 grad_norm: 5.3503 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.5773 loss: 1.5773 2022/09/07 05:03:51 - mmengine - INFO - Epoch(train) [42][2800/3757] lr: 1.0000e-02 eta: 9:38:15 time: 0.1613 data_time: 0.0098 memory: 7124 grad_norm: 5.3596 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8783 loss: 1.8783 2022/09/07 05:04:07 - mmengine - INFO - Epoch(train) [42][2900/3757] lr: 1.0000e-02 eta: 9:37:59 time: 0.1560 data_time: 0.0095 memory: 7124 grad_norm: 5.4253 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.8015 loss: 1.8015 2022/09/07 05:04:17 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:04:23 - mmengine - INFO - Epoch(train) [42][3000/3757] lr: 1.0000e-02 eta: 9:37:44 time: 0.1631 data_time: 0.0091 memory: 7124 grad_norm: 5.2620 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.5569 loss: 1.5569 2022/09/07 05:04:39 - mmengine - INFO - Epoch(train) [42][3100/3757] lr: 1.0000e-02 eta: 9:37:28 time: 0.1563 data_time: 0.0101 memory: 7124 grad_norm: 5.2848 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.8262 loss: 1.8262 2022/09/07 05:04:55 - mmengine - INFO - Epoch(train) [42][3200/3757] lr: 1.0000e-02 eta: 9:37:12 time: 0.1589 data_time: 0.0096 memory: 7124 grad_norm: 5.3299 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.4258 loss: 1.4258 2022/09/07 05:05:11 - mmengine - INFO - Epoch(train) [42][3300/3757] lr: 1.0000e-02 eta: 9:36:56 time: 0.1585 data_time: 0.0096 memory: 7124 grad_norm: 5.2316 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.8605 loss: 1.8605 2022/09/07 05:05:27 - mmengine - INFO - Epoch(train) [42][3400/3757] lr: 1.0000e-02 eta: 9:36:41 time: 0.1620 data_time: 0.0103 memory: 7124 grad_norm: 5.6911 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.1164 loss: 2.1164 2022/09/07 05:05:43 - mmengine - INFO - Epoch(train) [42][3500/3757] lr: 1.0000e-02 eta: 9:36:25 time: 0.1571 data_time: 0.0089 memory: 7124 grad_norm: 5.4526 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7468 loss: 1.7468 2022/09/07 05:05:59 - mmengine - INFO - Epoch(train) [42][3600/3757] lr: 1.0000e-02 eta: 9:36:09 time: 0.1575 data_time: 0.0100 memory: 7124 grad_norm: 5.4775 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7132 loss: 1.7132 2022/09/07 05:06:15 - mmengine - INFO - Epoch(train) [42][3700/3757] lr: 1.0000e-02 eta: 9:35:53 time: 0.1577 data_time: 0.0105 memory: 7124 grad_norm: 5.1485 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8377 loss: 1.8377 2022/09/07 05:06:24 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:06:24 - mmengine - INFO - Epoch(train) [42][3757/3757] lr: 1.0000e-02 eta: 9:35:47 time: 0.1359 data_time: 0.0072 memory: 7124 grad_norm: 5.3932 top1_acc: 0.7143 top5_acc: 1.0000 loss_cls: 1.6479 loss: 1.6479 2022/09/07 05:06:24 - mmengine - INFO - Saving checkpoint at 42 epochs 2022/09/07 05:08:43 - mmengine - INFO - Epoch(val) [42][100/310] eta: 0:04:16 time: 1.2208 data_time: 0.9147 memory: 7627 2022/09/07 05:10:57 - mmengine - INFO - Epoch(val) [42][200/310] eta: 0:02:09 time: 1.1734 data_time: 0.8748 memory: 7627 2022/09/07 05:13:03 - mmengine - INFO - Epoch(val) [42][300/310] eta: 0:00:12 time: 1.2686 data_time: 0.9680 memory: 7627 2022/09/07 05:13:21 - mmengine - INFO - Epoch(val) [42][310/310] acc/top1: 0.6557 acc/top5: 0.8660 acc/mean1: 0.6557 2022/09/07 05:13:39 - mmengine - INFO - Epoch(train) [43][100/3757] lr: 1.0000e-02 eta: 9:35:27 time: 0.1587 data_time: 0.0100 memory: 7627 grad_norm: 5.3953 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9592 loss: 1.9592 2022/09/07 05:13:55 - mmengine - INFO - Epoch(train) [43][200/3757] lr: 1.0000e-02 eta: 9:35:11 time: 0.1591 data_time: 0.0095 memory: 7124 grad_norm: 5.4878 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.5932 loss: 1.5932 2022/09/07 05:13:56 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:14:11 - mmengine - INFO - Epoch(train) [43][300/3757] lr: 1.0000e-02 eta: 9:34:55 time: 0.1587 data_time: 0.0105 memory: 7124 grad_norm: 5.4114 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.6502 loss: 1.6502 2022/09/07 05:14:27 - mmengine - INFO - Epoch(train) [43][400/3757] lr: 1.0000e-02 eta: 9:34:39 time: 0.1587 data_time: 0.0099 memory: 7124 grad_norm: 5.3457 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.7150 loss: 1.7150 2022/09/07 05:14:43 - mmengine - INFO - Epoch(train) [43][500/3757] lr: 1.0000e-02 eta: 9:34:24 time: 0.1586 data_time: 0.0111 memory: 7124 grad_norm: 5.3577 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.5892 loss: 1.5892 2022/09/07 05:14:59 - mmengine - INFO - Epoch(train) [43][600/3757] lr: 1.0000e-02 eta: 9:34:08 time: 0.1581 data_time: 0.0094 memory: 7124 grad_norm: 5.4774 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.8433 loss: 1.8433 2022/09/07 05:15:15 - mmengine - INFO - Epoch(train) [43][700/3757] lr: 1.0000e-02 eta: 9:33:52 time: 0.1602 data_time: 0.0106 memory: 7124 grad_norm: 5.3538 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.5874 loss: 1.5874 2022/09/07 05:15:31 - mmengine - INFO - Epoch(train) [43][800/3757] lr: 1.0000e-02 eta: 9:33:36 time: 0.1576 data_time: 0.0085 memory: 7124 grad_norm: 5.2748 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.5168 loss: 1.5168 2022/09/07 05:15:46 - mmengine - INFO - Epoch(train) [43][900/3757] lr: 1.0000e-02 eta: 9:33:20 time: 0.1574 data_time: 0.0105 memory: 7124 grad_norm: 5.5900 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.8087 loss: 1.8087 2022/09/07 05:16:02 - mmengine - INFO - Epoch(train) [43][1000/3757] lr: 1.0000e-02 eta: 9:33:04 time: 0.1557 data_time: 0.0103 memory: 7124 grad_norm: 5.2487 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.8170 loss: 1.8170 2022/09/07 05:16:18 - mmengine - INFO - Epoch(train) [43][1100/3757] lr: 1.0000e-02 eta: 9:32:49 time: 0.1649 data_time: 0.0105 memory: 7124 grad_norm: 5.4028 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5814 loss: 1.5814 2022/09/07 05:16:34 - mmengine - INFO - Epoch(train) [43][1200/3757] lr: 1.0000e-02 eta: 9:32:33 time: 0.1540 data_time: 0.0102 memory: 7124 grad_norm: 5.3909 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.1261 loss: 2.1261 2022/09/07 05:16:35 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:16:50 - mmengine - INFO - Epoch(train) [43][1300/3757] lr: 1.0000e-02 eta: 9:32:17 time: 0.1599 data_time: 0.0092 memory: 7124 grad_norm: 5.2173 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9038 loss: 1.9038 2022/09/07 05:17:06 - mmengine - INFO - Epoch(train) [43][1400/3757] lr: 1.0000e-02 eta: 9:32:01 time: 0.1573 data_time: 0.0115 memory: 7124 grad_norm: 5.2798 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.6932 loss: 1.6932 2022/09/07 05:17:22 - mmengine - INFO - Epoch(train) [43][1500/3757] lr: 1.0000e-02 eta: 9:31:45 time: 0.1576 data_time: 0.0103 memory: 7124 grad_norm: 5.5056 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8543 loss: 1.8543 2022/09/07 05:17:38 - mmengine - INFO - Epoch(train) [43][1600/3757] lr: 1.0000e-02 eta: 9:31:29 time: 0.1612 data_time: 0.0106 memory: 7124 grad_norm: 5.2695 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.6175 loss: 1.6175 2022/09/07 05:17:54 - mmengine - INFO - Epoch(train) [43][1700/3757] lr: 1.0000e-02 eta: 9:31:14 time: 0.1569 data_time: 0.0093 memory: 7124 grad_norm: 5.3799 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9055 loss: 1.9055 2022/09/07 05:18:10 - mmengine - INFO - Epoch(train) [43][1800/3757] lr: 1.0000e-02 eta: 9:30:58 time: 0.1617 data_time: 0.0099 memory: 7124 grad_norm: 5.4161 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.4936 loss: 1.4936 2022/09/07 05:18:26 - mmengine - INFO - Epoch(train) [43][1900/3757] lr: 1.0000e-02 eta: 9:30:42 time: 0.1563 data_time: 0.0098 memory: 7124 grad_norm: 5.2416 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 2.0674 loss: 2.0674 2022/09/07 05:18:41 - mmengine - INFO - Epoch(train) [43][2000/3757] lr: 1.0000e-02 eta: 9:30:26 time: 0.1558 data_time: 0.0090 memory: 7124 grad_norm: 5.3714 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 2.0004 loss: 2.0004 2022/09/07 05:18:58 - mmengine - INFO - Epoch(train) [43][2100/3757] lr: 1.0000e-02 eta: 9:30:10 time: 0.1595 data_time: 0.0102 memory: 7124 grad_norm: 5.3213 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.7269 loss: 1.7269 2022/09/07 05:19:14 - mmengine - INFO - Epoch(train) [43][2200/3757] lr: 1.0000e-02 eta: 9:29:55 time: 0.1551 data_time: 0.0092 memory: 7124 grad_norm: 5.4710 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7039 loss: 1.7039 2022/09/07 05:19:15 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:19:30 - mmengine - INFO - Epoch(train) [43][2300/3757] lr: 1.0000e-02 eta: 9:29:39 time: 0.1613 data_time: 0.0114 memory: 7124 grad_norm: 5.2576 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9864 loss: 1.9864 2022/09/07 05:19:46 - mmengine - INFO - Epoch(train) [43][2400/3757] lr: 1.0000e-02 eta: 9:29:23 time: 0.1557 data_time: 0.0103 memory: 7124 grad_norm: 5.5793 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7764 loss: 1.7764 2022/09/07 05:20:02 - mmengine - INFO - Epoch(train) [43][2500/3757] lr: 1.0000e-02 eta: 9:29:07 time: 0.1624 data_time: 0.0104 memory: 7124 grad_norm: 5.5205 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7173 loss: 1.7173 2022/09/07 05:20:18 - mmengine - INFO - Epoch(train) [43][2600/3757] lr: 1.0000e-02 eta: 9:28:52 time: 0.1587 data_time: 0.0117 memory: 7124 grad_norm: 5.4811 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 1.7847 loss: 1.7847 2022/09/07 05:20:34 - mmengine - INFO - Epoch(train) [43][2700/3757] lr: 1.0000e-02 eta: 9:28:36 time: 0.1573 data_time: 0.0111 memory: 7124 grad_norm: 5.3269 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.9761 loss: 1.9761 2022/09/07 05:20:49 - mmengine - INFO - Epoch(train) [43][2800/3757] lr: 1.0000e-02 eta: 9:28:20 time: 0.1587 data_time: 0.0103 memory: 7124 grad_norm: 5.5127 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8771 loss: 1.8771 2022/09/07 05:21:05 - mmengine - INFO - Epoch(train) [43][2900/3757] lr: 1.0000e-02 eta: 9:28:04 time: 0.1597 data_time: 0.0100 memory: 7124 grad_norm: 5.4184 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 2.0826 loss: 2.0826 2022/09/07 05:21:22 - mmengine - INFO - Epoch(train) [43][3000/3757] lr: 1.0000e-02 eta: 9:27:49 time: 0.1594 data_time: 0.0096 memory: 7124 grad_norm: 5.3255 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.7347 loss: 1.7347 2022/09/07 05:21:37 - mmengine - INFO - Epoch(train) [43][3100/3757] lr: 1.0000e-02 eta: 9:27:33 time: 0.1598 data_time: 0.0099 memory: 7124 grad_norm: 5.7449 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.6578 loss: 1.6578 2022/09/07 05:21:53 - mmengine - INFO - Epoch(train) [43][3200/3757] lr: 1.0000e-02 eta: 9:27:17 time: 0.1557 data_time: 0.0084 memory: 7124 grad_norm: 5.5593 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.5359 loss: 1.5359 2022/09/07 05:21:54 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:22:09 - mmengine - INFO - Epoch(train) [43][3300/3757] lr: 1.0000e-02 eta: 9:27:01 time: 0.1608 data_time: 0.0095 memory: 7124 grad_norm: 5.5956 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.6478 loss: 1.6478 2022/09/07 05:22:25 - mmengine - INFO - Epoch(train) [43][3400/3757] lr: 1.0000e-02 eta: 9:26:45 time: 0.1562 data_time: 0.0097 memory: 7124 grad_norm: 5.6514 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8566 loss: 1.8566 2022/09/07 05:22:41 - mmengine - INFO - Epoch(train) [43][3500/3757] lr: 1.0000e-02 eta: 9:26:30 time: 0.1539 data_time: 0.0089 memory: 7124 grad_norm: 5.4561 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6663 loss: 1.6663 2022/09/07 05:22:57 - mmengine - INFO - Epoch(train) [43][3600/3757] lr: 1.0000e-02 eta: 9:26:14 time: 0.1581 data_time: 0.0104 memory: 7124 grad_norm: 5.4470 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7645 loss: 1.7645 2022/09/07 05:23:13 - mmengine - INFO - Epoch(train) [43][3700/3757] lr: 1.0000e-02 eta: 9:25:58 time: 0.1582 data_time: 0.0111 memory: 7124 grad_norm: 5.2332 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7977 loss: 1.7977 2022/09/07 05:23:22 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:23:22 - mmengine - INFO - Epoch(train) [43][3757/3757] lr: 1.0000e-02 eta: 9:25:52 time: 0.1363 data_time: 0.0070 memory: 7124 grad_norm: 5.6204 top1_acc: 0.5714 top5_acc: 0.8571 loss_cls: 1.4809 loss: 1.4809 2022/09/07 05:23:22 - mmengine - INFO - Saving checkpoint at 43 epochs 2022/09/07 05:25:41 - mmengine - INFO - Epoch(val) [43][100/310] eta: 0:04:17 time: 1.2280 data_time: 0.9282 memory: 7627 2022/09/07 05:27:54 - mmengine - INFO - Epoch(val) [43][200/310] eta: 0:02:05 time: 1.1424 data_time: 0.8420 memory: 7627 2022/09/07 05:30:01 - mmengine - INFO - Epoch(val) [43][300/310] eta: 0:00:12 time: 1.2802 data_time: 0.9819 memory: 7627 2022/09/07 05:30:19 - mmengine - INFO - Epoch(val) [43][310/310] acc/top1: 0.6584 acc/top5: 0.8711 acc/mean1: 0.6582 2022/09/07 05:30:37 - mmengine - INFO - Epoch(train) [44][100/3757] lr: 1.0000e-02 eta: 9:25:32 time: 0.1585 data_time: 0.0089 memory: 7627 grad_norm: 5.4875 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.7500 loss: 1.7500 2022/09/07 05:30:53 - mmengine - INFO - Epoch(train) [44][200/3757] lr: 1.0000e-02 eta: 9:25:16 time: 0.1577 data_time: 0.0104 memory: 7124 grad_norm: 5.6012 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7598 loss: 1.7598 2022/09/07 05:31:09 - mmengine - INFO - Epoch(train) [44][300/3757] lr: 1.0000e-02 eta: 9:25:00 time: 0.1546 data_time: 0.0095 memory: 7124 grad_norm: 5.5313 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.5734 loss: 1.5734 2022/09/07 05:31:25 - mmengine - INFO - Epoch(train) [44][400/3757] lr: 1.0000e-02 eta: 9:24:44 time: 0.1575 data_time: 0.0104 memory: 7124 grad_norm: 5.5541 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6576 loss: 1.6576 2022/09/07 05:31:33 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:31:41 - mmengine - INFO - Epoch(train) [44][500/3757] lr: 1.0000e-02 eta: 9:24:29 time: 0.1670 data_time: 0.0106 memory: 7124 grad_norm: 5.4903 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.5130 loss: 1.5130 2022/09/07 05:31:57 - mmengine - INFO - Epoch(train) [44][600/3757] lr: 1.0000e-02 eta: 9:24:13 time: 0.1556 data_time: 0.0101 memory: 7124 grad_norm: 5.4005 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.7677 loss: 1.7677 2022/09/07 05:32:13 - mmengine - INFO - Epoch(train) [44][700/3757] lr: 1.0000e-02 eta: 9:23:57 time: 0.1577 data_time: 0.0111 memory: 7124 grad_norm: 5.4381 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5038 loss: 1.5038 2022/09/07 05:32:29 - mmengine - INFO - Epoch(train) [44][800/3757] lr: 1.0000e-02 eta: 9:23:42 time: 0.1605 data_time: 0.0113 memory: 7124 grad_norm: 5.4348 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7325 loss: 1.7325 2022/09/07 05:32:45 - mmengine - INFO - Epoch(train) [44][900/3757] lr: 1.0000e-02 eta: 9:23:26 time: 0.1564 data_time: 0.0095 memory: 7124 grad_norm: 5.4054 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6099 loss: 1.6099 2022/09/07 05:33:01 - mmengine - INFO - Epoch(train) [44][1000/3757] lr: 1.0000e-02 eta: 9:23:10 time: 0.1662 data_time: 0.0099 memory: 7124 grad_norm: 5.5562 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8411 loss: 1.8411 2022/09/07 05:33:17 - mmengine - INFO - Epoch(train) [44][1100/3757] lr: 1.0000e-02 eta: 9:22:54 time: 0.1586 data_time: 0.0105 memory: 7124 grad_norm: 5.6318 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.5804 loss: 1.5804 2022/09/07 05:33:33 - mmengine - INFO - Epoch(train) [44][1200/3757] lr: 1.0000e-02 eta: 9:22:39 time: 0.1573 data_time: 0.0103 memory: 7124 grad_norm: 5.4832 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.8698 loss: 1.8698 2022/09/07 05:33:49 - mmengine - INFO - Epoch(train) [44][1300/3757] lr: 1.0000e-02 eta: 9:22:23 time: 0.1601 data_time: 0.0097 memory: 7124 grad_norm: 5.3256 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.0715 loss: 2.0715 2022/09/07 05:34:05 - mmengine - INFO - Epoch(train) [44][1400/3757] lr: 1.0000e-02 eta: 9:22:07 time: 0.1604 data_time: 0.0109 memory: 7124 grad_norm: 5.3129 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7394 loss: 1.7394 2022/09/07 05:34:13 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:34:21 - mmengine - INFO - Epoch(train) [44][1500/3757] lr: 1.0000e-02 eta: 9:21:51 time: 0.1581 data_time: 0.0108 memory: 7124 grad_norm: 5.5485 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6032 loss: 1.6032 2022/09/07 05:34:37 - mmengine - INFO - Epoch(train) [44][1600/3757] lr: 1.0000e-02 eta: 9:21:35 time: 0.1607 data_time: 0.0094 memory: 7124 grad_norm: 5.1070 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7661 loss: 1.7661 2022/09/07 05:34:53 - mmengine - INFO - Epoch(train) [44][1700/3757] lr: 1.0000e-02 eta: 9:21:19 time: 0.1579 data_time: 0.0100 memory: 7124 grad_norm: 5.5107 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6814 loss: 1.6814 2022/09/07 05:35:09 - mmengine - INFO - Epoch(train) [44][1800/3757] lr: 1.0000e-02 eta: 9:21:04 time: 0.1650 data_time: 0.0108 memory: 7124 grad_norm: 5.2724 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8323 loss: 1.8323 2022/09/07 05:35:25 - mmengine - INFO - Epoch(train) [44][1900/3757] lr: 1.0000e-02 eta: 9:20:48 time: 0.1590 data_time: 0.0103 memory: 7124 grad_norm: 5.3405 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7304 loss: 1.7304 2022/09/07 05:35:41 - mmengine - INFO - Epoch(train) [44][2000/3757] lr: 1.0000e-02 eta: 9:20:32 time: 0.1577 data_time: 0.0093 memory: 7124 grad_norm: 5.3985 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.5705 loss: 1.5705 2022/09/07 05:35:57 - mmengine - INFO - Epoch(train) [44][2100/3757] lr: 1.0000e-02 eta: 9:20:16 time: 0.1578 data_time: 0.0088 memory: 7124 grad_norm: 5.4947 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8140 loss: 1.8140 2022/09/07 05:36:13 - mmengine - INFO - Epoch(train) [44][2200/3757] lr: 1.0000e-02 eta: 9:20:01 time: 0.1580 data_time: 0.0097 memory: 7124 grad_norm: 5.3598 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6693 loss: 1.6693 2022/09/07 05:36:29 - mmengine - INFO - Epoch(train) [44][2300/3757] lr: 1.0000e-02 eta: 9:19:45 time: 0.1621 data_time: 0.0093 memory: 7124 grad_norm: 5.2987 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.7160 loss: 1.7160 2022/09/07 05:36:45 - mmengine - INFO - Epoch(train) [44][2400/3757] lr: 1.0000e-02 eta: 9:19:29 time: 0.1597 data_time: 0.0089 memory: 7124 grad_norm: 5.4103 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.6154 loss: 1.6154 2022/09/07 05:36:53 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:37:01 - mmengine - INFO - Epoch(train) [44][2500/3757] lr: 1.0000e-02 eta: 9:19:14 time: 0.1571 data_time: 0.0095 memory: 7124 grad_norm: 5.3747 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.7134 loss: 1.7134 2022/09/07 05:37:17 - mmengine - INFO - Epoch(train) [44][2600/3757] lr: 1.0000e-02 eta: 9:18:58 time: 0.1565 data_time: 0.0096 memory: 7124 grad_norm: 5.4321 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 2.0262 loss: 2.0262 2022/09/07 05:37:33 - mmengine - INFO - Epoch(train) [44][2700/3757] lr: 1.0000e-02 eta: 9:18:42 time: 0.1619 data_time: 0.0111 memory: 7124 grad_norm: 5.1790 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9444 loss: 1.9444 2022/09/07 05:37:49 - mmengine - INFO - Epoch(train) [44][2800/3757] lr: 1.0000e-02 eta: 9:18:26 time: 0.1622 data_time: 0.0113 memory: 7124 grad_norm: 5.3242 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8938 loss: 1.8938 2022/09/07 05:38:05 - mmengine - INFO - Epoch(train) [44][2900/3757] lr: 1.0000e-02 eta: 9:18:11 time: 0.1566 data_time: 0.0098 memory: 7124 grad_norm: 5.5163 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8455 loss: 1.8455 2022/09/07 05:38:21 - mmengine - INFO - Epoch(train) [44][3000/3757] lr: 1.0000e-02 eta: 9:17:55 time: 0.1578 data_time: 0.0098 memory: 7124 grad_norm: 5.5196 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.8715 loss: 1.8715 2022/09/07 05:38:37 - mmengine - INFO - Epoch(train) [44][3100/3757] lr: 1.0000e-02 eta: 9:17:39 time: 0.1573 data_time: 0.0105 memory: 7124 grad_norm: 5.2685 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.4400 loss: 1.4400 2022/09/07 05:38:53 - mmengine - INFO - Epoch(train) [44][3200/3757] lr: 1.0000e-02 eta: 9:17:23 time: 0.1597 data_time: 0.0104 memory: 7124 grad_norm: 5.3608 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.5928 loss: 1.5928 2022/09/07 05:39:09 - mmengine - INFO - Epoch(train) [44][3300/3757] lr: 1.0000e-02 eta: 9:17:08 time: 0.1574 data_time: 0.0100 memory: 7124 grad_norm: 5.4774 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9621 loss: 1.9621 2022/09/07 05:39:25 - mmengine - INFO - Epoch(train) [44][3400/3757] lr: 1.0000e-02 eta: 9:16:52 time: 0.1609 data_time: 0.0116 memory: 7124 grad_norm: 5.4341 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7953 loss: 1.7953 2022/09/07 05:39:33 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:39:41 - mmengine - INFO - Epoch(train) [44][3500/3757] lr: 1.0000e-02 eta: 9:16:36 time: 0.1608 data_time: 0.0101 memory: 7124 grad_norm: 5.6506 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.5785 loss: 1.5785 2022/09/07 05:39:57 - mmengine - INFO - Epoch(train) [44][3600/3757] lr: 1.0000e-02 eta: 9:16:21 time: 0.1579 data_time: 0.0094 memory: 7124 grad_norm: 5.4346 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6993 loss: 1.6993 2022/09/07 05:40:13 - mmengine - INFO - Epoch(train) [44][3700/3757] lr: 1.0000e-02 eta: 9:16:05 time: 0.1573 data_time: 0.0098 memory: 7124 grad_norm: 5.2064 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.7636 loss: 1.7636 2022/09/07 05:40:22 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:40:22 - mmengine - INFO - Epoch(train) [44][3757/3757] lr: 1.0000e-02 eta: 9:15:58 time: 0.1368 data_time: 0.0073 memory: 7124 grad_norm: 5.4258 top1_acc: 0.7143 top5_acc: 0.7143 loss_cls: 1.5926 loss: 1.5926 2022/09/07 05:40:22 - mmengine - INFO - Saving checkpoint at 44 epochs 2022/09/07 05:42:40 - mmengine - INFO - Epoch(val) [44][100/310] eta: 0:03:59 time: 1.1399 data_time: 0.8412 memory: 7627 2022/09/07 05:44:57 - mmengine - INFO - Epoch(val) [44][200/310] eta: 0:02:22 time: 1.2947 data_time: 0.9949 memory: 7627 2022/09/07 05:47:01 - mmengine - INFO - Epoch(val) [44][300/310] eta: 0:00:11 time: 1.1280 data_time: 0.8235 memory: 7627 2022/09/07 05:47:18 - mmengine - INFO - Epoch(val) [44][310/310] acc/top1: 0.6629 acc/top5: 0.8709 acc/mean1: 0.6626 2022/09/07 05:47:18 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_37.pth is removed 2022/09/07 05:47:20 - mmengine - INFO - The best checkpoint with 0.6629 acc/top1 at 44 epoch is saved to best_acc/top1_epoch_44.pth. 2022/09/07 05:47:37 - mmengine - INFO - Epoch(train) [45][100/3757] lr: 1.0000e-02 eta: 9:15:37 time: 0.1619 data_time: 0.0092 memory: 7627 grad_norm: 5.1907 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8669 loss: 1.8669 2022/09/07 05:47:53 - mmengine - INFO - Epoch(train) [45][200/3757] lr: 1.0000e-02 eta: 9:15:22 time: 0.1595 data_time: 0.0107 memory: 7124 grad_norm: 5.5287 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.8030 loss: 1.8030 2022/09/07 05:48:09 - mmengine - INFO - Epoch(train) [45][300/3757] lr: 1.0000e-02 eta: 9:15:06 time: 0.1587 data_time: 0.0100 memory: 7124 grad_norm: 5.4386 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.7517 loss: 1.7517 2022/09/07 05:48:25 - mmengine - INFO - Epoch(train) [45][400/3757] lr: 1.0000e-02 eta: 9:14:50 time: 0.1577 data_time: 0.0095 memory: 7124 grad_norm: 5.4585 top1_acc: 0.1250 top5_acc: 0.8750 loss_cls: 1.8899 loss: 1.8899 2022/09/07 05:48:41 - mmengine - INFO - Epoch(train) [45][500/3757] lr: 1.0000e-02 eta: 9:14:34 time: 0.1614 data_time: 0.0101 memory: 7124 grad_norm: 5.2951 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.9441 loss: 1.9441 2022/09/07 05:48:56 - mmengine - INFO - Epoch(train) [45][600/3757] lr: 1.0000e-02 eta: 9:14:18 time: 0.1573 data_time: 0.0100 memory: 7124 grad_norm: 5.4714 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.5133 loss: 1.5133 2022/09/07 05:49:11 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:49:13 - mmengine - INFO - Epoch(train) [45][700/3757] lr: 1.0000e-02 eta: 9:14:03 time: 0.1665 data_time: 0.0102 memory: 7124 grad_norm: 5.4795 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.8494 loss: 1.8494 2022/09/07 05:49:28 - mmengine - INFO - Epoch(train) [45][800/3757] lr: 1.0000e-02 eta: 9:13:47 time: 0.1567 data_time: 0.0106 memory: 7124 grad_norm: 5.5079 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.6661 loss: 1.6661 2022/09/07 05:49:45 - mmengine - INFO - Epoch(train) [45][900/3757] lr: 1.0000e-02 eta: 9:13:31 time: 0.1611 data_time: 0.0105 memory: 7124 grad_norm: 5.4488 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.7707 loss: 1.7707 2022/09/07 05:50:01 - mmengine - INFO - Epoch(train) [45][1000/3757] lr: 1.0000e-02 eta: 9:13:15 time: 0.1593 data_time: 0.0123 memory: 7124 grad_norm: 5.5420 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.5278 loss: 1.5278 2022/09/07 05:50:17 - mmengine - INFO - Epoch(train) [45][1100/3757] lr: 1.0000e-02 eta: 9:13:00 time: 0.1592 data_time: 0.0100 memory: 7124 grad_norm: 5.2673 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8417 loss: 1.8417 2022/09/07 05:50:33 - mmengine - INFO - Epoch(train) [45][1200/3757] lr: 1.0000e-02 eta: 9:12:44 time: 0.1585 data_time: 0.0098 memory: 7124 grad_norm: 5.3640 top1_acc: 0.1250 top5_acc: 0.7500 loss_cls: 1.4606 loss: 1.4606 2022/09/07 05:50:49 - mmengine - INFO - Epoch(train) [45][1300/3757] lr: 1.0000e-02 eta: 9:12:28 time: 0.1639 data_time: 0.0112 memory: 7124 grad_norm: 5.3693 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.5106 loss: 1.5106 2022/09/07 05:51:05 - mmengine - INFO - Epoch(train) [45][1400/3757] lr: 1.0000e-02 eta: 9:12:12 time: 0.1555 data_time: 0.0091 memory: 7124 grad_norm: 5.4864 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.6992 loss: 1.6992 2022/09/07 05:51:21 - mmengine - INFO - Epoch(train) [45][1500/3757] lr: 1.0000e-02 eta: 9:11:57 time: 0.1592 data_time: 0.0112 memory: 7124 grad_norm: 5.5603 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.9068 loss: 1.9068 2022/09/07 05:51:37 - mmengine - INFO - Epoch(train) [45][1600/3757] lr: 1.0000e-02 eta: 9:11:41 time: 0.1589 data_time: 0.0091 memory: 7124 grad_norm: 5.4907 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8341 loss: 1.8341 2022/09/07 05:51:51 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:51:53 - mmengine - INFO - Epoch(train) [45][1700/3757] lr: 1.0000e-02 eta: 9:11:25 time: 0.1572 data_time: 0.0108 memory: 7124 grad_norm: 5.3003 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.5992 loss: 1.5992 2022/09/07 05:52:09 - mmengine - INFO - Epoch(train) [45][1800/3757] lr: 1.0000e-02 eta: 9:11:09 time: 0.1567 data_time: 0.0107 memory: 7124 grad_norm: 5.3331 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8121 loss: 1.8121 2022/09/07 05:52:25 - mmengine - INFO - Epoch(train) [45][1900/3757] lr: 1.0000e-02 eta: 9:10:54 time: 0.1608 data_time: 0.0109 memory: 7124 grad_norm: 5.7030 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9667 loss: 1.9667 2022/09/07 05:52:40 - mmengine - INFO - Epoch(train) [45][2000/3757] lr: 1.0000e-02 eta: 9:10:38 time: 0.1631 data_time: 0.0098 memory: 7124 grad_norm: 5.2274 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.5191 loss: 1.5191 2022/09/07 05:52:56 - mmengine - INFO - Epoch(train) [45][2100/3757] lr: 1.0000e-02 eta: 9:10:22 time: 0.1552 data_time: 0.0104 memory: 7124 grad_norm: 5.4370 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.6677 loss: 1.6677 2022/09/07 05:53:12 - mmengine - INFO - Epoch(train) [45][2200/3757] lr: 1.0000e-02 eta: 9:10:06 time: 0.1602 data_time: 0.0108 memory: 7124 grad_norm: 5.5106 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9448 loss: 1.9448 2022/09/07 05:53:28 - mmengine - INFO - Epoch(train) [45][2300/3757] lr: 1.0000e-02 eta: 9:09:50 time: 0.1559 data_time: 0.0103 memory: 7124 grad_norm: 5.3733 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.7016 loss: 1.7016 2022/09/07 05:53:44 - mmengine - INFO - Epoch(train) [45][2400/3757] lr: 1.0000e-02 eta: 9:09:35 time: 0.1557 data_time: 0.0091 memory: 7124 grad_norm: 5.5040 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.7228 loss: 1.7228 2022/09/07 05:54:00 - mmengine - INFO - Epoch(train) [45][2500/3757] lr: 1.0000e-02 eta: 9:09:19 time: 0.1580 data_time: 0.0097 memory: 7124 grad_norm: 5.3370 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.8341 loss: 1.8341 2022/09/07 05:54:16 - mmengine - INFO - Epoch(train) [45][2600/3757] lr: 1.0000e-02 eta: 9:09:03 time: 0.1606 data_time: 0.0097 memory: 7124 grad_norm: 5.4657 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.5335 loss: 1.5335 2022/09/07 05:54:31 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:54:32 - mmengine - INFO - Epoch(train) [45][2700/3757] lr: 1.0000e-02 eta: 9:08:47 time: 0.1588 data_time: 0.0095 memory: 7124 grad_norm: 5.2681 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.7836 loss: 1.7836 2022/09/07 05:54:48 - mmengine - INFO - Epoch(train) [45][2800/3757] lr: 1.0000e-02 eta: 9:08:31 time: 0.1626 data_time: 0.0090 memory: 7124 grad_norm: 5.2897 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.7528 loss: 1.7528 2022/09/07 05:55:04 - mmengine - INFO - Epoch(train) [45][2900/3757] lr: 1.0000e-02 eta: 9:08:16 time: 0.1620 data_time: 0.0095 memory: 7124 grad_norm: 5.4585 top1_acc: 0.2500 top5_acc: 0.8750 loss_cls: 1.7545 loss: 1.7545 2022/09/07 05:55:20 - mmengine - INFO - Epoch(train) [45][3000/3757] lr: 1.0000e-02 eta: 9:08:00 time: 0.1613 data_time: 0.0107 memory: 7124 grad_norm: 5.3624 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8866 loss: 1.8866 2022/09/07 05:55:36 - mmengine - INFO - Epoch(train) [45][3100/3757] lr: 1.0000e-02 eta: 9:07:44 time: 0.1566 data_time: 0.0109 memory: 7124 grad_norm: 5.4005 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.5322 loss: 1.5322 2022/09/07 05:55:53 - mmengine - INFO - Epoch(train) [45][3200/3757] lr: 1.0000e-02 eta: 9:07:29 time: 0.1599 data_time: 0.0105 memory: 7124 grad_norm: 5.4182 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.6366 loss: 1.6366 2022/09/07 05:56:09 - mmengine - INFO - Epoch(train) [45][3300/3757] lr: 1.0000e-02 eta: 9:07:13 time: 0.1546 data_time: 0.0107 memory: 7124 grad_norm: 5.5026 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.5495 loss: 1.5495 2022/09/07 05:56:25 - mmengine - INFO - Epoch(train) [45][3400/3757] lr: 1.0000e-02 eta: 9:06:58 time: 0.1566 data_time: 0.0097 memory: 7124 grad_norm: 5.4875 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9846 loss: 1.9846 2022/09/07 05:56:41 - mmengine - INFO - Epoch(train) [45][3500/3757] lr: 1.0000e-02 eta: 9:06:42 time: 0.1633 data_time: 0.0096 memory: 7124 grad_norm: 5.5148 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7298 loss: 1.7298 2022/09/07 05:56:57 - mmengine - INFO - Epoch(train) [45][3600/3757] lr: 1.0000e-02 eta: 9:06:26 time: 0.1563 data_time: 0.0092 memory: 7124 grad_norm: 5.4695 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8584 loss: 1.8584 2022/09/07 05:57:11 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:57:13 - mmengine - INFO - Epoch(train) [45][3700/3757] lr: 1.0000e-02 eta: 9:06:10 time: 0.1581 data_time: 0.0104 memory: 7124 grad_norm: 5.2914 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.6171 loss: 1.6171 2022/09/07 05:57:21 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 05:57:21 - mmengine - INFO - Epoch(train) [45][3757/3757] lr: 1.0000e-02 eta: 9:06:04 time: 0.1401 data_time: 0.0096 memory: 7124 grad_norm: 5.5959 top1_acc: 0.4286 top5_acc: 0.8571 loss_cls: 1.9527 loss: 1.9527 2022/09/07 05:57:21 - mmengine - INFO - Saving checkpoint at 45 epochs 2022/09/07 05:59:40 - mmengine - INFO - Epoch(val) [45][100/310] eta: 0:04:06 time: 1.1752 data_time: 0.8736 memory: 7627 2022/09/07 06:01:57 - mmengine - INFO - Epoch(val) [45][200/310] eta: 0:02:02 time: 1.1170 data_time: 0.8176 memory: 7627 2022/09/07 06:04:04 - mmengine - INFO - Epoch(val) [45][300/310] eta: 0:00:12 time: 1.2455 data_time: 0.9465 memory: 7627 2022/09/07 06:04:20 - mmengine - INFO - Epoch(val) [45][310/310] acc/top1: 0.6549 acc/top5: 0.8694 acc/mean1: 0.6547 2022/09/07 06:04:38 - mmengine - INFO - Epoch(train) [46][100/3757] lr: 1.0000e-02 eta: 9:05:44 time: 0.1619 data_time: 0.0094 memory: 7627 grad_norm: 5.5127 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.7937 loss: 1.7937 2022/09/07 06:04:54 - mmengine - INFO - Epoch(train) [46][200/3757] lr: 1.0000e-02 eta: 9:05:28 time: 0.1594 data_time: 0.0103 memory: 7124 grad_norm: 5.3562 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4165 loss: 1.4165 2022/09/07 06:05:10 - mmengine - INFO - Epoch(train) [46][300/3757] lr: 1.0000e-02 eta: 9:05:12 time: 0.1652 data_time: 0.0163 memory: 7124 grad_norm: 5.6262 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.5936 loss: 1.5936 2022/09/07 06:05:26 - mmengine - INFO - Epoch(train) [46][400/3757] lr: 1.0000e-02 eta: 9:04:57 time: 0.1632 data_time: 0.0098 memory: 7124 grad_norm: 5.1722 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6777 loss: 1.6777 2022/09/07 06:05:42 - mmengine - INFO - Epoch(train) [46][500/3757] lr: 1.0000e-02 eta: 9:04:41 time: 0.1570 data_time: 0.0105 memory: 7124 grad_norm: 5.2755 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6493 loss: 1.6493 2022/09/07 06:05:58 - mmengine - INFO - Epoch(train) [46][600/3757] lr: 1.0000e-02 eta: 9:04:25 time: 0.1579 data_time: 0.0104 memory: 7124 grad_norm: 5.3290 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.7514 loss: 1.7514 2022/09/07 06:06:14 - mmengine - INFO - Epoch(train) [46][700/3757] lr: 1.0000e-02 eta: 9:04:09 time: 0.1549 data_time: 0.0102 memory: 7124 grad_norm: 5.3216 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7195 loss: 1.7195 2022/09/07 06:06:30 - mmengine - INFO - Epoch(train) [46][800/3757] lr: 1.0000e-02 eta: 9:03:53 time: 0.1584 data_time: 0.0096 memory: 7124 grad_norm: 5.5362 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.5350 loss: 1.5350 2022/09/07 06:06:46 - mmengine - INFO - Epoch(train) [46][900/3757] lr: 1.0000e-02 eta: 9:03:38 time: 0.1597 data_time: 0.0105 memory: 7124 grad_norm: 5.4895 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8399 loss: 1.8399 2022/09/07 06:06:51 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:07:02 - mmengine - INFO - Epoch(train) [46][1000/3757] lr: 1.0000e-02 eta: 9:03:22 time: 0.1585 data_time: 0.0099 memory: 7124 grad_norm: 5.7299 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.7302 loss: 1.7302 2022/09/07 06:07:18 - mmengine - INFO - Epoch(train) [46][1100/3757] lr: 1.0000e-02 eta: 9:03:06 time: 0.1587 data_time: 0.0109 memory: 7124 grad_norm: 5.4817 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.5481 loss: 1.5481 2022/09/07 06:07:34 - mmengine - INFO - Epoch(train) [46][1200/3757] lr: 1.0000e-02 eta: 9:02:50 time: 0.1604 data_time: 0.0103 memory: 7124 grad_norm: 5.4988 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.8514 loss: 1.8514 2022/09/07 06:07:50 - mmengine - INFO - Epoch(train) [46][1300/3757] lr: 1.0000e-02 eta: 9:02:34 time: 0.1671 data_time: 0.0109 memory: 7124 grad_norm: 5.4710 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.7538 loss: 1.7538 2022/09/07 06:08:05 - mmengine - INFO - Epoch(train) [46][1400/3757] lr: 1.0000e-02 eta: 9:02:19 time: 0.1543 data_time: 0.0090 memory: 7124 grad_norm: 5.4813 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.8492 loss: 1.8492 2022/09/07 06:08:21 - mmengine - INFO - Epoch(train) [46][1500/3757] lr: 1.0000e-02 eta: 9:02:03 time: 0.1618 data_time: 0.0105 memory: 7124 grad_norm: 5.5612 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7154 loss: 1.7154 2022/09/07 06:08:38 - mmengine - INFO - Epoch(train) [46][1600/3757] lr: 1.0000e-02 eta: 9:01:47 time: 0.1602 data_time: 0.0113 memory: 7124 grad_norm: 5.4240 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.5546 loss: 1.5546 2022/09/07 06:08:54 - mmengine - INFO - Epoch(train) [46][1700/3757] lr: 1.0000e-02 eta: 9:01:31 time: 0.1621 data_time: 0.0096 memory: 7124 grad_norm: 5.5422 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7668 loss: 1.7668 2022/09/07 06:09:10 - mmengine - INFO - Epoch(train) [46][1800/3757] lr: 1.0000e-02 eta: 9:01:16 time: 0.1572 data_time: 0.0091 memory: 7124 grad_norm: 5.3980 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.7176 loss: 1.7176 2022/09/07 06:09:25 - mmengine - INFO - Epoch(train) [46][1900/3757] lr: 1.0000e-02 eta: 9:01:00 time: 0.1605 data_time: 0.0098 memory: 7124 grad_norm: 5.4939 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.7101 loss: 1.7101 2022/09/07 06:09:31 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:09:42 - mmengine - INFO - Epoch(train) [46][2000/3757] lr: 1.0000e-02 eta: 9:00:44 time: 0.1575 data_time: 0.0110 memory: 7124 grad_norm: 5.6638 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6813 loss: 1.6813 2022/09/07 06:09:58 - mmengine - INFO - Epoch(train) [46][2100/3757] lr: 1.0000e-02 eta: 9:00:28 time: 0.1609 data_time: 0.0102 memory: 7124 grad_norm: 5.7001 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8402 loss: 1.8402 2022/09/07 06:10:14 - mmengine - INFO - Epoch(train) [46][2200/3757] lr: 1.0000e-02 eta: 9:00:13 time: 0.1690 data_time: 0.0090 memory: 7124 grad_norm: 5.6700 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8709 loss: 1.8709 2022/09/07 06:10:30 - mmengine - INFO - Epoch(train) [46][2300/3757] lr: 1.0000e-02 eta: 8:59:57 time: 0.1576 data_time: 0.0107 memory: 7124 grad_norm: 5.6533 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7466 loss: 1.7466 2022/09/07 06:10:46 - mmengine - INFO - Epoch(train) [46][2400/3757] lr: 1.0000e-02 eta: 8:59:41 time: 0.1587 data_time: 0.0113 memory: 7124 grad_norm: 5.5809 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9144 loss: 1.9144 2022/09/07 06:11:01 - mmengine - INFO - Epoch(train) [46][2500/3757] lr: 1.0000e-02 eta: 8:59:25 time: 0.1587 data_time: 0.0091 memory: 7124 grad_norm: 5.5450 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6825 loss: 1.6825 2022/09/07 06:11:18 - mmengine - INFO - Epoch(train) [46][2600/3757] lr: 1.0000e-02 eta: 8:59:10 time: 0.1588 data_time: 0.0099 memory: 7124 grad_norm: 5.5532 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6469 loss: 1.6469 2022/09/07 06:11:33 - mmengine - INFO - Epoch(train) [46][2700/3757] lr: 1.0000e-02 eta: 8:58:54 time: 0.1624 data_time: 0.0104 memory: 7124 grad_norm: 5.2572 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6072 loss: 1.6072 2022/09/07 06:11:49 - mmengine - INFO - Epoch(train) [46][2800/3757] lr: 1.0000e-02 eta: 8:58:38 time: 0.1588 data_time: 0.0116 memory: 7124 grad_norm: 5.3220 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.4526 loss: 1.4526 2022/09/07 06:12:05 - mmengine - INFO - Epoch(train) [46][2900/3757] lr: 1.0000e-02 eta: 8:58:22 time: 0.1575 data_time: 0.0097 memory: 7124 grad_norm: 5.2378 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9035 loss: 1.9035 2022/09/07 06:12:11 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:12:21 - mmengine - INFO - Epoch(train) [46][3000/3757] lr: 1.0000e-02 eta: 8:58:06 time: 0.1595 data_time: 0.0118 memory: 7124 grad_norm: 5.5203 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.6736 loss: 1.6736 2022/09/07 06:12:38 - mmengine - INFO - Epoch(train) [46][3100/3757] lr: 1.0000e-02 eta: 8:57:51 time: 0.1588 data_time: 0.0096 memory: 7124 grad_norm: 5.6300 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 2.2161 loss: 2.2161 2022/09/07 06:12:54 - mmengine - INFO - Epoch(train) [46][3200/3757] lr: 1.0000e-02 eta: 8:57:35 time: 0.1612 data_time: 0.0098 memory: 7124 grad_norm: 5.5883 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.8947 loss: 1.8947 2022/09/07 06:13:09 - mmengine - INFO - Epoch(train) [46][3300/3757] lr: 1.0000e-02 eta: 8:57:19 time: 0.1596 data_time: 0.0110 memory: 7124 grad_norm: 5.2930 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.1002 loss: 2.1002 2022/09/07 06:13:25 - mmengine - INFO - Epoch(train) [46][3400/3757] lr: 1.0000e-02 eta: 8:57:03 time: 0.1591 data_time: 0.0124 memory: 7124 grad_norm: 5.2044 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5291 loss: 1.5291 2022/09/07 06:13:41 - mmengine - INFO - Epoch(train) [46][3500/3757] lr: 1.0000e-02 eta: 8:56:47 time: 0.1552 data_time: 0.0085 memory: 7124 grad_norm: 5.7060 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6534 loss: 1.6534 2022/09/07 06:13:57 - mmengine - INFO - Epoch(train) [46][3600/3757] lr: 1.0000e-02 eta: 8:56:32 time: 0.1580 data_time: 0.0099 memory: 7124 grad_norm: 5.3213 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.8281 loss: 1.8281 2022/09/07 06:14:13 - mmengine - INFO - Epoch(train) [46][3700/3757] lr: 1.0000e-02 eta: 8:56:16 time: 0.1550 data_time: 0.0103 memory: 7124 grad_norm: 5.4561 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.6957 loss: 1.6957 2022/09/07 06:14:22 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:14:22 - mmengine - INFO - Epoch(train) [46][3757/3757] lr: 1.0000e-02 eta: 8:56:10 time: 0.1352 data_time: 0.0069 memory: 7124 grad_norm: 5.5034 top1_acc: 0.5714 top5_acc: 0.8571 loss_cls: 1.8171 loss: 1.8171 2022/09/07 06:14:22 - mmengine - INFO - Saving checkpoint at 46 epochs 2022/09/07 06:16:42 - mmengine - INFO - Epoch(val) [46][100/310] eta: 0:04:39 time: 1.3296 data_time: 1.0279 memory: 7627 2022/09/07 06:18:55 - mmengine - INFO - Epoch(val) [46][200/310] eta: 0:01:59 time: 1.0878 data_time: 0.7852 memory: 7627 2022/09/07 06:21:02 - mmengine - INFO - Epoch(val) [46][300/310] eta: 0:00:12 time: 1.2563 data_time: 0.9592 memory: 7627 2022/09/07 06:21:20 - mmengine - INFO - Epoch(val) [46][310/310] acc/top1: 0.6648 acc/top5: 0.8704 acc/mean1: 0.6648 2022/09/07 06:21:20 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_44.pth is removed 2022/09/07 06:21:22 - mmengine - INFO - The best checkpoint with 0.6648 acc/top1 at 46 epoch is saved to best_acc/top1_epoch_46.pth. 2022/09/07 06:21:39 - mmengine - INFO - Epoch(train) [47][100/3757] lr: 1.0000e-02 eta: 8:55:49 time: 0.1586 data_time: 0.0099 memory: 7627 grad_norm: 5.5412 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.6885 loss: 1.6885 2022/09/07 06:21:51 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:21:55 - mmengine - INFO - Epoch(train) [47][200/3757] lr: 1.0000e-02 eta: 8:55:33 time: 0.1627 data_time: 0.0100 memory: 7124 grad_norm: 5.5849 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5547 loss: 1.5547 2022/09/07 06:22:11 - mmengine - INFO - Epoch(train) [47][300/3757] lr: 1.0000e-02 eta: 8:55:17 time: 0.1601 data_time: 0.0100 memory: 7124 grad_norm: 5.5343 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.6974 loss: 1.6974 2022/09/07 06:22:27 - mmengine - INFO - Epoch(train) [47][400/3757] lr: 1.0000e-02 eta: 8:55:01 time: 0.1550 data_time: 0.0099 memory: 7124 grad_norm: 5.5832 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.9115 loss: 1.9115 2022/09/07 06:22:43 - mmengine - INFO - Epoch(train) [47][500/3757] lr: 1.0000e-02 eta: 8:54:45 time: 0.1594 data_time: 0.0103 memory: 7124 grad_norm: 5.5277 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.9753 loss: 1.9753 2022/09/07 06:22:58 - mmengine - INFO - Epoch(train) [47][600/3757] lr: 1.0000e-02 eta: 8:54:29 time: 0.1583 data_time: 0.0099 memory: 7124 grad_norm: 5.6342 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.5691 loss: 1.5691 2022/09/07 06:23:14 - mmengine - INFO - Epoch(train) [47][700/3757] lr: 1.0000e-02 eta: 8:54:13 time: 0.1589 data_time: 0.0098 memory: 7124 grad_norm: 5.5367 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.6899 loss: 1.6899 2022/09/07 06:23:30 - mmengine - INFO - Epoch(train) [47][800/3757] lr: 1.0000e-02 eta: 8:53:58 time: 0.1563 data_time: 0.0098 memory: 7124 grad_norm: 5.4807 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5006 loss: 1.5006 2022/09/07 06:23:46 - mmengine - INFO - Epoch(train) [47][900/3757] lr: 1.0000e-02 eta: 8:53:42 time: 0.1572 data_time: 0.0111 memory: 7124 grad_norm: 5.5824 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.6114 loss: 1.6114 2022/09/07 06:24:02 - mmengine - INFO - Epoch(train) [47][1000/3757] lr: 1.0000e-02 eta: 8:53:26 time: 0.1564 data_time: 0.0097 memory: 7124 grad_norm: 5.6366 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.7952 loss: 1.7952 2022/09/07 06:24:18 - mmengine - INFO - Epoch(train) [47][1100/3757] lr: 1.0000e-02 eta: 8:53:10 time: 0.1576 data_time: 0.0099 memory: 7124 grad_norm: 5.3804 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.6909 loss: 1.6909 2022/09/07 06:24:31 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:24:34 - mmengine - INFO - Epoch(train) [47][1200/3757] lr: 1.0000e-02 eta: 8:52:54 time: 0.1590 data_time: 0.0110 memory: 7124 grad_norm: 5.2801 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.5963 loss: 1.5963 2022/09/07 06:24:50 - mmengine - INFO - Epoch(train) [47][1300/3757] lr: 1.0000e-02 eta: 8:52:39 time: 0.1588 data_time: 0.0098 memory: 7124 grad_norm: 5.3843 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9138 loss: 1.9138 2022/09/07 06:25:06 - mmengine - INFO - Epoch(train) [47][1400/3757] lr: 1.0000e-02 eta: 8:52:23 time: 0.1602 data_time: 0.0108 memory: 7124 grad_norm: 5.3299 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.8830 loss: 1.8830 2022/09/07 06:25:22 - mmengine - INFO - Epoch(train) [47][1500/3757] lr: 1.0000e-02 eta: 8:52:07 time: 0.1562 data_time: 0.0099 memory: 7124 grad_norm: 5.4663 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.7257 loss: 1.7257 2022/09/07 06:25:38 - mmengine - INFO - Epoch(train) [47][1600/3757] lr: 1.0000e-02 eta: 8:51:51 time: 0.1568 data_time: 0.0100 memory: 7124 grad_norm: 5.4216 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.7902 loss: 1.7902 2022/09/07 06:25:54 - mmengine - INFO - Epoch(train) [47][1700/3757] lr: 1.0000e-02 eta: 8:51:36 time: 0.1713 data_time: 0.0096 memory: 7124 grad_norm: 5.4592 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.5438 loss: 1.5438 2022/09/07 06:26:10 - mmengine - INFO - Epoch(train) [47][1800/3757] lr: 1.0000e-02 eta: 8:51:20 time: 0.1558 data_time: 0.0092 memory: 7124 grad_norm: 5.3786 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.8339 loss: 1.8339 2022/09/07 06:26:26 - mmengine - INFO - Epoch(train) [47][1900/3757] lr: 1.0000e-02 eta: 8:51:04 time: 0.1607 data_time: 0.0107 memory: 7124 grad_norm: 5.2786 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7464 loss: 1.7464 2022/09/07 06:26:42 - mmengine - INFO - Epoch(train) [47][2000/3757] lr: 1.0000e-02 eta: 8:50:48 time: 0.1572 data_time: 0.0109 memory: 7124 grad_norm: 5.4161 top1_acc: 0.3750 top5_acc: 0.3750 loss_cls: 1.7422 loss: 1.7422 2022/09/07 06:26:58 - mmengine - INFO - Epoch(train) [47][2100/3757] lr: 1.0000e-02 eta: 8:50:33 time: 0.1570 data_time: 0.0100 memory: 7124 grad_norm: 5.4732 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7534 loss: 1.7534 2022/09/07 06:27:11 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:27:14 - mmengine - INFO - Epoch(train) [47][2200/3757] lr: 1.0000e-02 eta: 8:50:17 time: 0.1606 data_time: 0.0100 memory: 7124 grad_norm: 5.2729 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.8000 loss: 1.8000 2022/09/07 06:27:30 - mmengine - INFO - Epoch(train) [47][2300/3757] lr: 1.0000e-02 eta: 8:50:01 time: 0.1597 data_time: 0.0106 memory: 7124 grad_norm: 5.5442 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.5570 loss: 1.5570 2022/09/07 06:27:46 - mmengine - INFO - Epoch(train) [47][2400/3757] lr: 1.0000e-02 eta: 8:49:45 time: 0.1593 data_time: 0.0110 memory: 7124 grad_norm: 5.5116 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6786 loss: 1.6786 2022/09/07 06:28:02 - mmengine - INFO - Epoch(train) [47][2500/3757] lr: 1.0000e-02 eta: 8:49:30 time: 0.1578 data_time: 0.0103 memory: 7124 grad_norm: 5.5397 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7097 loss: 1.7097 2022/09/07 06:28:18 - mmengine - INFO - Epoch(train) [47][2600/3757] lr: 1.0000e-02 eta: 8:49:14 time: 0.1605 data_time: 0.0104 memory: 7124 grad_norm: 5.3121 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6362 loss: 1.6362 2022/09/07 06:28:34 - mmengine - INFO - Epoch(train) [47][2700/3757] lr: 1.0000e-02 eta: 8:48:58 time: 0.1573 data_time: 0.0117 memory: 7124 grad_norm: 5.6478 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8086 loss: 1.8086 2022/09/07 06:28:50 - mmengine - INFO - Epoch(train) [47][2800/3757] lr: 1.0000e-02 eta: 8:48:42 time: 0.1578 data_time: 0.0117 memory: 7124 grad_norm: 5.3088 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.8499 loss: 1.8499 2022/09/07 06:29:06 - mmengine - INFO - Epoch(train) [47][2900/3757] lr: 1.0000e-02 eta: 8:48:27 time: 0.1586 data_time: 0.0092 memory: 7124 grad_norm: 5.6331 top1_acc: 0.1250 top5_acc: 0.6250 loss_cls: 1.9398 loss: 1.9398 2022/09/07 06:29:22 - mmengine - INFO - Epoch(train) [47][3000/3757] lr: 1.0000e-02 eta: 8:48:11 time: 0.1589 data_time: 0.0102 memory: 7124 grad_norm: 5.6181 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.6307 loss: 1.6307 2022/09/07 06:29:38 - mmengine - INFO - Epoch(train) [47][3100/3757] lr: 1.0000e-02 eta: 8:47:55 time: 0.1559 data_time: 0.0099 memory: 7124 grad_norm: 5.5427 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7965 loss: 1.7965 2022/09/07 06:29:51 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:29:54 - mmengine - INFO - Epoch(train) [47][3200/3757] lr: 1.0000e-02 eta: 8:47:39 time: 0.1629 data_time: 0.0099 memory: 7124 grad_norm: 5.3102 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.5694 loss: 1.5694 2022/09/07 06:30:10 - mmengine - INFO - Epoch(train) [47][3300/3757] lr: 1.0000e-02 eta: 8:47:24 time: 0.1571 data_time: 0.0105 memory: 7124 grad_norm: 5.4712 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.5420 loss: 1.5420 2022/09/07 06:30:26 - mmengine - INFO - Epoch(train) [47][3400/3757] lr: 1.0000e-02 eta: 8:47:08 time: 0.1594 data_time: 0.0104 memory: 7124 grad_norm: 5.4244 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6047 loss: 1.6047 2022/09/07 06:30:42 - mmengine - INFO - Epoch(train) [47][3500/3757] lr: 1.0000e-02 eta: 8:46:52 time: 0.1572 data_time: 0.0102 memory: 7124 grad_norm: 5.5156 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6644 loss: 1.6644 2022/09/07 06:30:59 - mmengine - INFO - Epoch(train) [47][3600/3757] lr: 1.0000e-02 eta: 8:46:36 time: 0.1602 data_time: 0.0101 memory: 7124 grad_norm: 5.4295 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.5734 loss: 1.5734 2022/09/07 06:31:15 - mmengine - INFO - Epoch(train) [47][3700/3757] lr: 1.0000e-02 eta: 8:46:21 time: 0.1607 data_time: 0.0101 memory: 7124 grad_norm: 5.6477 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 2.0060 loss: 2.0060 2022/09/07 06:31:23 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:31:23 - mmengine - INFO - Epoch(train) [47][3757/3757] lr: 1.0000e-02 eta: 8:46:14 time: 0.1369 data_time: 0.0072 memory: 7124 grad_norm: 5.5034 top1_acc: 0.4286 top5_acc: 0.5714 loss_cls: 2.1430 loss: 2.1430 2022/09/07 06:31:23 - mmengine - INFO - Saving checkpoint at 47 epochs 2022/09/07 06:33:42 - mmengine - INFO - Epoch(val) [47][100/310] eta: 0:04:05 time: 1.1690 data_time: 0.8644 memory: 7627 2022/09/07 06:35:57 - mmengine - INFO - Epoch(val) [47][200/310] eta: 0:02:17 time: 1.2469 data_time: 0.9467 memory: 7627 2022/09/07 06:38:02 - mmengine - INFO - Epoch(val) [47][300/310] eta: 0:00:11 time: 1.1774 data_time: 0.8722 memory: 7627 2022/09/07 06:38:22 - mmengine - INFO - Epoch(val) [47][310/310] acc/top1: 0.6536 acc/top5: 0.8703 acc/mean1: 0.6535 2022/09/07 06:38:40 - mmengine - INFO - Epoch(train) [48][100/3757] lr: 1.0000e-02 eta: 8:45:55 time: 0.1587 data_time: 0.0104 memory: 7627 grad_norm: 5.4224 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5885 loss: 1.5885 2022/09/07 06:38:56 - mmengine - INFO - Epoch(train) [48][200/3757] lr: 1.0000e-02 eta: 8:45:39 time: 0.1649 data_time: 0.0114 memory: 7124 grad_norm: 5.5549 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.9536 loss: 1.9536 2022/09/07 06:39:12 - mmengine - INFO - Epoch(train) [48][300/3757] lr: 1.0000e-02 eta: 8:45:23 time: 0.1561 data_time: 0.0105 memory: 7124 grad_norm: 5.3052 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.6017 loss: 1.6017 2022/09/07 06:39:28 - mmengine - INFO - Epoch(train) [48][400/3757] lr: 1.0000e-02 eta: 8:45:08 time: 0.1546 data_time: 0.0102 memory: 7124 grad_norm: 5.4366 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.5638 loss: 1.5638 2022/09/07 06:39:32 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:39:44 - mmengine - INFO - Epoch(train) [48][500/3757] lr: 1.0000e-02 eta: 8:44:52 time: 0.1610 data_time: 0.0091 memory: 7124 grad_norm: 5.3952 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.9204 loss: 1.9204 2022/09/07 06:40:00 - mmengine - INFO - Epoch(train) [48][600/3757] lr: 1.0000e-02 eta: 8:44:36 time: 0.1553 data_time: 0.0095 memory: 7124 grad_norm: 5.4998 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.3933 loss: 1.3933 2022/09/07 06:40:16 - mmengine - INFO - Epoch(train) [48][700/3757] lr: 1.0000e-02 eta: 8:44:21 time: 0.1588 data_time: 0.0091 memory: 7124 grad_norm: 5.5658 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7370 loss: 1.7370 2022/09/07 06:40:32 - mmengine - INFO - Epoch(train) [48][800/3757] lr: 1.0000e-02 eta: 8:44:05 time: 0.1585 data_time: 0.0109 memory: 7124 grad_norm: 5.3225 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.5335 loss: 1.5335 2022/09/07 06:40:49 - mmengine - INFO - Epoch(train) [48][900/3757] lr: 1.0000e-02 eta: 8:43:49 time: 0.1609 data_time: 0.0096 memory: 7124 grad_norm: 5.6668 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 2.0144 loss: 2.0144 2022/09/07 06:41:04 - mmengine - INFO - Epoch(train) [48][1000/3757] lr: 1.0000e-02 eta: 8:43:33 time: 0.1612 data_time: 0.0109 memory: 7124 grad_norm: 5.4677 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.6590 loss: 1.6590 2022/09/07 06:41:21 - mmengine - INFO - Epoch(train) [48][1100/3757] lr: 1.0000e-02 eta: 8:43:18 time: 0.1573 data_time: 0.0101 memory: 7124 grad_norm: 5.6676 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 2.1255 loss: 2.1255 2022/09/07 06:41:37 - mmengine - INFO - Epoch(train) [48][1200/3757] lr: 1.0000e-02 eta: 8:43:02 time: 0.1585 data_time: 0.0111 memory: 7124 grad_norm: 5.6429 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7014 loss: 1.7014 2022/09/07 06:41:53 - mmengine - INFO - Epoch(train) [48][1300/3757] lr: 1.0000e-02 eta: 8:42:46 time: 0.1560 data_time: 0.0101 memory: 7124 grad_norm: 5.4319 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5179 loss: 1.5179 2022/09/07 06:42:09 - mmengine - INFO - Epoch(train) [48][1400/3757] lr: 1.0000e-02 eta: 8:42:30 time: 0.1561 data_time: 0.0109 memory: 7124 grad_norm: 5.2497 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.4402 loss: 1.4402 2022/09/07 06:42:12 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:42:24 - mmengine - INFO - Epoch(train) [48][1500/3757] lr: 1.0000e-02 eta: 8:42:14 time: 0.1567 data_time: 0.0092 memory: 7124 grad_norm: 5.4220 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.7153 loss: 1.7153 2022/09/07 06:42:41 - mmengine - INFO - Epoch(train) [48][1600/3757] lr: 1.0000e-02 eta: 8:41:59 time: 0.1594 data_time: 0.0093 memory: 7124 grad_norm: 5.3894 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.5790 loss: 1.5790 2022/09/07 06:42:56 - mmengine - INFO - Epoch(train) [48][1700/3757] lr: 1.0000e-02 eta: 8:41:43 time: 0.1605 data_time: 0.0093 memory: 7124 grad_norm: 5.3585 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.9355 loss: 1.9355 2022/09/07 06:43:12 - mmengine - INFO - Epoch(train) [48][1800/3757] lr: 1.0000e-02 eta: 8:41:27 time: 0.1579 data_time: 0.0098 memory: 7124 grad_norm: 5.5322 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7821 loss: 1.7821 2022/09/07 06:43:29 - mmengine - INFO - Epoch(train) [48][1900/3757] lr: 1.0000e-02 eta: 8:41:11 time: 0.1571 data_time: 0.0112 memory: 7124 grad_norm: 5.5364 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8062 loss: 1.8062 2022/09/07 06:43:44 - mmengine - INFO - Epoch(train) [48][2000/3757] lr: 1.0000e-02 eta: 8:40:56 time: 0.1546 data_time: 0.0109 memory: 7124 grad_norm: 5.4181 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.7869 loss: 1.7869 2022/09/07 06:44:00 - mmengine - INFO - Epoch(train) [48][2100/3757] lr: 1.0000e-02 eta: 8:40:40 time: 0.1549 data_time: 0.0098 memory: 7124 grad_norm: 5.5218 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8940 loss: 1.8940 2022/09/07 06:44:16 - mmengine - INFO - Epoch(train) [48][2200/3757] lr: 1.0000e-02 eta: 8:40:24 time: 0.1559 data_time: 0.0098 memory: 7124 grad_norm: 5.3481 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7340 loss: 1.7340 2022/09/07 06:44:32 - mmengine - INFO - Epoch(train) [48][2300/3757] lr: 1.0000e-02 eta: 8:40:08 time: 0.1572 data_time: 0.0100 memory: 7124 grad_norm: 5.5570 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2879 loss: 1.2879 2022/09/07 06:44:48 - mmengine - INFO - Epoch(train) [48][2400/3757] lr: 1.0000e-02 eta: 8:39:52 time: 0.1591 data_time: 0.0096 memory: 7124 grad_norm: 5.4957 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8829 loss: 1.8829 2022/09/07 06:44:52 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:45:04 - mmengine - INFO - Epoch(train) [48][2500/3757] lr: 1.0000e-02 eta: 8:39:36 time: 0.1570 data_time: 0.0098 memory: 7124 grad_norm: 5.5952 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7778 loss: 1.7778 2022/09/07 06:45:20 - mmengine - INFO - Epoch(train) [48][2600/3757] lr: 1.0000e-02 eta: 8:39:21 time: 0.1573 data_time: 0.0101 memory: 7124 grad_norm: 5.4099 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.4716 loss: 1.4716 2022/09/07 06:45:36 - mmengine - INFO - Epoch(train) [48][2700/3757] lr: 1.0000e-02 eta: 8:39:05 time: 0.1652 data_time: 0.0109 memory: 7124 grad_norm: 5.6961 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6930 loss: 1.6930 2022/09/07 06:45:52 - mmengine - INFO - Epoch(train) [48][2800/3757] lr: 1.0000e-02 eta: 8:38:49 time: 0.1622 data_time: 0.0109 memory: 7124 grad_norm: 5.3585 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7237 loss: 1.7237 2022/09/07 06:46:08 - mmengine - INFO - Epoch(train) [48][2900/3757] lr: 1.0000e-02 eta: 8:38:33 time: 0.1551 data_time: 0.0110 memory: 7124 grad_norm: 5.4824 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6822 loss: 1.6822 2022/09/07 06:46:24 - mmengine - INFO - Epoch(train) [48][3000/3757] lr: 1.0000e-02 eta: 8:38:18 time: 0.1579 data_time: 0.0098 memory: 7124 grad_norm: 5.4398 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8286 loss: 1.8286 2022/09/07 06:46:40 - mmengine - INFO - Epoch(train) [48][3100/3757] lr: 1.0000e-02 eta: 8:38:02 time: 0.1562 data_time: 0.0093 memory: 7124 grad_norm: 5.3270 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.7313 loss: 1.7313 2022/09/07 06:46:56 - mmengine - INFO - Epoch(train) [48][3200/3757] lr: 1.0000e-02 eta: 8:37:46 time: 0.1566 data_time: 0.0090 memory: 7124 grad_norm: 5.3601 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7659 loss: 1.7659 2022/09/07 06:47:12 - mmengine - INFO - Epoch(train) [48][3300/3757] lr: 1.0000e-02 eta: 8:37:30 time: 0.1590 data_time: 0.0095 memory: 7124 grad_norm: 5.3670 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4818 loss: 1.4818 2022/09/07 06:47:28 - mmengine - INFO - Epoch(train) [48][3400/3757] lr: 1.0000e-02 eta: 8:37:14 time: 0.1633 data_time: 0.0097 memory: 7124 grad_norm: 5.3630 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.9251 loss: 1.9251 2022/09/07 06:47:31 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:47:44 - mmengine - INFO - Epoch(train) [48][3500/3757] lr: 1.0000e-02 eta: 8:36:59 time: 0.1569 data_time: 0.0109 memory: 7124 grad_norm: 5.3851 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 1.6262 loss: 1.6262 2022/09/07 06:48:00 - mmengine - INFO - Epoch(train) [48][3600/3757] lr: 1.0000e-02 eta: 8:36:43 time: 0.1628 data_time: 0.0088 memory: 7124 grad_norm: 5.8155 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8386 loss: 1.8386 2022/09/07 06:48:16 - mmengine - INFO - Epoch(train) [48][3700/3757] lr: 1.0000e-02 eta: 8:36:27 time: 0.1565 data_time: 0.0115 memory: 7124 grad_norm: 5.5955 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.9724 loss: 1.9724 2022/09/07 06:48:25 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:48:25 - mmengine - INFO - Epoch(train) [48][3757/3757] lr: 1.0000e-02 eta: 8:36:21 time: 0.1376 data_time: 0.0076 memory: 7124 grad_norm: 5.4326 top1_acc: 0.7143 top5_acc: 1.0000 loss_cls: 1.6176 loss: 1.6176 2022/09/07 06:48:25 - mmengine - INFO - Saving checkpoint at 48 epochs 2022/09/07 06:50:45 - mmengine - INFO - Epoch(val) [48][100/310] eta: 0:04:15 time: 1.2143 data_time: 0.9118 memory: 7627 2022/09/07 06:53:00 - mmengine - INFO - Epoch(val) [48][200/310] eta: 0:02:13 time: 1.2110 data_time: 0.9056 memory: 7627 2022/09/07 06:55:04 - mmengine - INFO - Epoch(val) [48][300/310] eta: 0:00:11 time: 1.1995 data_time: 0.8900 memory: 7627 2022/09/07 06:55:22 - mmengine - INFO - Epoch(val) [48][310/310] acc/top1: 0.6591 acc/top5: 0.8727 acc/mean1: 0.6589 2022/09/07 06:55:40 - mmengine - INFO - Epoch(train) [49][100/3757] lr: 1.0000e-02 eta: 8:36:02 time: 0.1568 data_time: 0.0111 memory: 7627 grad_norm: 5.3794 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.5411 loss: 1.5411 2022/09/07 06:55:56 - mmengine - INFO - Epoch(train) [49][200/3757] lr: 1.0000e-02 eta: 8:35:46 time: 0.1570 data_time: 0.0110 memory: 7124 grad_norm: 5.3618 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6889 loss: 1.6889 2022/09/07 06:56:12 - mmengine - INFO - Epoch(train) [49][300/3757] lr: 1.0000e-02 eta: 8:35:30 time: 0.1571 data_time: 0.0123 memory: 7124 grad_norm: 5.6689 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.9201 loss: 1.9201 2022/09/07 06:56:28 - mmengine - INFO - Epoch(train) [49][400/3757] lr: 1.0000e-02 eta: 8:35:14 time: 0.1568 data_time: 0.0113 memory: 7124 grad_norm: 5.1234 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.9209 loss: 1.9209 2022/09/07 06:56:44 - mmengine - INFO - Epoch(train) [49][500/3757] lr: 1.0000e-02 eta: 8:34:58 time: 0.1575 data_time: 0.0103 memory: 7124 grad_norm: 5.6238 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.7362 loss: 1.7362 2022/09/07 06:57:00 - mmengine - INFO - Epoch(train) [49][600/3757] lr: 1.0000e-02 eta: 8:34:42 time: 0.1570 data_time: 0.0109 memory: 7124 grad_norm: 5.2936 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.7288 loss: 1.7288 2022/09/07 06:57:10 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:57:16 - mmengine - INFO - Epoch(train) [49][700/3757] lr: 1.0000e-02 eta: 8:34:27 time: 0.1579 data_time: 0.0111 memory: 7124 grad_norm: 5.4054 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.4563 loss: 1.4563 2022/09/07 06:57:32 - mmengine - INFO - Epoch(train) [49][800/3757] lr: 1.0000e-02 eta: 8:34:11 time: 0.1575 data_time: 0.0109 memory: 7124 grad_norm: 5.4892 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7578 loss: 1.7578 2022/09/07 06:57:48 - mmengine - INFO - Epoch(train) [49][900/3757] lr: 1.0000e-02 eta: 8:33:55 time: 0.1599 data_time: 0.0110 memory: 7124 grad_norm: 5.3687 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.6283 loss: 1.6283 2022/09/07 06:58:04 - mmengine - INFO - Epoch(train) [49][1000/3757] lr: 1.0000e-02 eta: 8:33:39 time: 0.1590 data_time: 0.0108 memory: 7124 grad_norm: 5.7245 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8144 loss: 1.8144 2022/09/07 06:58:20 - mmengine - INFO - Epoch(train) [49][1100/3757] lr: 1.0000e-02 eta: 8:33:24 time: 0.1539 data_time: 0.0102 memory: 7124 grad_norm: 5.3839 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7260 loss: 1.7260 2022/09/07 06:58:36 - mmengine - INFO - Epoch(train) [49][1200/3757] lr: 1.0000e-02 eta: 8:33:08 time: 0.1601 data_time: 0.0116 memory: 7124 grad_norm: 5.4632 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 2.0388 loss: 2.0388 2022/09/07 06:58:52 - mmengine - INFO - Epoch(train) [49][1300/3757] lr: 1.0000e-02 eta: 8:32:52 time: 0.1573 data_time: 0.0111 memory: 7124 grad_norm: 5.4869 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6744 loss: 1.6744 2022/09/07 06:59:08 - mmengine - INFO - Epoch(train) [49][1400/3757] lr: 1.0000e-02 eta: 8:32:36 time: 0.1627 data_time: 0.0107 memory: 7124 grad_norm: 5.2806 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.9303 loss: 1.9303 2022/09/07 06:59:24 - mmengine - INFO - Epoch(train) [49][1500/3757] lr: 1.0000e-02 eta: 8:32:21 time: 0.1583 data_time: 0.0109 memory: 7124 grad_norm: 5.7201 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7630 loss: 1.7630 2022/09/07 06:59:40 - mmengine - INFO - Epoch(train) [49][1600/3757] lr: 1.0000e-02 eta: 8:32:05 time: 0.1573 data_time: 0.0103 memory: 7124 grad_norm: 5.6240 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.7202 loss: 1.7202 2022/09/07 06:59:50 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 06:59:56 - mmengine - INFO - Epoch(train) [49][1700/3757] lr: 1.0000e-02 eta: 8:31:49 time: 0.1558 data_time: 0.0106 memory: 7124 grad_norm: 5.2837 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7912 loss: 1.7912 2022/09/07 07:00:12 - mmengine - INFO - Epoch(train) [49][1800/3757] lr: 1.0000e-02 eta: 8:31:33 time: 0.1622 data_time: 0.0095 memory: 7124 grad_norm: 5.4098 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.9503 loss: 1.9503 2022/09/07 07:00:28 - mmengine - INFO - Epoch(train) [49][1900/3757] lr: 1.0000e-02 eta: 8:31:17 time: 0.1631 data_time: 0.0110 memory: 7124 grad_norm: 5.2859 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.7888 loss: 1.7888 2022/09/07 07:00:44 - mmengine - INFO - Epoch(train) [49][2000/3757] lr: 1.0000e-02 eta: 8:31:02 time: 0.1592 data_time: 0.0105 memory: 7124 grad_norm: 5.5713 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.5707 loss: 1.5707 2022/09/07 07:01:00 - mmengine - INFO - Epoch(train) [49][2100/3757] lr: 1.0000e-02 eta: 8:30:46 time: 0.1589 data_time: 0.0109 memory: 7124 grad_norm: 5.7281 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.8788 loss: 1.8788 2022/09/07 07:01:16 - mmengine - INFO - Epoch(train) [49][2200/3757] lr: 1.0000e-02 eta: 8:30:30 time: 0.1567 data_time: 0.0124 memory: 7124 grad_norm: 5.3800 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7685 loss: 1.7685 2022/09/07 07:01:32 - mmengine - INFO - Epoch(train) [49][2300/3757] lr: 1.0000e-02 eta: 8:30:14 time: 0.1557 data_time: 0.0108 memory: 7124 grad_norm: 5.4526 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6114 loss: 1.6114 2022/09/07 07:01:48 - mmengine - INFO - Epoch(train) [49][2400/3757] lr: 1.0000e-02 eta: 8:29:58 time: 0.1612 data_time: 0.0105 memory: 7124 grad_norm: 5.4881 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6004 loss: 1.6004 2022/09/07 07:02:04 - mmengine - INFO - Epoch(train) [49][2500/3757] lr: 1.0000e-02 eta: 8:29:42 time: 0.1563 data_time: 0.0103 memory: 7124 grad_norm: 5.3153 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.8536 loss: 1.8536 2022/09/07 07:02:20 - mmengine - INFO - Epoch(train) [49][2600/3757] lr: 1.0000e-02 eta: 8:29:26 time: 0.1604 data_time: 0.0112 memory: 7124 grad_norm: 5.4873 top1_acc: 0.0000 top5_acc: 0.7500 loss_cls: 2.0249 loss: 2.0249 2022/09/07 07:02:30 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:02:36 - mmengine - INFO - Epoch(train) [49][2700/3757] lr: 1.0000e-02 eta: 8:29:11 time: 0.1579 data_time: 0.0114 memory: 7124 grad_norm: 5.4533 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.8205 loss: 1.8205 2022/09/07 07:02:52 - mmengine - INFO - Epoch(train) [49][2800/3757] lr: 1.0000e-02 eta: 8:28:55 time: 0.1583 data_time: 0.0118 memory: 7124 grad_norm: 5.3884 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7262 loss: 1.7262 2022/09/07 07:03:08 - mmengine - INFO - Epoch(train) [49][2900/3757] lr: 1.0000e-02 eta: 8:28:39 time: 0.1600 data_time: 0.0114 memory: 7124 grad_norm: 5.6363 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.7075 loss: 1.7075 2022/09/07 07:03:24 - mmengine - INFO - Epoch(train) [49][3000/3757] lr: 1.0000e-02 eta: 8:28:24 time: 0.1614 data_time: 0.0102 memory: 7124 grad_norm: 5.4306 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.9964 loss: 1.9964 2022/09/07 07:03:40 - mmengine - INFO - Epoch(train) [49][3100/3757] lr: 1.0000e-02 eta: 8:28:08 time: 0.1549 data_time: 0.0102 memory: 7124 grad_norm: 5.5188 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.7760 loss: 1.7760 2022/09/07 07:03:56 - mmengine - INFO - Epoch(train) [49][3200/3757] lr: 1.0000e-02 eta: 8:27:52 time: 0.1592 data_time: 0.0105 memory: 7124 grad_norm: 5.6386 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.9429 loss: 1.9429 2022/09/07 07:04:12 - mmengine - INFO - Epoch(train) [49][3300/3757] lr: 1.0000e-02 eta: 8:27:36 time: 0.1582 data_time: 0.0105 memory: 7124 grad_norm: 5.3486 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7426 loss: 1.7426 2022/09/07 07:04:28 - mmengine - INFO - Epoch(train) [49][3400/3757] lr: 1.0000e-02 eta: 8:27:21 time: 0.1598 data_time: 0.0106 memory: 7124 grad_norm: 5.5623 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.5000 loss: 1.5000 2022/09/07 07:04:44 - mmengine - INFO - Epoch(train) [49][3500/3757] lr: 1.0000e-02 eta: 8:27:05 time: 0.1569 data_time: 0.0104 memory: 7124 grad_norm: 5.6784 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.7718 loss: 1.7718 2022/09/07 07:05:00 - mmengine - INFO - Epoch(train) [49][3600/3757] lr: 1.0000e-02 eta: 8:26:49 time: 0.1587 data_time: 0.0135 memory: 7124 grad_norm: 5.3352 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6841 loss: 1.6841 2022/09/07 07:05:10 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:05:16 - mmengine - INFO - Epoch(train) [49][3700/3757] lr: 1.0000e-02 eta: 8:26:33 time: 0.1573 data_time: 0.0115 memory: 7124 grad_norm: 5.8066 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.8256 loss: 1.8256 2022/09/07 07:05:25 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:05:25 - mmengine - INFO - Epoch(train) [49][3757/3757] lr: 1.0000e-02 eta: 8:26:27 time: 0.1362 data_time: 0.0069 memory: 7124 grad_norm: 5.4072 top1_acc: 0.8571 top5_acc: 0.8571 loss_cls: 1.5219 loss: 1.5219 2022/09/07 07:05:25 - mmengine - INFO - Saving checkpoint at 49 epochs 2022/09/07 07:07:42 - mmengine - INFO - Epoch(val) [49][100/310] eta: 0:03:59 time: 1.1419 data_time: 0.8433 memory: 7627 2022/09/07 07:10:00 - mmengine - INFO - Epoch(val) [49][200/310] eta: 0:02:28 time: 1.3512 data_time: 1.0489 memory: 7627 2022/09/07 07:12:03 - mmengine - INFO - Epoch(val) [49][300/310] eta: 0:00:11 time: 1.1524 data_time: 0.8529 memory: 7627 2022/09/07 07:12:21 - mmengine - INFO - Epoch(val) [49][310/310] acc/top1: 0.6605 acc/top5: 0.8702 acc/mean1: 0.6606 2022/09/07 07:12:39 - mmengine - INFO - Epoch(train) [50][100/3757] lr: 1.0000e-02 eta: 8:26:07 time: 0.1619 data_time: 0.0145 memory: 7627 grad_norm: 5.6407 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8842 loss: 1.8842 2022/09/07 07:12:55 - mmengine - INFO - Epoch(train) [50][200/3757] lr: 1.0000e-02 eta: 8:25:51 time: 0.1578 data_time: 0.0104 memory: 7124 grad_norm: 5.6006 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.8005 loss: 1.8005 2022/09/07 07:13:11 - mmengine - INFO - Epoch(train) [50][300/3757] lr: 1.0000e-02 eta: 8:25:35 time: 0.1601 data_time: 0.0110 memory: 7124 grad_norm: 5.4763 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 2.0591 loss: 2.0591 2022/09/07 07:13:27 - mmengine - INFO - Epoch(train) [50][400/3757] lr: 1.0000e-02 eta: 8:25:19 time: 0.1551 data_time: 0.0103 memory: 7124 grad_norm: 5.5126 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.7642 loss: 1.7642 2022/09/07 07:13:43 - mmengine - INFO - Epoch(train) [50][500/3757] lr: 1.0000e-02 eta: 8:25:04 time: 0.1560 data_time: 0.0100 memory: 7124 grad_norm: 5.4569 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.5046 loss: 1.5046 2022/09/07 07:13:59 - mmengine - INFO - Epoch(train) [50][600/3757] lr: 1.0000e-02 eta: 8:24:48 time: 0.1606 data_time: 0.0108 memory: 7124 grad_norm: 5.3640 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.5777 loss: 1.5777 2022/09/07 07:14:15 - mmengine - INFO - Epoch(train) [50][700/3757] lr: 1.0000e-02 eta: 8:24:32 time: 0.1582 data_time: 0.0104 memory: 7124 grad_norm: 5.3882 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4978 loss: 1.4978 2022/09/07 07:14:31 - mmengine - INFO - Epoch(train) [50][800/3757] lr: 1.0000e-02 eta: 8:24:16 time: 0.1619 data_time: 0.0107 memory: 7124 grad_norm: 5.1442 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7824 loss: 1.7824 2022/09/07 07:14:47 - mmengine - INFO - Epoch(train) [50][900/3757] lr: 1.0000e-02 eta: 8:24:01 time: 0.1585 data_time: 0.0110 memory: 7124 grad_norm: 5.6159 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.6814 loss: 1.6814 2022/09/07 07:14:48 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:15:03 - mmengine - INFO - Epoch(train) [50][1000/3757] lr: 1.0000e-02 eta: 8:23:45 time: 0.1629 data_time: 0.0180 memory: 7124 grad_norm: 5.4291 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.8055 loss: 1.8055 2022/09/07 07:15:19 - mmengine - INFO - Epoch(train) [50][1100/3757] lr: 1.0000e-02 eta: 8:23:29 time: 0.1605 data_time: 0.0114 memory: 7124 grad_norm: 5.6287 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.7941 loss: 1.7941 2022/09/07 07:15:35 - mmengine - INFO - Epoch(train) [50][1200/3757] lr: 1.0000e-02 eta: 8:23:13 time: 0.1612 data_time: 0.0118 memory: 7124 grad_norm: 5.5378 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.7064 loss: 1.7064 2022/09/07 07:15:51 - mmengine - INFO - Epoch(train) [50][1300/3757] lr: 1.0000e-02 eta: 8:22:57 time: 0.1603 data_time: 0.0115 memory: 7124 grad_norm: 5.6995 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.6708 loss: 1.6708 2022/09/07 07:16:07 - mmengine - INFO - Epoch(train) [50][1400/3757] lr: 1.0000e-02 eta: 8:22:42 time: 0.1594 data_time: 0.0114 memory: 7124 grad_norm: 5.2679 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7125 loss: 1.7125 2022/09/07 07:16:23 - mmengine - INFO - Epoch(train) [50][1500/3757] lr: 1.0000e-02 eta: 8:22:26 time: 0.1581 data_time: 0.0116 memory: 7124 grad_norm: 5.1469 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.6589 loss: 1.6589 2022/09/07 07:16:39 - mmengine - INFO - Epoch(train) [50][1600/3757] lr: 1.0000e-02 eta: 8:22:10 time: 0.1619 data_time: 0.0161 memory: 7124 grad_norm: 5.8656 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.7543 loss: 1.7543 2022/09/07 07:16:55 - mmengine - INFO - Epoch(train) [50][1700/3757] lr: 1.0000e-02 eta: 8:21:54 time: 0.1582 data_time: 0.0104 memory: 7124 grad_norm: 5.6404 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6665 loss: 1.6665 2022/09/07 07:17:11 - mmengine - INFO - Epoch(train) [50][1800/3757] lr: 1.0000e-02 eta: 8:21:38 time: 0.1576 data_time: 0.0125 memory: 7124 grad_norm: 5.6447 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.9692 loss: 1.9692 2022/09/07 07:17:27 - mmengine - INFO - Epoch(train) [50][1900/3757] lr: 1.0000e-02 eta: 8:21:22 time: 0.1551 data_time: 0.0103 memory: 7124 grad_norm: 5.4936 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.7990 loss: 1.7990 2022/09/07 07:17:28 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:17:43 - mmengine - INFO - Epoch(train) [50][2000/3757] lr: 1.0000e-02 eta: 8:21:07 time: 0.1592 data_time: 0.0129 memory: 7124 grad_norm: 5.4297 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.9888 loss: 1.9888 2022/09/07 07:17:59 - mmengine - INFO - Epoch(train) [50][2100/3757] lr: 1.0000e-02 eta: 8:20:51 time: 0.1603 data_time: 0.0108 memory: 7124 grad_norm: 5.5010 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.6541 loss: 1.6541 2022/09/07 07:18:15 - mmengine - INFO - Epoch(train) [50][2200/3757] lr: 1.0000e-02 eta: 8:20:35 time: 0.1569 data_time: 0.0118 memory: 7124 grad_norm: 5.5500 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6428 loss: 1.6428 2022/09/07 07:18:31 - mmengine - INFO - Epoch(train) [50][2300/3757] lr: 1.0000e-02 eta: 8:20:19 time: 0.1567 data_time: 0.0114 memory: 7124 grad_norm: 5.6661 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.9031 loss: 1.9031 2022/09/07 07:18:47 - mmengine - INFO - Epoch(train) [50][2400/3757] lr: 1.0000e-02 eta: 8:20:04 time: 0.1570 data_time: 0.0112 memory: 7124 grad_norm: 5.3566 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.7206 loss: 1.7206 2022/09/07 07:19:03 - mmengine - INFO - Epoch(train) [50][2500/3757] lr: 1.0000e-02 eta: 8:19:48 time: 0.1586 data_time: 0.0097 memory: 7124 grad_norm: 5.3507 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.7259 loss: 1.7259 2022/09/07 07:19:19 - mmengine - INFO - Epoch(train) [50][2600/3757] lr: 1.0000e-02 eta: 8:19:32 time: 0.1621 data_time: 0.0119 memory: 7124 grad_norm: 5.1375 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.6626 loss: 1.6626 2022/09/07 07:19:35 - mmengine - INFO - Epoch(train) [50][2700/3757] lr: 1.0000e-02 eta: 8:19:16 time: 0.1583 data_time: 0.0121 memory: 7124 grad_norm: 5.5967 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.5947 loss: 1.5947 2022/09/07 07:19:51 - mmengine - INFO - Epoch(train) [50][2800/3757] lr: 1.0000e-02 eta: 8:19:00 time: 0.1607 data_time: 0.0097 memory: 7124 grad_norm: 5.6287 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7938 loss: 1.7938 2022/09/07 07:20:07 - mmengine - INFO - Epoch(train) [50][2900/3757] lr: 1.0000e-02 eta: 8:18:45 time: 0.1600 data_time: 0.0111 memory: 7124 grad_norm: 5.5992 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.5788 loss: 1.5788 2022/09/07 07:20:08 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:20:23 - mmengine - INFO - Epoch(train) [50][3000/3757] lr: 1.0000e-02 eta: 8:18:29 time: 0.1564 data_time: 0.0122 memory: 7124 grad_norm: 5.3326 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.8405 loss: 1.8405 2022/09/07 07:20:39 - mmengine - INFO - Epoch(train) [50][3100/3757] lr: 1.0000e-02 eta: 8:18:13 time: 0.1599 data_time: 0.0107 memory: 7124 grad_norm: 5.5111 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 2.0041 loss: 2.0041 2022/09/07 07:20:55 - mmengine - INFO - Epoch(train) [50][3200/3757] lr: 1.0000e-02 eta: 8:17:57 time: 0.1578 data_time: 0.0115 memory: 7124 grad_norm: 5.4827 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.6978 loss: 1.6978 2022/09/07 07:21:11 - mmengine - INFO - Epoch(train) [50][3300/3757] lr: 1.0000e-02 eta: 8:17:42 time: 0.1612 data_time: 0.0112 memory: 7124 grad_norm: 5.6694 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.6410 loss: 1.6410 2022/09/07 07:21:27 - mmengine - INFO - Epoch(train) [50][3400/3757] lr: 1.0000e-02 eta: 8:17:26 time: 0.1559 data_time: 0.0107 memory: 7124 grad_norm: 5.4061 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.3984 loss: 1.3984 2022/09/07 07:21:43 - mmengine - INFO - Epoch(train) [50][3500/3757] lr: 1.0000e-02 eta: 8:17:10 time: 0.1568 data_time: 0.0107 memory: 7124 grad_norm: 5.5426 top1_acc: 0.3750 top5_acc: 1.0000 loss_cls: 1.5100 loss: 1.5100 2022/09/07 07:21:59 - mmengine - INFO - Epoch(train) [50][3600/3757] lr: 1.0000e-02 eta: 8:16:54 time: 0.1577 data_time: 0.0108 memory: 7124 grad_norm: 5.7460 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.9978 loss: 1.9978 2022/09/07 07:22:15 - mmengine - INFO - Epoch(train) [50][3700/3757] lr: 1.0000e-02 eta: 8:16:38 time: 0.1612 data_time: 0.0107 memory: 7124 grad_norm: 5.2395 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.7225 loss: 1.7225 2022/09/07 07:22:23 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:22:23 - mmengine - INFO - Epoch(train) [50][3757/3757] lr: 1.0000e-02 eta: 8:16:32 time: 0.1386 data_time: 0.0086 memory: 7124 grad_norm: 5.5492 top1_acc: 0.7143 top5_acc: 1.0000 loss_cls: 1.9440 loss: 1.9440 2022/09/07 07:22:23 - mmengine - INFO - Saving checkpoint at 50 epochs 2022/09/07 07:24:42 - mmengine - INFO - Epoch(val) [50][100/310] eta: 0:04:12 time: 1.2020 data_time: 0.8930 memory: 7627 2022/09/07 07:26:57 - mmengine - INFO - Epoch(val) [50][200/310] eta: 0:02:20 time: 1.2761 data_time: 0.9678 memory: 7627 2022/09/07 07:29:04 - mmengine - INFO - Epoch(val) [50][300/310] eta: 0:00:12 time: 1.2996 data_time: 0.9981 memory: 7627 2022/09/07 07:29:22 - mmengine - INFO - Epoch(val) [50][310/310] acc/top1: 0.6624 acc/top5: 0.8739 acc/mean1: 0.6623 2022/09/07 07:29:40 - mmengine - INFO - Epoch(train) [51][100/3757] lr: 1.0000e-03 eta: 8:16:13 time: 0.1577 data_time: 0.0110 memory: 7627 grad_norm: 5.1543 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3778 loss: 1.3778 2022/09/07 07:29:48 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:29:56 - mmengine - INFO - Epoch(train) [51][200/3757] lr: 1.0000e-03 eta: 8:15:57 time: 0.1593 data_time: 0.0103 memory: 7124 grad_norm: 5.0314 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2892 loss: 1.2892 2022/09/07 07:30:12 - mmengine - INFO - Epoch(train) [51][300/3757] lr: 1.0000e-03 eta: 8:15:41 time: 0.1565 data_time: 0.0094 memory: 7124 grad_norm: 5.2574 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.4965 loss: 1.4965 2022/09/07 07:30:28 - mmengine - INFO - Epoch(train) [51][400/3757] lr: 1.0000e-03 eta: 8:15:25 time: 0.1563 data_time: 0.0117 memory: 7124 grad_norm: 5.0818 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3662 loss: 1.3662 2022/09/07 07:30:44 - mmengine - INFO - Epoch(train) [51][500/3757] lr: 1.0000e-03 eta: 8:15:10 time: 0.1598 data_time: 0.0104 memory: 7124 grad_norm: 5.2940 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4394 loss: 1.4394 2022/09/07 07:31:00 - mmengine - INFO - Epoch(train) [51][600/3757] lr: 1.0000e-03 eta: 8:14:54 time: 0.1663 data_time: 0.0105 memory: 7124 grad_norm: 5.2694 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.3636 loss: 1.3636 2022/09/07 07:31:16 - mmengine - INFO - Epoch(train) [51][700/3757] lr: 1.0000e-03 eta: 8:14:38 time: 0.1683 data_time: 0.0106 memory: 7124 grad_norm: 5.3776 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.2655 loss: 1.2655 2022/09/07 07:31:32 - mmengine - INFO - Epoch(train) [51][800/3757] lr: 1.0000e-03 eta: 8:14:23 time: 0.1630 data_time: 0.0105 memory: 7124 grad_norm: 5.3675 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.7522 loss: 1.7522 2022/09/07 07:31:48 - mmengine - INFO - Epoch(train) [51][900/3757] lr: 1.0000e-03 eta: 8:14:07 time: 0.1589 data_time: 0.0107 memory: 7124 grad_norm: 5.4121 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.4899 loss: 1.4899 2022/09/07 07:32:04 - mmengine - INFO - Epoch(train) [51][1000/3757] lr: 1.0000e-03 eta: 8:13:51 time: 0.1604 data_time: 0.0110 memory: 7124 grad_norm: 5.4044 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.4532 loss: 1.4532 2022/09/07 07:32:20 - mmengine - INFO - Epoch(train) [51][1100/3757] lr: 1.0000e-03 eta: 8:13:35 time: 0.1600 data_time: 0.0096 memory: 7124 grad_norm: 5.4843 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.5592 loss: 1.5592 2022/09/07 07:32:28 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:32:36 - mmengine - INFO - Epoch(train) [51][1200/3757] lr: 1.0000e-03 eta: 8:13:19 time: 0.1588 data_time: 0.0105 memory: 7124 grad_norm: 5.1886 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.3869 loss: 1.3869 2022/09/07 07:32:52 - mmengine - INFO - Epoch(train) [51][1300/3757] lr: 1.0000e-03 eta: 8:13:04 time: 0.1598 data_time: 0.0113 memory: 7124 grad_norm: 5.3936 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.5426 loss: 1.5426 2022/09/07 07:33:08 - mmengine - INFO - Epoch(train) [51][1400/3757] lr: 1.0000e-03 eta: 8:12:48 time: 0.1571 data_time: 0.0107 memory: 7124 grad_norm: 5.3460 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.6214 loss: 1.6214 2022/09/07 07:33:24 - mmengine - INFO - Epoch(train) [51][1500/3757] lr: 1.0000e-03 eta: 8:12:32 time: 0.1570 data_time: 0.0097 memory: 7124 grad_norm: 5.5951 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3982 loss: 1.3982 2022/09/07 07:33:40 - mmengine - INFO - Epoch(train) [51][1600/3757] lr: 1.0000e-03 eta: 8:12:16 time: 0.1597 data_time: 0.0100 memory: 7124 grad_norm: 5.5746 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6707 loss: 1.6707 2022/09/07 07:33:56 - mmengine - INFO - Epoch(train) [51][1700/3757] lr: 1.0000e-03 eta: 8:12:00 time: 0.1571 data_time: 0.0106 memory: 7124 grad_norm: 5.7624 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.6368 loss: 1.6368 2022/09/07 07:34:12 - mmengine - INFO - Epoch(train) [51][1800/3757] lr: 1.0000e-03 eta: 8:11:44 time: 0.1613 data_time: 0.0115 memory: 7124 grad_norm: 5.2697 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.6763 loss: 1.6763 2022/09/07 07:34:28 - mmengine - INFO - Epoch(train) [51][1900/3757] lr: 1.0000e-03 eta: 8:11:29 time: 0.1593 data_time: 0.0087 memory: 7124 grad_norm: 5.4155 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.4397 loss: 1.4397 2022/09/07 07:34:44 - mmengine - INFO - Epoch(train) [51][2000/3757] lr: 1.0000e-03 eta: 8:11:13 time: 0.1563 data_time: 0.0110 memory: 7124 grad_norm: 5.5586 top1_acc: 0.2500 top5_acc: 0.5000 loss_cls: 1.6621 loss: 1.6621 2022/09/07 07:35:00 - mmengine - INFO - Epoch(train) [51][2100/3757] lr: 1.0000e-03 eta: 8:10:57 time: 0.1577 data_time: 0.0107 memory: 7124 grad_norm: 5.5255 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.7556 loss: 1.7556 2022/09/07 07:35:08 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:35:16 - mmengine - INFO - Epoch(train) [51][2200/3757] lr: 1.0000e-03 eta: 8:10:41 time: 0.1581 data_time: 0.0113 memory: 7124 grad_norm: 5.6771 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.3883 loss: 1.3883 2022/09/07 07:35:32 - mmengine - INFO - Epoch(train) [51][2300/3757] lr: 1.0000e-03 eta: 8:10:25 time: 0.1605 data_time: 0.0105 memory: 7124 grad_norm: 5.5484 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.2278 loss: 1.2278 2022/09/07 07:35:48 - mmengine - INFO - Epoch(train) [51][2400/3757] lr: 1.0000e-03 eta: 8:10:10 time: 0.1578 data_time: 0.0114 memory: 7124 grad_norm: 5.5362 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.5827 loss: 1.5827 2022/09/07 07:36:04 - mmengine - INFO - Epoch(train) [51][2500/3757] lr: 1.0000e-03 eta: 8:09:54 time: 0.1571 data_time: 0.0091 memory: 7124 grad_norm: 5.3975 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.4881 loss: 1.4881 2022/09/07 07:36:20 - mmengine - INFO - Epoch(train) [51][2600/3757] lr: 1.0000e-03 eta: 8:09:38 time: 0.1593 data_time: 0.0105 memory: 7124 grad_norm: 5.3274 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1818 loss: 1.1818 2022/09/07 07:36:36 - mmengine - INFO - Epoch(train) [51][2700/3757] lr: 1.0000e-03 eta: 8:09:22 time: 0.1588 data_time: 0.0109 memory: 7124 grad_norm: 5.4963 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.4570 loss: 1.4570 2022/09/07 07:36:52 - mmengine - INFO - Epoch(train) [51][2800/3757] lr: 1.0000e-03 eta: 8:09:06 time: 0.1601 data_time: 0.0102 memory: 7124 grad_norm: 5.5599 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.5292 loss: 1.5292 2022/09/07 07:37:08 - mmengine - INFO - Epoch(train) [51][2900/3757] lr: 1.0000e-03 eta: 8:08:51 time: 0.1575 data_time: 0.0105 memory: 7124 grad_norm: 5.5342 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.4854 loss: 1.4854 2022/09/07 07:37:24 - mmengine - INFO - Epoch(train) [51][3000/3757] lr: 1.0000e-03 eta: 8:08:35 time: 0.1590 data_time: 0.0105 memory: 7124 grad_norm: 5.4871 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2506 loss: 1.2506 2022/09/07 07:37:40 - mmengine - INFO - Epoch(train) [51][3100/3757] lr: 1.0000e-03 eta: 8:08:19 time: 0.1607 data_time: 0.0112 memory: 7124 grad_norm: 5.3938 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2481 loss: 1.2481 2022/09/07 07:37:48 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:37:56 - mmengine - INFO - Epoch(train) [51][3200/3757] lr: 1.0000e-03 eta: 8:08:03 time: 0.1574 data_time: 0.0097 memory: 7124 grad_norm: 5.3495 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.4499 loss: 1.4499 2022/09/07 07:38:12 - mmengine - INFO - Epoch(train) [51][3300/3757] lr: 1.0000e-03 eta: 8:07:47 time: 0.1604 data_time: 0.0115 memory: 7124 grad_norm: 5.2056 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2401 loss: 1.2401 2022/09/07 07:38:28 - mmengine - INFO - Epoch(train) [51][3400/3757] lr: 1.0000e-03 eta: 8:07:31 time: 0.1555 data_time: 0.0115 memory: 7124 grad_norm: 5.6106 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3955 loss: 1.3955 2022/09/07 07:38:44 - mmengine - INFO - Epoch(train) [51][3500/3757] lr: 1.0000e-03 eta: 8:07:16 time: 0.1598 data_time: 0.0114 memory: 7124 grad_norm: 5.4253 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.3026 loss: 1.3026 2022/09/07 07:39:00 - mmengine - INFO - Epoch(train) [51][3600/3757] lr: 1.0000e-03 eta: 8:07:00 time: 0.1595 data_time: 0.0115 memory: 7124 grad_norm: 5.3555 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3340 loss: 1.3340 2022/09/07 07:39:16 - mmengine - INFO - Epoch(train) [51][3700/3757] lr: 1.0000e-03 eta: 8:06:44 time: 0.1562 data_time: 0.0104 memory: 7124 grad_norm: 5.3899 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1981 loss: 1.1981 2022/09/07 07:39:24 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:39:24 - mmengine - INFO - Epoch(train) [51][3757/3757] lr: 1.0000e-03 eta: 8:06:38 time: 0.1352 data_time: 0.0075 memory: 7124 grad_norm: 5.3491 top1_acc: 0.7143 top5_acc: 1.0000 loss_cls: 1.1149 loss: 1.1149 2022/09/07 07:39:24 - mmengine - INFO - Saving checkpoint at 51 epochs 2022/09/07 07:41:43 - mmengine - INFO - Epoch(val) [51][100/310] eta: 0:04:05 time: 1.1707 data_time: 0.8667 memory: 7627 2022/09/07 07:44:00 - mmengine - INFO - Epoch(val) [51][200/310] eta: 0:02:29 time: 1.3610 data_time: 1.0464 memory: 7627 2022/09/07 07:46:04 - mmengine - INFO - Epoch(val) [51][300/310] eta: 0:00:11 time: 1.1047 data_time: 0.8004 memory: 7627 2022/09/07 07:46:22 - mmengine - INFO - Epoch(val) [51][310/310] acc/top1: 0.7243 acc/top5: 0.9040 acc/mean1: 0.7243 2022/09/07 07:46:22 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_46.pth is removed 2022/09/07 07:46:24 - mmengine - INFO - The best checkpoint with 0.7243 acc/top1 at 51 epoch is saved to best_acc/top1_epoch_51.pth. 2022/09/07 07:46:41 - mmengine - INFO - Epoch(train) [52][100/3757] lr: 1.0000e-03 eta: 8:06:18 time: 0.1728 data_time: 0.0106 memory: 7627 grad_norm: 5.4096 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2786 loss: 1.2786 2022/09/07 07:46:57 - mmengine - INFO - Epoch(train) [52][200/3757] lr: 1.0000e-03 eta: 8:06:02 time: 0.1563 data_time: 0.0128 memory: 7124 grad_norm: 5.3387 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.4411 loss: 1.4411 2022/09/07 07:47:13 - mmengine - INFO - Epoch(train) [52][300/3757] lr: 1.0000e-03 eta: 8:05:46 time: 0.1583 data_time: 0.0108 memory: 7124 grad_norm: 5.6785 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.3571 loss: 1.3571 2022/09/07 07:47:28 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:47:29 - mmengine - INFO - Epoch(train) [52][400/3757] lr: 1.0000e-03 eta: 8:05:30 time: 0.1582 data_time: 0.0115 memory: 7124 grad_norm: 5.3895 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.4057 loss: 1.4057 2022/09/07 07:47:45 - mmengine - INFO - Epoch(train) [52][500/3757] lr: 1.0000e-03 eta: 8:05:14 time: 0.1634 data_time: 0.0118 memory: 7124 grad_norm: 5.5515 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.4394 loss: 1.4394 2022/09/07 07:48:01 - mmengine - INFO - Epoch(train) [52][600/3757] lr: 1.0000e-03 eta: 8:04:58 time: 0.1590 data_time: 0.0114 memory: 7124 grad_norm: 5.4872 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.0928 loss: 1.0928 2022/09/07 07:48:17 - mmengine - INFO - Epoch(train) [52][700/3757] lr: 1.0000e-03 eta: 8:04:42 time: 0.1594 data_time: 0.0110 memory: 7124 grad_norm: 5.3004 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3329 loss: 1.3329 2022/09/07 07:48:33 - mmengine - INFO - Epoch(train) [52][800/3757] lr: 1.0000e-03 eta: 8:04:27 time: 0.1626 data_time: 0.0107 memory: 7124 grad_norm: 5.3275 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3787 loss: 1.3787 2022/09/07 07:48:49 - mmengine - INFO - Epoch(train) [52][900/3757] lr: 1.0000e-03 eta: 8:04:11 time: 0.1608 data_time: 0.0115 memory: 7124 grad_norm: 5.6626 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3997 loss: 1.3997 2022/09/07 07:49:05 - mmengine - INFO - Epoch(train) [52][1000/3757] lr: 1.0000e-03 eta: 8:03:55 time: 0.1627 data_time: 0.0106 memory: 7124 grad_norm: 5.5484 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5111 loss: 1.5111 2022/09/07 07:49:21 - mmengine - INFO - Epoch(train) [52][1100/3757] lr: 1.0000e-03 eta: 8:03:39 time: 0.1621 data_time: 0.0124 memory: 7124 grad_norm: 5.4176 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0014 loss: 1.0014 2022/09/07 07:49:37 - mmengine - INFO - Epoch(train) [52][1200/3757] lr: 1.0000e-03 eta: 8:03:23 time: 0.1594 data_time: 0.0110 memory: 7124 grad_norm: 5.6546 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.4790 loss: 1.4790 2022/09/07 07:49:53 - mmengine - INFO - Epoch(train) [52][1300/3757] lr: 1.0000e-03 eta: 8:03:08 time: 0.1604 data_time: 0.0103 memory: 7124 grad_norm: 5.7214 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.4577 loss: 1.4577 2022/09/07 07:50:08 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:50:09 - mmengine - INFO - Epoch(train) [52][1400/3757] lr: 1.0000e-03 eta: 8:02:52 time: 0.1575 data_time: 0.0118 memory: 7124 grad_norm: 5.5837 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.3393 loss: 1.3393 2022/09/07 07:50:25 - mmengine - INFO - Epoch(train) [52][1500/3757] lr: 1.0000e-03 eta: 8:02:36 time: 0.1555 data_time: 0.0122 memory: 7124 grad_norm: 5.5419 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.2609 loss: 1.2609 2022/09/07 07:50:41 - mmengine - INFO - Epoch(train) [52][1600/3757] lr: 1.0000e-03 eta: 8:02:20 time: 0.1611 data_time: 0.0107 memory: 7124 grad_norm: 5.5795 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.4513 loss: 1.4513 2022/09/07 07:50:57 - mmengine - INFO - Epoch(train) [52][1700/3757] lr: 1.0000e-03 eta: 8:02:04 time: 0.1562 data_time: 0.0107 memory: 7124 grad_norm: 5.5873 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.3861 loss: 1.3861 2022/09/07 07:51:13 - mmengine - INFO - Epoch(train) [52][1800/3757] lr: 1.0000e-03 eta: 8:01:49 time: 0.1594 data_time: 0.0133 memory: 7124 grad_norm: 5.5494 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2913 loss: 1.2913 2022/09/07 07:51:29 - mmengine - INFO - Epoch(train) [52][1900/3757] lr: 1.0000e-03 eta: 8:01:33 time: 0.1544 data_time: 0.0104 memory: 7124 grad_norm: 5.7164 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2410 loss: 1.2410 2022/09/07 07:51:45 - mmengine - INFO - Epoch(train) [52][2000/3757] lr: 1.0000e-03 eta: 8:01:17 time: 0.1604 data_time: 0.0105 memory: 7124 grad_norm: 5.3137 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1860 loss: 1.1860 2022/09/07 07:52:01 - mmengine - INFO - Epoch(train) [52][2100/3757] lr: 1.0000e-03 eta: 8:01:01 time: 0.1554 data_time: 0.0102 memory: 7124 grad_norm: 5.7687 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2558 loss: 1.2558 2022/09/07 07:52:17 - mmengine - INFO - Epoch(train) [52][2200/3757] lr: 1.0000e-03 eta: 8:00:45 time: 0.1573 data_time: 0.0114 memory: 7124 grad_norm: 5.4456 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2890 loss: 1.2890 2022/09/07 07:52:33 - mmengine - INFO - Epoch(train) [52][2300/3757] lr: 1.0000e-03 eta: 8:00:29 time: 0.1580 data_time: 0.0114 memory: 7124 grad_norm: 5.3249 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3138 loss: 1.3138 2022/09/07 07:52:47 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:52:49 - mmengine - INFO - Epoch(train) [52][2400/3757] lr: 1.0000e-03 eta: 8:00:13 time: 0.1539 data_time: 0.0103 memory: 7124 grad_norm: 5.7594 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.5111 loss: 1.5111 2022/09/07 07:53:05 - mmengine - INFO - Epoch(train) [52][2500/3757] lr: 1.0000e-03 eta: 7:59:58 time: 0.1609 data_time: 0.0129 memory: 7124 grad_norm: 5.6199 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2461 loss: 1.2461 2022/09/07 07:53:20 - mmengine - INFO - Epoch(train) [52][2600/3757] lr: 1.0000e-03 eta: 7:59:42 time: 0.1571 data_time: 0.0107 memory: 7124 grad_norm: 5.5608 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2736 loss: 1.2736 2022/09/07 07:53:36 - mmengine - INFO - Epoch(train) [52][2700/3757] lr: 1.0000e-03 eta: 7:59:26 time: 0.1581 data_time: 0.0111 memory: 7124 grad_norm: 5.6693 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.5038 loss: 1.5038 2022/09/07 07:53:52 - mmengine - INFO - Epoch(train) [52][2800/3757] lr: 1.0000e-03 eta: 7:59:10 time: 0.1602 data_time: 0.0102 memory: 7124 grad_norm: 5.4400 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2281 loss: 1.2281 2022/09/07 07:54:08 - mmengine - INFO - Epoch(train) [52][2900/3757] lr: 1.0000e-03 eta: 7:58:54 time: 0.1554 data_time: 0.0111 memory: 7124 grad_norm: 5.7055 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.3316 loss: 1.3316 2022/09/07 07:54:24 - mmengine - INFO - Epoch(train) [52][3000/3757] lr: 1.0000e-03 eta: 7:58:38 time: 0.1605 data_time: 0.0116 memory: 7124 grad_norm: 5.4700 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.3493 loss: 1.3493 2022/09/07 07:54:40 - mmengine - INFO - Epoch(train) [52][3100/3757] lr: 1.0000e-03 eta: 7:58:22 time: 0.1536 data_time: 0.0107 memory: 7124 grad_norm: 5.6861 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.3258 loss: 1.3258 2022/09/07 07:54:56 - mmengine - INFO - Epoch(train) [52][3200/3757] lr: 1.0000e-03 eta: 7:58:07 time: 0.1586 data_time: 0.0105 memory: 7124 grad_norm: 5.7615 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2249 loss: 1.2249 2022/09/07 07:55:12 - mmengine - INFO - Epoch(train) [52][3300/3757] lr: 1.0000e-03 eta: 7:57:51 time: 0.1690 data_time: 0.0115 memory: 7124 grad_norm: 5.8284 top1_acc: 0.2500 top5_acc: 1.0000 loss_cls: 1.4298 loss: 1.4298 2022/09/07 07:55:27 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:55:28 - mmengine - INFO - Epoch(train) [52][3400/3757] lr: 1.0000e-03 eta: 7:57:35 time: 0.1571 data_time: 0.0111 memory: 7124 grad_norm: 5.4730 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0604 loss: 1.0604 2022/09/07 07:55:44 - mmengine - INFO - Epoch(train) [52][3500/3757] lr: 1.0000e-03 eta: 7:57:19 time: 0.1568 data_time: 0.0103 memory: 7124 grad_norm: 5.6262 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.5032 loss: 1.5032 2022/09/07 07:56:00 - mmengine - INFO - Epoch(train) [52][3600/3757] lr: 1.0000e-03 eta: 7:57:04 time: 0.1538 data_time: 0.0106 memory: 7124 grad_norm: 5.6756 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.5950 loss: 1.5950 2022/09/07 07:56:16 - mmengine - INFO - Epoch(train) [52][3700/3757] lr: 1.0000e-03 eta: 7:56:48 time: 0.1562 data_time: 0.0109 memory: 7124 grad_norm: 5.4242 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2620 loss: 1.2620 2022/09/07 07:56:25 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 07:56:25 - mmengine - INFO - Epoch(train) [52][3757/3757] lr: 1.0000e-03 eta: 7:56:41 time: 0.1369 data_time: 0.0073 memory: 7124 grad_norm: 5.6695 top1_acc: 0.5714 top5_acc: 0.8571 loss_cls: 1.5517 loss: 1.5517 2022/09/07 07:56:25 - mmengine - INFO - Saving checkpoint at 52 epochs 2022/09/07 07:58:43 - mmengine - INFO - Epoch(val) [52][100/310] eta: 0:04:04 time: 1.1649 data_time: 0.8598 memory: 7627 2022/09/07 08:01:01 - mmengine - INFO - Epoch(val) [52][200/310] eta: 0:02:29 time: 1.3573 data_time: 1.0517 memory: 7627 2022/09/07 08:03:05 - mmengine - INFO - Epoch(val) [52][300/310] eta: 0:00:11 time: 1.1265 data_time: 0.8251 memory: 7627 2022/09/07 08:03:22 - mmengine - INFO - Epoch(val) [52][310/310] acc/top1: 0.7309 acc/top5: 0.9079 acc/mean1: 0.7309 2022/09/07 08:03:22 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_51.pth is removed 2022/09/07 08:03:24 - mmengine - INFO - The best checkpoint with 0.7309 acc/top1 at 52 epoch is saved to best_acc/top1_epoch_52.pth. 2022/09/07 08:03:40 - mmengine - INFO - Epoch(train) [53][100/3757] lr: 1.0000e-03 eta: 7:56:21 time: 0.1557 data_time: 0.0116 memory: 7627 grad_norm: 5.8217 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4312 loss: 1.4312 2022/09/07 08:03:56 - mmengine - INFO - Epoch(train) [53][200/3757] lr: 1.0000e-03 eta: 7:56:05 time: 0.1572 data_time: 0.0099 memory: 7124 grad_norm: 5.6918 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.2782 loss: 1.2782 2022/09/07 08:04:12 - mmengine - INFO - Epoch(train) [53][300/3757] lr: 1.0000e-03 eta: 7:55:49 time: 0.1573 data_time: 0.0105 memory: 7124 grad_norm: 5.6473 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3703 loss: 1.3703 2022/09/07 08:04:28 - mmengine - INFO - Epoch(train) [53][400/3757] lr: 1.0000e-03 eta: 7:55:34 time: 0.1542 data_time: 0.0108 memory: 7124 grad_norm: 5.5291 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.3113 loss: 1.3113 2022/09/07 08:04:44 - mmengine - INFO - Epoch(train) [53][500/3757] lr: 1.0000e-03 eta: 7:55:18 time: 0.1576 data_time: 0.0109 memory: 7124 grad_norm: 5.5485 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4269 loss: 1.4269 2022/09/07 08:05:00 - mmengine - INFO - Epoch(train) [53][600/3757] lr: 1.0000e-03 eta: 7:55:02 time: 0.1591 data_time: 0.0104 memory: 7124 grad_norm: 5.4676 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.3321 loss: 1.3321 2022/09/07 08:05:06 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:05:16 - mmengine - INFO - Epoch(train) [53][700/3757] lr: 1.0000e-03 eta: 7:54:46 time: 0.1551 data_time: 0.0118 memory: 7124 grad_norm: 5.4375 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2660 loss: 1.2660 2022/09/07 08:05:32 - mmengine - INFO - Epoch(train) [53][800/3757] lr: 1.0000e-03 eta: 7:54:30 time: 0.1542 data_time: 0.0094 memory: 7124 grad_norm: 5.7232 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2232 loss: 1.2232 2022/09/07 08:05:48 - mmengine - INFO - Epoch(train) [53][900/3757] lr: 1.0000e-03 eta: 7:54:14 time: 0.1535 data_time: 0.0118 memory: 7124 grad_norm: 5.5141 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1880 loss: 1.1880 2022/09/07 08:06:03 - mmengine - INFO - Epoch(train) [53][1000/3757] lr: 1.0000e-03 eta: 7:53:58 time: 0.1582 data_time: 0.0115 memory: 7124 grad_norm: 5.6097 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.5023 loss: 1.5023 2022/09/07 08:06:19 - mmengine - INFO - Epoch(train) [53][1100/3757] lr: 1.0000e-03 eta: 7:53:42 time: 0.1590 data_time: 0.0104 memory: 7124 grad_norm: 5.5076 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.6720 loss: 1.6720 2022/09/07 08:06:35 - mmengine - INFO - Epoch(train) [53][1200/3757] lr: 1.0000e-03 eta: 7:53:26 time: 0.1583 data_time: 0.0111 memory: 7124 grad_norm: 5.5965 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2017 loss: 1.2017 2022/09/07 08:06:51 - mmengine - INFO - Epoch(train) [53][1300/3757] lr: 1.0000e-03 eta: 7:53:11 time: 0.1586 data_time: 0.0103 memory: 7124 grad_norm: 5.6593 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.4121 loss: 1.4121 2022/09/07 08:07:07 - mmengine - INFO - Epoch(train) [53][1400/3757] lr: 1.0000e-03 eta: 7:52:55 time: 0.1565 data_time: 0.0110 memory: 7124 grad_norm: 5.6667 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2744 loss: 1.2744 2022/09/07 08:07:23 - mmengine - INFO - Epoch(train) [53][1500/3757] lr: 1.0000e-03 eta: 7:52:39 time: 0.1596 data_time: 0.0123 memory: 7124 grad_norm: 5.6219 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2972 loss: 1.2972 2022/09/07 08:07:39 - mmengine - INFO - Epoch(train) [53][1600/3757] lr: 1.0000e-03 eta: 7:52:23 time: 0.1592 data_time: 0.0105 memory: 7124 grad_norm: 5.3991 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3214 loss: 1.3214 2022/09/07 08:07:45 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:07:56 - mmengine - INFO - Epoch(train) [53][1700/3757] lr: 1.0000e-03 eta: 7:52:08 time: 0.1775 data_time: 0.0114 memory: 7124 grad_norm: 5.4038 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.1493 loss: 1.1493 2022/09/07 08:08:11 - mmengine - INFO - Epoch(train) [53][1800/3757] lr: 1.0000e-03 eta: 7:51:52 time: 0.1560 data_time: 0.0124 memory: 7124 grad_norm: 5.5728 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.3373 loss: 1.3373 2022/09/07 08:08:27 - mmengine - INFO - Epoch(train) [53][1900/3757] lr: 1.0000e-03 eta: 7:51:36 time: 0.1562 data_time: 0.0105 memory: 7124 grad_norm: 5.5188 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.3661 loss: 1.3661 2022/09/07 08:08:43 - mmengine - INFO - Epoch(train) [53][2000/3757] lr: 1.0000e-03 eta: 7:51:20 time: 0.1635 data_time: 0.0128 memory: 7124 grad_norm: 5.4323 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0772 loss: 1.0772 2022/09/07 08:08:59 - mmengine - INFO - Epoch(train) [53][2100/3757] lr: 1.0000e-03 eta: 7:51:04 time: 0.1568 data_time: 0.0121 memory: 7124 grad_norm: 5.7303 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.4700 loss: 1.4700 2022/09/07 08:09:15 - mmengine - INFO - Epoch(train) [53][2200/3757] lr: 1.0000e-03 eta: 7:50:48 time: 0.1587 data_time: 0.0104 memory: 7124 grad_norm: 5.6437 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.3627 loss: 1.3627 2022/09/07 08:09:31 - mmengine - INFO - Epoch(train) [53][2300/3757] lr: 1.0000e-03 eta: 7:50:32 time: 0.1546 data_time: 0.0115 memory: 7124 grad_norm: 5.7082 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2341 loss: 1.2341 2022/09/07 08:09:47 - mmengine - INFO - Epoch(train) [53][2400/3757] lr: 1.0000e-03 eta: 7:50:16 time: 0.1584 data_time: 0.0116 memory: 7124 grad_norm: 5.4969 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3850 loss: 1.3850 2022/09/07 08:10:03 - mmengine - INFO - Epoch(train) [53][2500/3757] lr: 1.0000e-03 eta: 7:50:00 time: 0.1582 data_time: 0.0110 memory: 7124 grad_norm: 5.6624 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0952 loss: 1.0952 2022/09/07 08:10:19 - mmengine - INFO - Epoch(train) [53][2600/3757] lr: 1.0000e-03 eta: 7:49:44 time: 0.1558 data_time: 0.0107 memory: 7124 grad_norm: 5.2794 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.4457 loss: 1.4457 2022/09/07 08:10:24 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:10:34 - mmengine - INFO - Epoch(train) [53][2700/3757] lr: 1.0000e-03 eta: 7:49:29 time: 0.1577 data_time: 0.0110 memory: 7124 grad_norm: 5.5623 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1971 loss: 1.1971 2022/09/07 08:10:50 - mmengine - INFO - Epoch(train) [53][2800/3757] lr: 1.0000e-03 eta: 7:49:13 time: 0.1572 data_time: 0.0122 memory: 7124 grad_norm: 5.6162 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2724 loss: 1.2724 2022/09/07 08:11:06 - mmengine - INFO - Epoch(train) [53][2900/3757] lr: 1.0000e-03 eta: 7:48:57 time: 0.1587 data_time: 0.0117 memory: 7124 grad_norm: 5.8752 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.5156 loss: 1.5156 2022/09/07 08:11:22 - mmengine - INFO - Epoch(train) [53][3000/3757] lr: 1.0000e-03 eta: 7:48:41 time: 0.1556 data_time: 0.0113 memory: 7124 grad_norm: 5.6973 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1595 loss: 1.1595 2022/09/07 08:11:38 - mmengine - INFO - Epoch(train) [53][3100/3757] lr: 1.0000e-03 eta: 7:48:25 time: 0.1576 data_time: 0.0117 memory: 7124 grad_norm: 5.6929 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.3028 loss: 1.3028 2022/09/07 08:11:54 - mmengine - INFO - Epoch(train) [53][3200/3757] lr: 1.0000e-03 eta: 7:48:09 time: 0.1637 data_time: 0.0116 memory: 7124 grad_norm: 5.8881 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.5495 loss: 1.5495 2022/09/07 08:12:10 - mmengine - INFO - Epoch(train) [53][3300/3757] lr: 1.0000e-03 eta: 7:47:53 time: 0.1574 data_time: 0.0100 memory: 7124 grad_norm: 5.8697 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.4042 loss: 1.4042 2022/09/07 08:12:26 - mmengine - INFO - Epoch(train) [53][3400/3757] lr: 1.0000e-03 eta: 7:47:37 time: 0.1556 data_time: 0.0109 memory: 7124 grad_norm: 5.8644 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.4064 loss: 1.4064 2022/09/07 08:12:42 - mmengine - INFO - Epoch(train) [53][3500/3757] lr: 1.0000e-03 eta: 7:47:22 time: 0.1552 data_time: 0.0096 memory: 7124 grad_norm: 5.7754 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.4489 loss: 1.4489 2022/09/07 08:12:58 - mmengine - INFO - Epoch(train) [53][3600/3757] lr: 1.0000e-03 eta: 7:47:06 time: 0.1606 data_time: 0.0120 memory: 7124 grad_norm: 5.9277 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1105 loss: 1.1105 2022/09/07 08:13:04 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:13:14 - mmengine - INFO - Epoch(train) [53][3700/3757] lr: 1.0000e-03 eta: 7:46:50 time: 0.1599 data_time: 0.0114 memory: 7124 grad_norm: 5.9078 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.4660 loss: 1.4660 2022/09/07 08:13:22 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:13:22 - mmengine - INFO - Epoch(train) [53][3757/3757] lr: 1.0000e-03 eta: 7:46:44 time: 0.1354 data_time: 0.0072 memory: 7124 grad_norm: 5.5821 top1_acc: 0.5714 top5_acc: 0.7143 loss_cls: 1.2116 loss: 1.2116 2022/09/07 08:13:22 - mmengine - INFO - Saving checkpoint at 53 epochs 2022/09/07 08:15:40 - mmengine - INFO - Epoch(val) [53][100/310] eta: 0:03:47 time: 1.0826 data_time: 0.7817 memory: 7627 2022/09/07 08:17:57 - mmengine - INFO - Epoch(val) [53][200/310] eta: 0:02:25 time: 1.3264 data_time: 1.0186 memory: 7627 2022/09/07 08:20:02 - mmengine - INFO - Epoch(val) [53][300/310] eta: 0:00:11 time: 1.1907 data_time: 0.8853 memory: 7627 2022/09/07 08:20:19 - mmengine - INFO - Epoch(val) [53][310/310] acc/top1: 0.7337 acc/top5: 0.9099 acc/mean1: 0.7337 2022/09/07 08:20:19 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_52.pth is removed 2022/09/07 08:20:21 - mmengine - INFO - The best checkpoint with 0.7337 acc/top1 at 53 epoch is saved to best_acc/top1_epoch_53.pth. 2022/09/07 08:20:38 - mmengine - INFO - Epoch(train) [54][100/3757] lr: 1.0000e-03 eta: 7:46:23 time: 0.1626 data_time: 0.0104 memory: 7627 grad_norm: 5.7893 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2475 loss: 1.2475 2022/09/07 08:20:54 - mmengine - INFO - Epoch(train) [54][200/3757] lr: 1.0000e-03 eta: 7:46:08 time: 0.1598 data_time: 0.0119 memory: 7124 grad_norm: 5.7273 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.3861 loss: 1.3861 2022/09/07 08:21:10 - mmengine - INFO - Epoch(train) [54][300/3757] lr: 1.0000e-03 eta: 7:45:52 time: 0.1569 data_time: 0.0110 memory: 7124 grad_norm: 5.8651 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.1821 loss: 1.1821 2022/09/07 08:21:26 - mmengine - INFO - Epoch(train) [54][400/3757] lr: 1.0000e-03 eta: 7:45:36 time: 0.1665 data_time: 0.0101 memory: 7124 grad_norm: 5.6006 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2780 loss: 1.2780 2022/09/07 08:21:42 - mmengine - INFO - Epoch(train) [54][500/3757] lr: 1.0000e-03 eta: 7:45:20 time: 0.1585 data_time: 0.0121 memory: 7124 grad_norm: 5.7350 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.5186 loss: 1.5186 2022/09/07 08:21:58 - mmengine - INFO - Epoch(train) [54][600/3757] lr: 1.0000e-03 eta: 7:45:04 time: 0.1558 data_time: 0.0106 memory: 7124 grad_norm: 5.7043 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.4829 loss: 1.4829 2022/09/07 08:22:14 - mmengine - INFO - Epoch(train) [54][700/3757] lr: 1.0000e-03 eta: 7:44:48 time: 0.1554 data_time: 0.0111 memory: 7124 grad_norm: 5.6777 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2178 loss: 1.2178 2022/09/07 08:22:30 - mmengine - INFO - Epoch(train) [54][800/3757] lr: 1.0000e-03 eta: 7:44:32 time: 0.1566 data_time: 0.0098 memory: 7124 grad_norm: 5.7896 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.3719 loss: 1.3719 2022/09/07 08:22:42 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:22:46 - mmengine - INFO - Epoch(train) [54][900/3757] lr: 1.0000e-03 eta: 7:44:16 time: 0.1570 data_time: 0.0124 memory: 7124 grad_norm: 5.7483 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2722 loss: 1.2722 2022/09/07 08:23:01 - mmengine - INFO - Epoch(train) [54][1000/3757] lr: 1.0000e-03 eta: 7:44:00 time: 0.1571 data_time: 0.0109 memory: 7124 grad_norm: 5.6441 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1349 loss: 1.1349 2022/09/07 08:23:17 - mmengine - INFO - Epoch(train) [54][1100/3757] lr: 1.0000e-03 eta: 7:43:45 time: 0.1566 data_time: 0.0117 memory: 7124 grad_norm: 5.8360 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.3137 loss: 1.3137 2022/09/07 08:23:33 - mmengine - INFO - Epoch(train) [54][1200/3757] lr: 1.0000e-03 eta: 7:43:29 time: 0.1597 data_time: 0.0119 memory: 7124 grad_norm: 5.6649 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2891 loss: 1.2891 2022/09/07 08:23:49 - mmengine - INFO - Epoch(train) [54][1300/3757] lr: 1.0000e-03 eta: 7:43:13 time: 0.1555 data_time: 0.0112 memory: 7124 grad_norm: 5.7307 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.3832 loss: 1.3832 2022/09/07 08:24:05 - mmengine - INFO - Epoch(train) [54][1400/3757] lr: 1.0000e-03 eta: 7:42:57 time: 0.1576 data_time: 0.0121 memory: 7124 grad_norm: 5.5666 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1446 loss: 1.1446 2022/09/07 08:24:20 - mmengine - INFO - Epoch(train) [54][1500/3757] lr: 1.0000e-03 eta: 7:42:41 time: 0.1541 data_time: 0.0110 memory: 7124 grad_norm: 5.6593 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3659 loss: 1.3659 2022/09/07 08:24:36 - mmengine - INFO - Epoch(train) [54][1600/3757] lr: 1.0000e-03 eta: 7:42:25 time: 0.1560 data_time: 0.0110 memory: 7124 grad_norm: 5.6052 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.2744 loss: 1.2744 2022/09/07 08:24:52 - mmengine - INFO - Epoch(train) [54][1700/3757] lr: 1.0000e-03 eta: 7:42:09 time: 0.1568 data_time: 0.0102 memory: 7124 grad_norm: 5.4942 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2599 loss: 1.2599 2022/09/07 08:25:08 - mmengine - INFO - Epoch(train) [54][1800/3757] lr: 1.0000e-03 eta: 7:41:53 time: 0.1558 data_time: 0.0108 memory: 7124 grad_norm: 5.8235 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3727 loss: 1.3727 2022/09/07 08:25:21 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:25:24 - mmengine - INFO - Epoch(train) [54][1900/3757] lr: 1.0000e-03 eta: 7:41:37 time: 0.1551 data_time: 0.0101 memory: 7124 grad_norm: 5.7663 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2045 loss: 1.2045 2022/09/07 08:25:40 - mmengine - INFO - Epoch(train) [54][2000/3757] lr: 1.0000e-03 eta: 7:41:21 time: 0.1576 data_time: 0.0117 memory: 7124 grad_norm: 5.7263 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.3259 loss: 1.3259 2022/09/07 08:25:56 - mmengine - INFO - Epoch(train) [54][2100/3757] lr: 1.0000e-03 eta: 7:41:05 time: 0.1577 data_time: 0.0129 memory: 7124 grad_norm: 5.7282 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.6045 loss: 1.6045 2022/09/07 08:26:12 - mmengine - INFO - Epoch(train) [54][2200/3757] lr: 1.0000e-03 eta: 7:40:49 time: 0.1548 data_time: 0.0107 memory: 7124 grad_norm: 5.8291 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.2707 loss: 1.2707 2022/09/07 08:26:28 - mmengine - INFO - Epoch(train) [54][2300/3757] lr: 1.0000e-03 eta: 7:40:33 time: 0.1547 data_time: 0.0119 memory: 7124 grad_norm: 5.6838 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2979 loss: 1.2979 2022/09/07 08:26:44 - mmengine - INFO - Epoch(train) [54][2400/3757] lr: 1.0000e-03 eta: 7:40:18 time: 0.1588 data_time: 0.0119 memory: 7124 grad_norm: 5.7602 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.2831 loss: 1.2831 2022/09/07 08:27:00 - mmengine - INFO - Epoch(train) [54][2500/3757] lr: 1.0000e-03 eta: 7:40:02 time: 0.1535 data_time: 0.0100 memory: 7124 grad_norm: 5.5199 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2112 loss: 1.2112 2022/09/07 08:27:15 - mmengine - INFO - Epoch(train) [54][2600/3757] lr: 1.0000e-03 eta: 7:39:46 time: 0.1596 data_time: 0.0112 memory: 7124 grad_norm: 5.8668 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0553 loss: 1.0553 2022/09/07 08:27:31 - mmengine - INFO - Epoch(train) [54][2700/3757] lr: 1.0000e-03 eta: 7:39:30 time: 0.1574 data_time: 0.0108 memory: 7124 grad_norm: 5.4515 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1708 loss: 1.1708 2022/09/07 08:27:47 - mmengine - INFO - Epoch(train) [54][2800/3757] lr: 1.0000e-03 eta: 7:39:14 time: 0.1606 data_time: 0.0122 memory: 7124 grad_norm: 5.6162 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.3710 loss: 1.3710 2022/09/07 08:28:00 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:28:03 - mmengine - INFO - Epoch(train) [54][2900/3757] lr: 1.0000e-03 eta: 7:38:58 time: 0.1603 data_time: 0.0113 memory: 7124 grad_norm: 5.6991 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.3087 loss: 1.3087 2022/09/07 08:28:19 - mmengine - INFO - Epoch(train) [54][3000/3757] lr: 1.0000e-03 eta: 7:38:42 time: 0.1587 data_time: 0.0117 memory: 7124 grad_norm: 5.7246 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2250 loss: 1.2250 2022/09/07 08:28:35 - mmengine - INFO - Epoch(train) [54][3100/3757] lr: 1.0000e-03 eta: 7:38:26 time: 0.1603 data_time: 0.0112 memory: 7124 grad_norm: 5.6984 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.3891 loss: 1.3891 2022/09/07 08:28:51 - mmengine - INFO - Epoch(train) [54][3200/3757] lr: 1.0000e-03 eta: 7:38:10 time: 0.1562 data_time: 0.0110 memory: 7124 grad_norm: 5.6830 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.2602 loss: 1.2602 2022/09/07 08:29:07 - mmengine - INFO - Epoch(train) [54][3300/3757] lr: 1.0000e-03 eta: 7:37:54 time: 0.1609 data_time: 0.0122 memory: 7124 grad_norm: 5.8340 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2386 loss: 1.2386 2022/09/07 08:29:22 - mmengine - INFO - Epoch(train) [54][3400/3757] lr: 1.0000e-03 eta: 7:37:39 time: 0.1544 data_time: 0.0112 memory: 7124 grad_norm: 5.7296 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.3000 loss: 1.3000 2022/09/07 08:29:38 - mmengine - INFO - Epoch(train) [54][3500/3757] lr: 1.0000e-03 eta: 7:37:23 time: 0.1549 data_time: 0.0113 memory: 7124 grad_norm: 5.8329 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.3734 loss: 1.3734 2022/09/07 08:29:54 - mmengine - INFO - Epoch(train) [54][3600/3757] lr: 1.0000e-03 eta: 7:37:07 time: 0.1554 data_time: 0.0114 memory: 7124 grad_norm: 5.8844 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2946 loss: 1.2946 2022/09/07 08:30:10 - mmengine - INFO - Epoch(train) [54][3700/3757] lr: 1.0000e-03 eta: 7:36:51 time: 0.1561 data_time: 0.0110 memory: 7124 grad_norm: 5.6721 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3981 loss: 1.3981 2022/09/07 08:30:19 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:30:19 - mmengine - INFO - Epoch(train) [54][3757/3757] lr: 1.0000e-03 eta: 7:36:44 time: 0.1391 data_time: 0.0077 memory: 7124 grad_norm: 5.5464 top1_acc: 0.5714 top5_acc: 1.0000 loss_cls: 1.1838 loss: 1.1838 2022/09/07 08:30:19 - mmengine - INFO - Saving checkpoint at 54 epochs 2022/09/07 08:32:38 - mmengine - INFO - Epoch(val) [54][100/310] eta: 0:04:28 time: 1.2794 data_time: 0.9761 memory: 7627 2022/09/07 08:34:53 - mmengine - INFO - Epoch(val) [54][200/310] eta: 0:02:19 time: 1.2712 data_time: 0.9675 memory: 7627 2022/09/07 08:36:58 - mmengine - INFO - Epoch(val) [54][300/310] eta: 0:00:12 time: 1.2536 data_time: 0.9513 memory: 7627 2022/09/07 08:37:17 - mmengine - INFO - Epoch(val) [54][310/310] acc/top1: 0.7369 acc/top5: 0.9109 acc/mean1: 0.7369 2022/09/07 08:37:17 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_53.pth is removed 2022/09/07 08:37:19 - mmengine - INFO - The best checkpoint with 0.7369 acc/top1 at 54 epoch is saved to best_acc/top1_epoch_54.pth. 2022/09/07 08:37:36 - mmengine - INFO - Epoch(train) [55][100/3757] lr: 1.0000e-03 eta: 7:36:24 time: 0.1575 data_time: 0.0104 memory: 7627 grad_norm: 5.9817 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3046 loss: 1.3046 2022/09/07 08:37:39 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:37:52 - mmengine - INFO - Epoch(train) [55][200/3757] lr: 1.0000e-03 eta: 7:36:09 time: 0.1587 data_time: 0.0104 memory: 7124 grad_norm: 5.7813 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.2036 loss: 1.2036 2022/09/07 08:38:08 - mmengine - INFO - Epoch(train) [55][300/3757] lr: 1.0000e-03 eta: 7:35:53 time: 0.1583 data_time: 0.0110 memory: 7124 grad_norm: 5.6506 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.1979 loss: 1.1979 2022/09/07 08:38:24 - mmengine - INFO - Epoch(train) [55][400/3757] lr: 1.0000e-03 eta: 7:35:37 time: 0.1563 data_time: 0.0104 memory: 7124 grad_norm: 5.6026 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.5317 loss: 1.5317 2022/09/07 08:38:40 - mmengine - INFO - Epoch(train) [55][500/3757] lr: 1.0000e-03 eta: 7:35:21 time: 0.1598 data_time: 0.0093 memory: 7124 grad_norm: 5.7058 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3077 loss: 1.3077 2022/09/07 08:38:56 - mmengine - INFO - Epoch(train) [55][600/3757] lr: 1.0000e-03 eta: 7:35:05 time: 0.1566 data_time: 0.0120 memory: 7124 grad_norm: 5.9010 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.3575 loss: 1.3575 2022/09/07 08:39:12 - mmengine - INFO - Epoch(train) [55][700/3757] lr: 1.0000e-03 eta: 7:34:49 time: 0.1571 data_time: 0.0110 memory: 7124 grad_norm: 5.6496 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.2813 loss: 1.2813 2022/09/07 08:39:28 - mmengine - INFO - Epoch(train) [55][800/3757] lr: 1.0000e-03 eta: 7:34:34 time: 0.1608 data_time: 0.0103 memory: 7124 grad_norm: 5.8373 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2226 loss: 1.2226 2022/09/07 08:39:44 - mmengine - INFO - Epoch(train) [55][900/3757] lr: 1.0000e-03 eta: 7:34:18 time: 0.1582 data_time: 0.0108 memory: 7124 grad_norm: 5.6582 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2450 loss: 1.2450 2022/09/07 08:40:00 - mmengine - INFO - Epoch(train) [55][1000/3757] lr: 1.0000e-03 eta: 7:34:02 time: 0.1566 data_time: 0.0099 memory: 7124 grad_norm: 5.8747 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.3262 loss: 1.3262 2022/09/07 08:40:16 - mmengine - INFO - Epoch(train) [55][1100/3757] lr: 1.0000e-03 eta: 7:33:46 time: 0.1614 data_time: 0.0114 memory: 7124 grad_norm: 5.6745 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3838 loss: 1.3838 2022/09/07 08:40:20 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:40:32 - mmengine - INFO - Epoch(train) [55][1200/3757] lr: 1.0000e-03 eta: 7:33:31 time: 0.1575 data_time: 0.0110 memory: 7124 grad_norm: 5.6500 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1638 loss: 1.1638 2022/09/07 08:40:48 - mmengine - INFO - Epoch(train) [55][1300/3757] lr: 1.0000e-03 eta: 7:33:15 time: 0.1644 data_time: 0.0108 memory: 7124 grad_norm: 5.8153 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3482 loss: 1.3482 2022/09/07 08:41:04 - mmengine - INFO - Epoch(train) [55][1400/3757] lr: 1.0000e-03 eta: 7:32:59 time: 0.1578 data_time: 0.0113 memory: 7124 grad_norm: 5.6231 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2924 loss: 1.2924 2022/09/07 08:41:20 - mmengine - INFO - Epoch(train) [55][1500/3757] lr: 1.0000e-03 eta: 7:32:43 time: 0.1587 data_time: 0.0111 memory: 7124 grad_norm: 5.7880 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.5954 loss: 1.5954 2022/09/07 08:41:36 - mmengine - INFO - Epoch(train) [55][1600/3757] lr: 1.0000e-03 eta: 7:32:27 time: 0.1565 data_time: 0.0123 memory: 7124 grad_norm: 5.8431 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1791 loss: 1.1791 2022/09/07 08:41:52 - mmengine - INFO - Epoch(train) [55][1700/3757] lr: 1.0000e-03 eta: 7:32:11 time: 0.1582 data_time: 0.0103 memory: 7124 grad_norm: 5.8879 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.4715 loss: 1.4715 2022/09/07 08:42:08 - mmengine - INFO - Epoch(train) [55][1800/3757] lr: 1.0000e-03 eta: 7:31:55 time: 0.1564 data_time: 0.0101 memory: 7124 grad_norm: 5.8771 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1848 loss: 1.1848 2022/09/07 08:42:24 - mmengine - INFO - Epoch(train) [55][1900/3757] lr: 1.0000e-03 eta: 7:31:40 time: 0.1605 data_time: 0.0109 memory: 7124 grad_norm: 5.6698 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8789 loss: 0.8789 2022/09/07 08:42:40 - mmengine - INFO - Epoch(train) [55][2000/3757] lr: 1.0000e-03 eta: 7:31:24 time: 0.1555 data_time: 0.0111 memory: 7124 grad_norm: 5.7058 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2413 loss: 1.2413 2022/09/07 08:42:55 - mmengine - INFO - Epoch(train) [55][2100/3757] lr: 1.0000e-03 eta: 7:31:08 time: 0.1567 data_time: 0.0104 memory: 7124 grad_norm: 5.6506 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.1669 loss: 1.1669 2022/09/07 08:42:59 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:43:11 - mmengine - INFO - Epoch(train) [55][2200/3757] lr: 1.0000e-03 eta: 7:30:52 time: 0.1568 data_time: 0.0108 memory: 7124 grad_norm: 5.6662 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1146 loss: 1.1146 2022/09/07 08:43:28 - mmengine - INFO - Epoch(train) [55][2300/3757] lr: 1.0000e-03 eta: 7:30:36 time: 0.1585 data_time: 0.0116 memory: 7124 grad_norm: 5.6739 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1155 loss: 1.1155 2022/09/07 08:43:43 - mmengine - INFO - Epoch(train) [55][2400/3757] lr: 1.0000e-03 eta: 7:30:20 time: 0.1583 data_time: 0.0108 memory: 7124 grad_norm: 5.5499 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.3861 loss: 1.3861 2022/09/07 08:43:59 - mmengine - INFO - Epoch(train) [55][2500/3757] lr: 1.0000e-03 eta: 7:30:05 time: 0.1628 data_time: 0.0117 memory: 7124 grad_norm: 5.6567 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.2901 loss: 1.2901 2022/09/07 08:44:15 - mmengine - INFO - Epoch(train) [55][2600/3757] lr: 1.0000e-03 eta: 7:29:49 time: 0.1556 data_time: 0.0105 memory: 7124 grad_norm: 5.7965 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2950 loss: 1.2950 2022/09/07 08:44:31 - mmengine - INFO - Epoch(train) [55][2700/3757] lr: 1.0000e-03 eta: 7:29:33 time: 0.1580 data_time: 0.0124 memory: 7124 grad_norm: 5.8164 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.4255 loss: 1.4255 2022/09/07 08:44:48 - mmengine - INFO - Epoch(train) [55][2800/3757] lr: 1.0000e-03 eta: 7:29:17 time: 0.1617 data_time: 0.0117 memory: 7124 grad_norm: 6.0258 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3676 loss: 1.3676 2022/09/07 08:45:04 - mmengine - INFO - Epoch(train) [55][2900/3757] lr: 1.0000e-03 eta: 7:29:02 time: 0.1580 data_time: 0.0112 memory: 7124 grad_norm: 5.8551 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1990 loss: 1.1990 2022/09/07 08:45:20 - mmengine - INFO - Epoch(train) [55][3000/3757] lr: 1.0000e-03 eta: 7:28:46 time: 0.1609 data_time: 0.0113 memory: 7124 grad_norm: 5.6452 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.4684 loss: 1.4684 2022/09/07 08:45:36 - mmengine - INFO - Epoch(train) [55][3100/3757] lr: 1.0000e-03 eta: 7:28:30 time: 0.1600 data_time: 0.0106 memory: 7124 grad_norm: 5.7240 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.4368 loss: 1.4368 2022/09/07 08:45:39 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:45:52 - mmengine - INFO - Epoch(train) [55][3200/3757] lr: 1.0000e-03 eta: 7:28:14 time: 0.1550 data_time: 0.0105 memory: 7124 grad_norm: 5.7622 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2915 loss: 1.2915 2022/09/07 08:46:08 - mmengine - INFO - Epoch(train) [55][3300/3757] lr: 1.0000e-03 eta: 7:27:58 time: 0.1617 data_time: 0.0109 memory: 7124 grad_norm: 5.8597 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.4994 loss: 1.4994 2022/09/07 08:46:24 - mmengine - INFO - Epoch(train) [55][3400/3757] lr: 1.0000e-03 eta: 7:27:43 time: 0.1583 data_time: 0.0107 memory: 7124 grad_norm: 5.9578 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1262 loss: 1.1262 2022/09/07 08:46:40 - mmengine - INFO - Epoch(train) [55][3500/3757] lr: 1.0000e-03 eta: 7:27:27 time: 0.1587 data_time: 0.0102 memory: 7124 grad_norm: 5.6938 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1182 loss: 1.1182 2022/09/07 08:46:55 - mmengine - INFO - Epoch(train) [55][3600/3757] lr: 1.0000e-03 eta: 7:27:11 time: 0.1554 data_time: 0.0109 memory: 7124 grad_norm: 5.7074 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.2066 loss: 1.2066 2022/09/07 08:47:12 - mmengine - INFO - Epoch(train) [55][3700/3757] lr: 1.0000e-03 eta: 7:26:55 time: 0.1610 data_time: 0.0127 memory: 7124 grad_norm: 5.8006 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2174 loss: 1.2174 2022/09/07 08:47:20 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:47:20 - mmengine - INFO - Epoch(train) [55][3757/3757] lr: 1.0000e-03 eta: 7:26:49 time: 0.1386 data_time: 0.0074 memory: 7124 grad_norm: 5.9332 top1_acc: 0.7143 top5_acc: 1.0000 loss_cls: 1.2679 loss: 1.2679 2022/09/07 08:47:20 - mmengine - INFO - Saving checkpoint at 55 epochs 2022/09/07 08:49:40 - mmengine - INFO - Epoch(val) [55][100/310] eta: 0:04:13 time: 1.2048 data_time: 0.8975 memory: 7627 2022/09/07 08:51:57 - mmengine - INFO - Epoch(val) [55][200/310] eta: 0:02:19 time: 1.2648 data_time: 0.9554 memory: 7627 2022/09/07 08:54:02 - mmengine - INFO - Epoch(val) [55][300/310] eta: 0:00:11 time: 1.1166 data_time: 0.8137 memory: 7627 2022/09/07 08:54:19 - mmengine - INFO - Epoch(val) [55][310/310] acc/top1: 0.7342 acc/top5: 0.9117 acc/mean1: 0.7342 2022/09/07 08:54:37 - mmengine - INFO - Epoch(train) [56][100/3757] lr: 1.0000e-03 eta: 7:26:29 time: 0.1612 data_time: 0.0113 memory: 7627 grad_norm: 5.9116 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3599 loss: 1.3599 2022/09/07 08:54:53 - mmengine - INFO - Epoch(train) [56][200/3757] lr: 1.0000e-03 eta: 7:26:14 time: 0.1576 data_time: 0.0114 memory: 7124 grad_norm: 5.5734 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0836 loss: 1.0836 2022/09/07 08:55:09 - mmengine - INFO - Epoch(train) [56][300/3757] lr: 1.0000e-03 eta: 7:25:58 time: 0.1583 data_time: 0.0116 memory: 7124 grad_norm: 5.9507 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2577 loss: 1.2577 2022/09/07 08:55:19 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:55:25 - mmengine - INFO - Epoch(train) [56][400/3757] lr: 1.0000e-03 eta: 7:25:42 time: 0.1563 data_time: 0.0110 memory: 7124 grad_norm: 5.7482 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2623 loss: 1.2623 2022/09/07 08:55:41 - mmengine - INFO - Epoch(train) [56][500/3757] lr: 1.0000e-03 eta: 7:25:26 time: 0.1609 data_time: 0.0130 memory: 7124 grad_norm: 5.9584 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5303 loss: 1.5303 2022/09/07 08:55:57 - mmengine - INFO - Epoch(train) [56][600/3757] lr: 1.0000e-03 eta: 7:25:10 time: 0.1561 data_time: 0.0109 memory: 7124 grad_norm: 5.7764 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1890 loss: 1.1890 2022/09/07 08:56:13 - mmengine - INFO - Epoch(train) [56][700/3757] lr: 1.0000e-03 eta: 7:24:54 time: 0.1602 data_time: 0.0102 memory: 7124 grad_norm: 5.5457 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2437 loss: 1.2437 2022/09/07 08:56:29 - mmengine - INFO - Epoch(train) [56][800/3757] lr: 1.0000e-03 eta: 7:24:38 time: 0.1576 data_time: 0.0105 memory: 7124 grad_norm: 5.8173 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.4366 loss: 1.4366 2022/09/07 08:56:45 - mmengine - INFO - Epoch(train) [56][900/3757] lr: 1.0000e-03 eta: 7:24:23 time: 0.1602 data_time: 0.0121 memory: 7124 grad_norm: 5.6351 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1311 loss: 1.1311 2022/09/07 08:57:01 - mmengine - INFO - Epoch(train) [56][1000/3757] lr: 1.0000e-03 eta: 7:24:07 time: 0.1581 data_time: 0.0087 memory: 7124 grad_norm: 5.6836 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1220 loss: 1.1220 2022/09/07 08:57:17 - mmengine - INFO - Epoch(train) [56][1100/3757] lr: 1.0000e-03 eta: 7:23:51 time: 0.1576 data_time: 0.0105 memory: 7124 grad_norm: 6.0810 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2379 loss: 1.2379 2022/09/07 08:57:33 - mmengine - INFO - Epoch(train) [56][1200/3757] lr: 1.0000e-03 eta: 7:23:35 time: 0.1559 data_time: 0.0113 memory: 7124 grad_norm: 5.7202 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.4647 loss: 1.4647 2022/09/07 08:57:49 - mmengine - INFO - Epoch(train) [56][1300/3757] lr: 1.0000e-03 eta: 7:23:20 time: 0.1594 data_time: 0.0108 memory: 7124 grad_norm: 5.8773 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5741 loss: 1.5741 2022/09/07 08:57:59 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 08:58:05 - mmengine - INFO - Epoch(train) [56][1400/3757] lr: 1.0000e-03 eta: 7:23:04 time: 0.1587 data_time: 0.0113 memory: 7124 grad_norm: 5.8127 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1489 loss: 1.1489 2022/09/07 08:58:21 - mmengine - INFO - Epoch(train) [56][1500/3757] lr: 1.0000e-03 eta: 7:22:48 time: 0.1588 data_time: 0.0103 memory: 7124 grad_norm: 5.7340 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.0515 loss: 1.0515 2022/09/07 08:58:37 - mmengine - INFO - Epoch(train) [56][1600/3757] lr: 1.0000e-03 eta: 7:22:32 time: 0.1572 data_time: 0.0122 memory: 7124 grad_norm: 5.7249 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2051 loss: 1.2051 2022/09/07 08:58:53 - mmengine - INFO - Epoch(train) [56][1700/3757] lr: 1.0000e-03 eta: 7:22:16 time: 0.1550 data_time: 0.0116 memory: 7124 grad_norm: 5.8249 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0962 loss: 1.0962 2022/09/07 08:59:08 - mmengine - INFO - Epoch(train) [56][1800/3757] lr: 1.0000e-03 eta: 7:22:00 time: 0.1556 data_time: 0.0110 memory: 7124 grad_norm: 5.7510 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.3292 loss: 1.3292 2022/09/07 08:59:25 - mmengine - INFO - Epoch(train) [56][1900/3757] lr: 1.0000e-03 eta: 7:21:45 time: 0.1588 data_time: 0.0119 memory: 7124 grad_norm: 5.8410 top1_acc: 0.2500 top5_acc: 0.8750 loss_cls: 1.0743 loss: 1.0743 2022/09/07 08:59:40 - mmengine - INFO - Epoch(train) [56][2000/3757] lr: 1.0000e-03 eta: 7:21:29 time: 0.1569 data_time: 0.0109 memory: 7124 grad_norm: 5.5688 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3176 loss: 1.3176 2022/09/07 08:59:56 - mmengine - INFO - Epoch(train) [56][2100/3757] lr: 1.0000e-03 eta: 7:21:13 time: 0.1584 data_time: 0.0108 memory: 7124 grad_norm: 5.9844 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3635 loss: 1.3635 2022/09/07 09:00:12 - mmengine - INFO - Epoch(train) [56][2200/3757] lr: 1.0000e-03 eta: 7:20:57 time: 0.1561 data_time: 0.0109 memory: 7124 grad_norm: 5.5024 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.3572 loss: 1.3572 2022/09/07 09:00:28 - mmengine - INFO - Epoch(train) [56][2300/3757] lr: 1.0000e-03 eta: 7:20:41 time: 0.1576 data_time: 0.0107 memory: 7124 grad_norm: 5.7674 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1787 loss: 1.1787 2022/09/07 09:00:39 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:00:44 - mmengine - INFO - Epoch(train) [56][2400/3757] lr: 1.0000e-03 eta: 7:20:25 time: 0.1574 data_time: 0.0119 memory: 7124 grad_norm: 5.6806 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.4105 loss: 1.4105 2022/09/07 09:01:00 - mmengine - INFO - Epoch(train) [56][2500/3757] lr: 1.0000e-03 eta: 7:20:10 time: 0.1588 data_time: 0.0111 memory: 7124 grad_norm: 5.7986 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2215 loss: 1.2215 2022/09/07 09:01:16 - mmengine - INFO - Epoch(train) [56][2600/3757] lr: 1.0000e-03 eta: 7:19:54 time: 0.1568 data_time: 0.0107 memory: 7124 grad_norm: 5.9009 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.3166 loss: 1.3166 2022/09/07 09:01:32 - mmengine - INFO - Epoch(train) [56][2700/3757] lr: 1.0000e-03 eta: 7:19:38 time: 0.1556 data_time: 0.0097 memory: 7124 grad_norm: 5.9745 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.2369 loss: 1.2369 2022/09/07 09:01:49 - mmengine - INFO - Epoch(train) [56][2800/3757] lr: 1.0000e-03 eta: 7:19:22 time: 0.1633 data_time: 0.0109 memory: 7124 grad_norm: 5.7732 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2346 loss: 1.2346 2022/09/07 09:02:05 - mmengine - INFO - Epoch(train) [56][2900/3757] lr: 1.0000e-03 eta: 7:19:06 time: 0.1566 data_time: 0.0103 memory: 7124 grad_norm: 6.0171 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3157 loss: 1.3157 2022/09/07 09:02:20 - mmengine - INFO - Epoch(train) [56][3000/3757] lr: 1.0000e-03 eta: 7:18:51 time: 0.1630 data_time: 0.0136 memory: 7124 grad_norm: 5.8985 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1306 loss: 1.1306 2022/09/07 09:02:36 - mmengine - INFO - Epoch(train) [56][3100/3757] lr: 1.0000e-03 eta: 7:18:35 time: 0.1573 data_time: 0.0113 memory: 7124 grad_norm: 5.7851 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2401 loss: 1.2401 2022/09/07 09:02:52 - mmengine - INFO - Epoch(train) [56][3200/3757] lr: 1.0000e-03 eta: 7:18:19 time: 0.1590 data_time: 0.0106 memory: 7124 grad_norm: 5.7515 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2122 loss: 1.2122 2022/09/07 09:03:09 - mmengine - INFO - Epoch(train) [56][3300/3757] lr: 1.0000e-03 eta: 7:18:03 time: 0.1593 data_time: 0.0114 memory: 7124 grad_norm: 5.9200 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1738 loss: 1.1738 2022/09/07 09:03:19 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:03:24 - mmengine - INFO - Epoch(train) [56][3400/3757] lr: 1.0000e-03 eta: 7:17:47 time: 0.1600 data_time: 0.0107 memory: 7124 grad_norm: 5.7085 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0946 loss: 1.0946 2022/09/07 09:03:40 - mmengine - INFO - Epoch(train) [56][3500/3757] lr: 1.0000e-03 eta: 7:17:31 time: 0.1602 data_time: 0.0095 memory: 7124 grad_norm: 6.2037 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5517 loss: 1.5517 2022/09/07 09:03:56 - mmengine - INFO - Epoch(train) [56][3600/3757] lr: 1.0000e-03 eta: 7:17:16 time: 0.1585 data_time: 0.0108 memory: 7124 grad_norm: 5.7478 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1292 loss: 1.1292 2022/09/07 09:04:12 - mmengine - INFO - Epoch(train) [56][3700/3757] lr: 1.0000e-03 eta: 7:17:00 time: 0.1569 data_time: 0.0123 memory: 7124 grad_norm: 5.7867 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.4046 loss: 1.4046 2022/09/07 09:04:21 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:04:21 - mmengine - INFO - Epoch(train) [56][3757/3757] lr: 1.0000e-03 eta: 7:16:54 time: 0.1355 data_time: 0.0081 memory: 7124 grad_norm: 5.8206 top1_acc: 0.7143 top5_acc: 0.8571 loss_cls: 1.2834 loss: 1.2834 2022/09/07 09:04:21 - mmengine - INFO - Saving checkpoint at 56 epochs 2022/09/07 09:06:40 - mmengine - INFO - Epoch(val) [56][100/310] eta: 0:04:05 time: 1.1672 data_time: 0.8636 memory: 7627 2022/09/07 09:08:57 - mmengine - INFO - Epoch(val) [56][200/310] eta: 0:02:24 time: 1.3115 data_time: 1.0069 memory: 7627 2022/09/07 09:11:02 - mmengine - INFO - Epoch(val) [56][300/310] eta: 0:00:11 time: 1.1130 data_time: 0.8098 memory: 7627 2022/09/07 09:11:18 - mmengine - INFO - Epoch(val) [56][310/310] acc/top1: 0.7378 acc/top5: 0.9121 acc/mean1: 0.7377 2022/09/07 09:11:18 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_54.pth is removed 2022/09/07 09:11:20 - mmengine - INFO - The best checkpoint with 0.7378 acc/top1 at 56 epoch is saved to best_acc/top1_epoch_56.pth. 2022/09/07 09:11:37 - mmengine - INFO - Epoch(train) [57][100/3757] lr: 1.0000e-03 eta: 7:16:34 time: 0.1608 data_time: 0.0112 memory: 7627 grad_norm: 5.6355 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.2842 loss: 1.2842 2022/09/07 09:11:53 - mmengine - INFO - Epoch(train) [57][200/3757] lr: 1.0000e-03 eta: 7:16:18 time: 0.1596 data_time: 0.0115 memory: 7124 grad_norm: 5.9221 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2132 loss: 1.2132 2022/09/07 09:12:09 - mmengine - INFO - Epoch(train) [57][300/3757] lr: 1.0000e-03 eta: 7:16:02 time: 0.1670 data_time: 0.0117 memory: 7124 grad_norm: 5.9211 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.0906 loss: 1.0906 2022/09/07 09:12:25 - mmengine - INFO - Epoch(train) [57][400/3757] lr: 1.0000e-03 eta: 7:15:46 time: 0.1598 data_time: 0.0112 memory: 7124 grad_norm: 5.9007 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1452 loss: 1.1452 2022/09/07 09:12:41 - mmengine - INFO - Epoch(train) [57][500/3757] lr: 1.0000e-03 eta: 7:15:30 time: 0.1617 data_time: 0.0122 memory: 7124 grad_norm: 5.8406 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.4807 loss: 1.4807 2022/09/07 09:12:57 - mmengine - INFO - Epoch(train) [57][600/3757] lr: 1.0000e-03 eta: 7:15:15 time: 0.1583 data_time: 0.0107 memory: 7124 grad_norm: 6.0264 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2425 loss: 1.2425 2022/09/07 09:12:58 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:13:13 - mmengine - INFO - Epoch(train) [57][700/3757] lr: 1.0000e-03 eta: 7:14:59 time: 0.1636 data_time: 0.0110 memory: 7124 grad_norm: 5.8893 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1342 loss: 1.1342 2022/09/07 09:13:29 - mmengine - INFO - Epoch(train) [57][800/3757] lr: 1.0000e-03 eta: 7:14:43 time: 0.1562 data_time: 0.0121 memory: 7124 grad_norm: 5.8782 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.3574 loss: 1.3574 2022/09/07 09:13:45 - mmengine - INFO - Epoch(train) [57][900/3757] lr: 1.0000e-03 eta: 7:14:27 time: 0.1560 data_time: 0.0099 memory: 7124 grad_norm: 5.9779 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1747 loss: 1.1747 2022/09/07 09:14:01 - mmengine - INFO - Epoch(train) [57][1000/3757] lr: 1.0000e-03 eta: 7:14:11 time: 0.1596 data_time: 0.0114 memory: 7124 grad_norm: 6.1072 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1388 loss: 1.1388 2022/09/07 09:14:17 - mmengine - INFO - Epoch(train) [57][1100/3757] lr: 1.0000e-03 eta: 7:13:55 time: 0.1571 data_time: 0.0113 memory: 7124 grad_norm: 5.9927 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.3025 loss: 1.3025 2022/09/07 09:14:33 - mmengine - INFO - Epoch(train) [57][1200/3757] lr: 1.0000e-03 eta: 7:13:40 time: 0.1585 data_time: 0.0132 memory: 7124 grad_norm: 5.8940 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1461 loss: 1.1461 2022/09/07 09:14:49 - mmengine - INFO - Epoch(train) [57][1300/3757] lr: 1.0000e-03 eta: 7:13:24 time: 0.1585 data_time: 0.0124 memory: 7124 grad_norm: 5.7648 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2395 loss: 1.2395 2022/09/07 09:15:05 - mmengine - INFO - Epoch(train) [57][1400/3757] lr: 1.0000e-03 eta: 7:13:08 time: 0.1585 data_time: 0.0112 memory: 7124 grad_norm: 6.2227 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3838 loss: 1.3838 2022/09/07 09:15:20 - mmengine - INFO - Epoch(train) [57][1500/3757] lr: 1.0000e-03 eta: 7:12:52 time: 0.1609 data_time: 0.0119 memory: 7124 grad_norm: 5.9398 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2346 loss: 1.2346 2022/09/07 09:15:36 - mmengine - INFO - Epoch(train) [57][1600/3757] lr: 1.0000e-03 eta: 7:12:36 time: 0.1577 data_time: 0.0111 memory: 7124 grad_norm: 5.7235 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1849 loss: 1.1849 2022/09/07 09:15:38 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:15:52 - mmengine - INFO - Epoch(train) [57][1700/3757] lr: 1.0000e-03 eta: 7:12:21 time: 0.1590 data_time: 0.0102 memory: 7124 grad_norm: 6.1806 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2811 loss: 1.2811 2022/09/07 09:16:08 - mmengine - INFO - Epoch(train) [57][1800/3757] lr: 1.0000e-03 eta: 7:12:05 time: 0.1578 data_time: 0.0104 memory: 7124 grad_norm: 5.8888 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.1709 loss: 1.1709 2022/09/07 09:16:25 - mmengine - INFO - Epoch(train) [57][1900/3757] lr: 1.0000e-03 eta: 7:11:49 time: 0.1630 data_time: 0.0101 memory: 7124 grad_norm: 6.0355 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3992 loss: 1.3992 2022/09/07 09:16:41 - mmengine - INFO - Epoch(train) [57][2000/3757] lr: 1.0000e-03 eta: 7:11:33 time: 0.1666 data_time: 0.0105 memory: 7124 grad_norm: 5.8410 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1558 loss: 1.1558 2022/09/07 09:16:57 - mmengine - INFO - Epoch(train) [57][2100/3757] lr: 1.0000e-03 eta: 7:11:17 time: 0.1557 data_time: 0.0102 memory: 7124 grad_norm: 5.7978 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.1952 loss: 1.1952 2022/09/07 09:17:13 - mmengine - INFO - Epoch(train) [57][2200/3757] lr: 1.0000e-03 eta: 7:11:02 time: 0.1563 data_time: 0.0114 memory: 7124 grad_norm: 5.7784 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2688 loss: 1.2688 2022/09/07 09:17:29 - mmengine - INFO - Epoch(train) [57][2300/3757] lr: 1.0000e-03 eta: 7:10:46 time: 0.1549 data_time: 0.0110 memory: 7124 grad_norm: 6.0390 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0635 loss: 1.0635 2022/09/07 09:17:45 - mmengine - INFO - Epoch(train) [57][2400/3757] lr: 1.0000e-03 eta: 7:10:30 time: 0.1552 data_time: 0.0105 memory: 7124 grad_norm: 5.8015 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0105 loss: 1.0105 2022/09/07 09:18:01 - mmengine - INFO - Epoch(train) [57][2500/3757] lr: 1.0000e-03 eta: 7:10:14 time: 0.1590 data_time: 0.0104 memory: 7124 grad_norm: 5.7047 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9686 loss: 0.9686 2022/09/07 09:18:17 - mmengine - INFO - Epoch(train) [57][2600/3757] lr: 1.0000e-03 eta: 7:09:58 time: 0.1611 data_time: 0.0118 memory: 7124 grad_norm: 5.8428 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3107 loss: 1.3107 2022/09/07 09:18:18 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:18:33 - mmengine - INFO - Epoch(train) [57][2700/3757] lr: 1.0000e-03 eta: 7:09:43 time: 0.1617 data_time: 0.0110 memory: 7124 grad_norm: 5.8620 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.4026 loss: 1.4026 2022/09/07 09:18:49 - mmengine - INFO - Epoch(train) [57][2800/3757] lr: 1.0000e-03 eta: 7:09:27 time: 0.1586 data_time: 0.0107 memory: 7124 grad_norm: 6.1622 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2992 loss: 1.2992 2022/09/07 09:19:05 - mmengine - INFO - Epoch(train) [57][2900/3757] lr: 1.0000e-03 eta: 7:09:11 time: 0.1614 data_time: 0.0093 memory: 7124 grad_norm: 6.0405 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4457 loss: 1.4457 2022/09/07 09:19:21 - mmengine - INFO - Epoch(train) [57][3000/3757] lr: 1.0000e-03 eta: 7:08:55 time: 0.1644 data_time: 0.0111 memory: 7124 grad_norm: 5.8755 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2119 loss: 1.2119 2022/09/07 09:19:37 - mmengine - INFO - Epoch(train) [57][3100/3757] lr: 1.0000e-03 eta: 7:08:39 time: 0.1572 data_time: 0.0110 memory: 7124 grad_norm: 5.7918 top1_acc: 0.3750 top5_acc: 1.0000 loss_cls: 1.2147 loss: 1.2147 2022/09/07 09:19:52 - mmengine - INFO - Epoch(train) [57][3200/3757] lr: 1.0000e-03 eta: 7:08:23 time: 0.1573 data_time: 0.0119 memory: 7124 grad_norm: 5.8863 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1570 loss: 1.1570 2022/09/07 09:20:08 - mmengine - INFO - Epoch(train) [57][3300/3757] lr: 1.0000e-03 eta: 7:08:08 time: 0.1573 data_time: 0.0116 memory: 7124 grad_norm: 6.0033 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2343 loss: 1.2343 2022/09/07 09:20:24 - mmengine - INFO - Epoch(train) [57][3400/3757] lr: 1.0000e-03 eta: 7:07:52 time: 0.1578 data_time: 0.0108 memory: 7124 grad_norm: 5.8351 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.4119 loss: 1.4119 2022/09/07 09:20:40 - mmengine - INFO - Epoch(train) [57][3500/3757] lr: 1.0000e-03 eta: 7:07:36 time: 0.1562 data_time: 0.0113 memory: 7124 grad_norm: 5.9605 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3552 loss: 1.3552 2022/09/07 09:20:56 - mmengine - INFO - Epoch(train) [57][3600/3757] lr: 1.0000e-03 eta: 7:07:20 time: 0.1577 data_time: 0.0111 memory: 7124 grad_norm: 6.0300 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.2723 loss: 1.2723 2022/09/07 09:20:58 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:21:12 - mmengine - INFO - Epoch(train) [57][3700/3757] lr: 1.0000e-03 eta: 7:07:04 time: 0.1593 data_time: 0.0100 memory: 7124 grad_norm: 5.9688 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2756 loss: 1.2756 2022/09/07 09:21:21 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:21:21 - mmengine - INFO - Epoch(train) [57][3757/3757] lr: 1.0000e-03 eta: 7:06:58 time: 0.1390 data_time: 0.0079 memory: 7124 grad_norm: 5.9245 top1_acc: 0.5714 top5_acc: 0.7143 loss_cls: 1.2524 loss: 1.2524 2022/09/07 09:21:21 - mmengine - INFO - Saving checkpoint at 57 epochs 2022/09/07 09:23:39 - mmengine - INFO - Epoch(val) [57][100/310] eta: 0:03:55 time: 1.1226 data_time: 0.8199 memory: 7627 2022/09/07 09:25:56 - mmengine - INFO - Epoch(val) [57][200/310] eta: 0:02:26 time: 1.3355 data_time: 1.0334 memory: 7627 2022/09/07 09:28:00 - mmengine - INFO - Epoch(val) [57][300/310] eta: 0:00:11 time: 1.1066 data_time: 0.8098 memory: 7627 2022/09/07 09:28:17 - mmengine - INFO - Epoch(val) [57][310/310] acc/top1: 0.7409 acc/top5: 0.9137 acc/mean1: 0.7408 2022/09/07 09:28:17 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_56.pth is removed 2022/09/07 09:28:20 - mmengine - INFO - The best checkpoint with 0.7409 acc/top1 at 57 epoch is saved to best_acc/top1_epoch_57.pth. 2022/09/07 09:28:37 - mmengine - INFO - Epoch(train) [58][100/3757] lr: 1.0000e-03 eta: 7:06:38 time: 0.1573 data_time: 0.0102 memory: 7627 grad_norm: 6.0374 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2662 loss: 1.2662 2022/09/07 09:28:53 - mmengine - INFO - Epoch(train) [58][200/3757] lr: 1.0000e-03 eta: 7:06:22 time: 0.1615 data_time: 0.0096 memory: 7124 grad_norm: 5.8524 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0452 loss: 1.0452 2022/09/07 09:29:09 - mmengine - INFO - Epoch(train) [58][300/3757] lr: 1.0000e-03 eta: 7:06:06 time: 0.1638 data_time: 0.0114 memory: 7124 grad_norm: 6.0015 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2425 loss: 1.2425 2022/09/07 09:29:25 - mmengine - INFO - Epoch(train) [58][400/3757] lr: 1.0000e-03 eta: 7:05:50 time: 0.1577 data_time: 0.0105 memory: 7124 grad_norm: 5.8912 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1157 loss: 1.1157 2022/09/07 09:29:41 - mmengine - INFO - Epoch(train) [58][500/3757] lr: 1.0000e-03 eta: 7:05:34 time: 0.1572 data_time: 0.0112 memory: 7124 grad_norm: 6.1233 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3330 loss: 1.3330 2022/09/07 09:29:57 - mmengine - INFO - Epoch(train) [58][600/3757] lr: 1.0000e-03 eta: 7:05:18 time: 0.1545 data_time: 0.0093 memory: 7124 grad_norm: 6.0086 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.0571 loss: 1.0571 2022/09/07 09:30:13 - mmengine - INFO - Epoch(train) [58][700/3757] lr: 1.0000e-03 eta: 7:05:03 time: 0.1556 data_time: 0.0102 memory: 7124 grad_norm: 6.1151 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4255 loss: 1.4255 2022/09/07 09:30:28 - mmengine - INFO - Epoch(train) [58][800/3757] lr: 1.0000e-03 eta: 7:04:47 time: 0.1607 data_time: 0.0107 memory: 7124 grad_norm: 5.7340 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1768 loss: 1.1768 2022/09/07 09:30:37 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:30:45 - mmengine - INFO - Epoch(train) [58][900/3757] lr: 1.0000e-03 eta: 7:04:31 time: 0.1567 data_time: 0.0098 memory: 7124 grad_norm: 5.8452 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3031 loss: 1.3031 2022/09/07 09:31:00 - mmengine - INFO - Epoch(train) [58][1000/3757] lr: 1.0000e-03 eta: 7:04:15 time: 0.1591 data_time: 0.0103 memory: 7124 grad_norm: 5.7865 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1127 loss: 1.1127 2022/09/07 09:31:16 - mmengine - INFO - Epoch(train) [58][1100/3757] lr: 1.0000e-03 eta: 7:03:59 time: 0.1548 data_time: 0.0100 memory: 7124 grad_norm: 5.8848 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2966 loss: 1.2966 2022/09/07 09:31:32 - mmengine - INFO - Epoch(train) [58][1200/3757] lr: 1.0000e-03 eta: 7:03:44 time: 0.1552 data_time: 0.0115 memory: 7124 grad_norm: 5.9290 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.3293 loss: 1.3293 2022/09/07 09:31:48 - mmengine - INFO - Epoch(train) [58][1300/3757] lr: 1.0000e-03 eta: 7:03:28 time: 0.1591 data_time: 0.0109 memory: 7124 grad_norm: 6.0153 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2693 loss: 1.2693 2022/09/07 09:32:04 - mmengine - INFO - Epoch(train) [58][1400/3757] lr: 1.0000e-03 eta: 7:03:12 time: 0.1592 data_time: 0.0090 memory: 7124 grad_norm: 5.9323 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1041 loss: 1.1041 2022/09/07 09:32:20 - mmengine - INFO - Epoch(train) [58][1500/3757] lr: 1.0000e-03 eta: 7:02:56 time: 0.1607 data_time: 0.0123 memory: 7124 grad_norm: 6.1252 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1026 loss: 1.1026 2022/09/07 09:32:36 - mmengine - INFO - Epoch(train) [58][1600/3757] lr: 1.0000e-03 eta: 7:02:40 time: 0.1672 data_time: 0.0098 memory: 7124 grad_norm: 5.5797 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.2288 loss: 1.2288 2022/09/07 09:32:52 - mmengine - INFO - Epoch(train) [58][1700/3757] lr: 1.0000e-03 eta: 7:02:24 time: 0.1581 data_time: 0.0093 memory: 7124 grad_norm: 6.0979 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.3380 loss: 1.3380 2022/09/07 09:33:08 - mmengine - INFO - Epoch(train) [58][1800/3757] lr: 1.0000e-03 eta: 7:02:09 time: 0.1567 data_time: 0.0093 memory: 7124 grad_norm: 5.9876 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0058 loss: 1.0058 2022/09/07 09:33:16 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:33:24 - mmengine - INFO - Epoch(train) [58][1900/3757] lr: 1.0000e-03 eta: 7:01:53 time: 0.1588 data_time: 0.0124 memory: 7124 grad_norm: 6.0619 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2926 loss: 1.2926 2022/09/07 09:33:40 - mmengine - INFO - Epoch(train) [58][2000/3757] lr: 1.0000e-03 eta: 7:01:37 time: 0.1572 data_time: 0.0096 memory: 7124 grad_norm: 5.8836 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1128 loss: 1.1128 2022/09/07 09:33:56 - mmengine - INFO - Epoch(train) [58][2100/3757] lr: 1.0000e-03 eta: 7:01:21 time: 0.1569 data_time: 0.0128 memory: 7124 grad_norm: 6.0577 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1446 loss: 1.1446 2022/09/07 09:34:12 - mmengine - INFO - Epoch(train) [58][2200/3757] lr: 1.0000e-03 eta: 7:01:05 time: 0.1575 data_time: 0.0101 memory: 7124 grad_norm: 6.0168 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1751 loss: 1.1751 2022/09/07 09:34:28 - mmengine - INFO - Epoch(train) [58][2300/3757] lr: 1.0000e-03 eta: 7:00:49 time: 0.1563 data_time: 0.0106 memory: 7124 grad_norm: 6.0221 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1924 loss: 1.1924 2022/09/07 09:34:44 - mmengine - INFO - Epoch(train) [58][2400/3757] lr: 1.0000e-03 eta: 7:00:33 time: 0.1618 data_time: 0.0096 memory: 7124 grad_norm: 6.2250 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.2672 loss: 1.2672 2022/09/07 09:34:59 - mmengine - INFO - Epoch(train) [58][2500/3757] lr: 1.0000e-03 eta: 7:00:17 time: 0.1578 data_time: 0.0094 memory: 7124 grad_norm: 6.1586 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2007 loss: 1.2007 2022/09/07 09:35:15 - mmengine - INFO - Epoch(train) [58][2600/3757] lr: 1.0000e-03 eta: 7:00:02 time: 0.1576 data_time: 0.0095 memory: 7124 grad_norm: 5.9746 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3002 loss: 1.3002 2022/09/07 09:35:31 - mmengine - INFO - Epoch(train) [58][2700/3757] lr: 1.0000e-03 eta: 6:59:46 time: 0.1578 data_time: 0.0098 memory: 7124 grad_norm: 5.8593 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.1575 loss: 1.1575 2022/09/07 09:35:47 - mmengine - INFO - Epoch(train) [58][2800/3757] lr: 1.0000e-03 eta: 6:59:30 time: 0.1559 data_time: 0.0104 memory: 7124 grad_norm: 6.0600 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2787 loss: 1.2787 2022/09/07 09:35:55 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:36:03 - mmengine - INFO - Epoch(train) [58][2900/3757] lr: 1.0000e-03 eta: 6:59:14 time: 0.1640 data_time: 0.0104 memory: 7124 grad_norm: 6.0618 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.4648 loss: 1.4648 2022/09/07 09:36:19 - mmengine - INFO - Epoch(train) [58][3000/3757] lr: 1.0000e-03 eta: 6:58:58 time: 0.1562 data_time: 0.0109 memory: 7124 grad_norm: 5.9113 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 1.1147 loss: 1.1147 2022/09/07 09:36:35 - mmengine - INFO - Epoch(train) [58][3100/3757] lr: 1.0000e-03 eta: 6:58:42 time: 0.1591 data_time: 0.0102 memory: 7124 grad_norm: 5.8427 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.3061 loss: 1.3061 2022/09/07 09:36:51 - mmengine - INFO - Epoch(train) [58][3200/3757] lr: 1.0000e-03 eta: 6:58:26 time: 0.1595 data_time: 0.0101 memory: 7124 grad_norm: 6.0605 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.3978 loss: 1.3978 2022/09/07 09:37:07 - mmengine - INFO - Epoch(train) [58][3300/3757] lr: 1.0000e-03 eta: 6:58:11 time: 0.1589 data_time: 0.0104 memory: 7124 grad_norm: 5.7886 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.3157 loss: 1.3157 2022/09/07 09:37:23 - mmengine - INFO - Epoch(train) [58][3400/3757] lr: 1.0000e-03 eta: 6:57:55 time: 0.1612 data_time: 0.0096 memory: 7124 grad_norm: 6.0457 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2895 loss: 1.2895 2022/09/07 09:37:39 - mmengine - INFO - Epoch(train) [58][3500/3757] lr: 1.0000e-03 eta: 6:57:39 time: 0.1586 data_time: 0.0118 memory: 7124 grad_norm: 6.0390 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0984 loss: 1.0984 2022/09/07 09:37:55 - mmengine - INFO - Epoch(train) [58][3600/3757] lr: 1.0000e-03 eta: 6:57:23 time: 0.1579 data_time: 0.0100 memory: 7124 grad_norm: 6.0843 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1920 loss: 1.1920 2022/09/07 09:38:11 - mmengine - INFO - Epoch(train) [58][3700/3757] lr: 1.0000e-03 eta: 6:57:07 time: 0.1552 data_time: 0.0107 memory: 7124 grad_norm: 5.9367 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.3927 loss: 1.3927 2022/09/07 09:38:19 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:38:19 - mmengine - INFO - Epoch(train) [58][3757/3757] lr: 1.0000e-03 eta: 6:57:01 time: 0.1368 data_time: 0.0077 memory: 7124 grad_norm: 5.5913 top1_acc: 0.7143 top5_acc: 0.8571 loss_cls: 1.1313 loss: 1.1313 2022/09/07 09:38:20 - mmengine - INFO - Saving checkpoint at 58 epochs 2022/09/07 09:40:39 - mmengine - INFO - Epoch(val) [58][100/310] eta: 0:04:06 time: 1.1731 data_time: 0.8759 memory: 7627 2022/09/07 09:42:53 - mmengine - INFO - Epoch(val) [58][200/310] eta: 0:02:14 time: 1.2257 data_time: 0.9258 memory: 7627 2022/09/07 09:44:59 - mmengine - INFO - Epoch(val) [58][300/310] eta: 0:00:12 time: 1.2381 data_time: 0.9373 memory: 7627 2022/09/07 09:45:19 - mmengine - INFO - Epoch(val) [58][310/310] acc/top1: 0.7372 acc/top5: 0.9128 acc/mean1: 0.7371 2022/09/07 09:45:36 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:45:37 - mmengine - INFO - Epoch(train) [59][100/3757] lr: 1.0000e-03 eta: 6:56:42 time: 0.1630 data_time: 0.0100 memory: 7627 grad_norm: 5.9432 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3600 loss: 1.3600 2022/09/07 09:45:53 - mmengine - INFO - Epoch(train) [59][200/3757] lr: 1.0000e-03 eta: 6:56:26 time: 0.1625 data_time: 0.0099 memory: 7124 grad_norm: 5.5899 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0130 loss: 1.0130 2022/09/07 09:46:09 - mmengine - INFO - Epoch(train) [59][300/3757] lr: 1.0000e-03 eta: 6:56:10 time: 0.1601 data_time: 0.0099 memory: 7124 grad_norm: 5.7284 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1228 loss: 1.1228 2022/09/07 09:46:25 - mmengine - INFO - Epoch(train) [59][400/3757] lr: 1.0000e-03 eta: 6:55:54 time: 0.1587 data_time: 0.0092 memory: 7124 grad_norm: 6.1256 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1454 loss: 1.1454 2022/09/07 09:46:40 - mmengine - INFO - Epoch(train) [59][500/3757] lr: 1.0000e-03 eta: 6:55:38 time: 0.1542 data_time: 0.0109 memory: 7124 grad_norm: 5.9369 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.3931 loss: 1.3931 2022/09/07 09:46:56 - mmengine - INFO - Epoch(train) [59][600/3757] lr: 1.0000e-03 eta: 6:55:22 time: 0.1597 data_time: 0.0095 memory: 7124 grad_norm: 6.1042 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3862 loss: 1.3862 2022/09/07 09:47:12 - mmengine - INFO - Epoch(train) [59][700/3757] lr: 1.0000e-03 eta: 6:55:07 time: 0.1572 data_time: 0.0097 memory: 7124 grad_norm: 5.5926 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9707 loss: 0.9707 2022/09/07 09:47:28 - mmengine - INFO - Epoch(train) [59][800/3757] lr: 1.0000e-03 eta: 6:54:51 time: 0.1668 data_time: 0.0121 memory: 7124 grad_norm: 5.9158 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1920 loss: 1.1920 2022/09/07 09:47:44 - mmengine - INFO - Epoch(train) [59][900/3757] lr: 1.0000e-03 eta: 6:54:35 time: 0.1583 data_time: 0.0102 memory: 7124 grad_norm: 6.0657 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1942 loss: 1.1942 2022/09/07 09:48:00 - mmengine - INFO - Epoch(train) [59][1000/3757] lr: 1.0000e-03 eta: 6:54:19 time: 0.1557 data_time: 0.0099 memory: 7124 grad_norm: 6.0380 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3098 loss: 1.3098 2022/09/07 09:48:15 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:48:16 - mmengine - INFO - Epoch(train) [59][1100/3757] lr: 1.0000e-03 eta: 6:54:03 time: 0.1561 data_time: 0.0101 memory: 7124 grad_norm: 5.9994 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1532 loss: 1.1532 2022/09/07 09:48:32 - mmengine - INFO - Epoch(train) [59][1200/3757] lr: 1.0000e-03 eta: 6:53:47 time: 0.1570 data_time: 0.0103 memory: 7124 grad_norm: 6.2144 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2574 loss: 1.2574 2022/09/07 09:48:48 - mmengine - INFO - Epoch(train) [59][1300/3757] lr: 1.0000e-03 eta: 6:53:32 time: 0.1590 data_time: 0.0107 memory: 7124 grad_norm: 5.9419 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1781 loss: 1.1781 2022/09/07 09:49:04 - mmengine - INFO - Epoch(train) [59][1400/3757] lr: 1.0000e-03 eta: 6:53:16 time: 0.1642 data_time: 0.0104 memory: 7124 grad_norm: 6.0214 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1544 loss: 1.1544 2022/09/07 09:49:20 - mmengine - INFO - Epoch(train) [59][1500/3757] lr: 1.0000e-03 eta: 6:53:00 time: 0.1557 data_time: 0.0107 memory: 7124 grad_norm: 5.9947 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3807 loss: 1.3807 2022/09/07 09:49:36 - mmengine - INFO - Epoch(train) [59][1600/3757] lr: 1.0000e-03 eta: 6:52:44 time: 0.1572 data_time: 0.0101 memory: 7124 grad_norm: 5.9568 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0449 loss: 1.0449 2022/09/07 09:49:52 - mmengine - INFO - Epoch(train) [59][1700/3757] lr: 1.0000e-03 eta: 6:52:28 time: 0.1549 data_time: 0.0094 memory: 7124 grad_norm: 5.8643 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.3756 loss: 1.3756 2022/09/07 09:50:08 - mmengine - INFO - Epoch(train) [59][1800/3757] lr: 1.0000e-03 eta: 6:52:12 time: 0.1545 data_time: 0.0111 memory: 7124 grad_norm: 5.9874 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.4479 loss: 1.4479 2022/09/07 09:50:24 - mmengine - INFO - Epoch(train) [59][1900/3757] lr: 1.0000e-03 eta: 6:51:57 time: 0.1652 data_time: 0.0114 memory: 7124 grad_norm: 5.8192 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.2458 loss: 1.2458 2022/09/07 09:50:40 - mmengine - INFO - Epoch(train) [59][2000/3757] lr: 1.0000e-03 eta: 6:51:41 time: 0.1538 data_time: 0.0091 memory: 7124 grad_norm: 6.0047 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.5393 loss: 1.5393 2022/09/07 09:50:55 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:50:56 - mmengine - INFO - Epoch(train) [59][2100/3757] lr: 1.0000e-03 eta: 6:51:25 time: 0.1575 data_time: 0.0119 memory: 7124 grad_norm: 6.1985 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2980 loss: 1.2980 2022/09/07 09:51:12 - mmengine - INFO - Epoch(train) [59][2200/3757] lr: 1.0000e-03 eta: 6:51:09 time: 0.1577 data_time: 0.0102 memory: 7124 grad_norm: 6.0500 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4500 loss: 1.4500 2022/09/07 09:51:28 - mmengine - INFO - Epoch(train) [59][2300/3757] lr: 1.0000e-03 eta: 6:50:53 time: 0.1567 data_time: 0.0110 memory: 7124 grad_norm: 5.8443 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1148 loss: 1.1148 2022/09/07 09:51:44 - mmengine - INFO - Epoch(train) [59][2400/3757] lr: 1.0000e-03 eta: 6:50:37 time: 0.1703 data_time: 0.0100 memory: 7124 grad_norm: 6.1414 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.3434 loss: 1.3434 2022/09/07 09:52:00 - mmengine - INFO - Epoch(train) [59][2500/3757] lr: 1.0000e-03 eta: 6:50:21 time: 0.1538 data_time: 0.0108 memory: 7124 grad_norm: 5.8658 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1433 loss: 1.1433 2022/09/07 09:52:16 - mmengine - INFO - Epoch(train) [59][2600/3757] lr: 1.0000e-03 eta: 6:50:06 time: 0.1609 data_time: 0.0109 memory: 7124 grad_norm: 6.0259 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0202 loss: 1.0202 2022/09/07 09:52:31 - mmengine - INFO - Epoch(train) [59][2700/3757] lr: 1.0000e-03 eta: 6:49:50 time: 0.1596 data_time: 0.0113 memory: 7124 grad_norm: 6.1801 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3474 loss: 1.3474 2022/09/07 09:52:48 - mmengine - INFO - Epoch(train) [59][2800/3757] lr: 1.0000e-03 eta: 6:49:34 time: 0.1579 data_time: 0.0114 memory: 7124 grad_norm: 6.0647 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.3430 loss: 1.3430 2022/09/07 09:53:04 - mmengine - INFO - Epoch(train) [59][2900/3757] lr: 1.0000e-03 eta: 6:49:18 time: 0.1687 data_time: 0.0109 memory: 7124 grad_norm: 6.2761 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2799 loss: 1.2799 2022/09/07 09:53:20 - mmengine - INFO - Epoch(train) [59][3000/3757] lr: 1.0000e-03 eta: 6:49:02 time: 0.1608 data_time: 0.0108 memory: 7124 grad_norm: 5.9000 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.2702 loss: 1.2702 2022/09/07 09:53:35 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:53:36 - mmengine - INFO - Epoch(train) [59][3100/3757] lr: 1.0000e-03 eta: 6:48:47 time: 0.1609 data_time: 0.0119 memory: 7124 grad_norm: 5.9657 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1271 loss: 1.1271 2022/09/07 09:53:51 - mmengine - INFO - Epoch(train) [59][3200/3757] lr: 1.0000e-03 eta: 6:48:31 time: 0.1570 data_time: 0.0101 memory: 7124 grad_norm: 5.9573 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.1424 loss: 1.1424 2022/09/07 09:54:07 - mmengine - INFO - Epoch(train) [59][3300/3757] lr: 1.0000e-03 eta: 6:48:15 time: 0.1566 data_time: 0.0116 memory: 7124 grad_norm: 5.9030 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1123 loss: 1.1123 2022/09/07 09:54:23 - mmengine - INFO - Epoch(train) [59][3400/3757] lr: 1.0000e-03 eta: 6:47:59 time: 0.1619 data_time: 0.0103 memory: 7124 grad_norm: 6.0821 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1557 loss: 1.1557 2022/09/07 09:54:39 - mmengine - INFO - Epoch(train) [59][3500/3757] lr: 1.0000e-03 eta: 6:47:43 time: 0.1571 data_time: 0.0107 memory: 7124 grad_norm: 6.1893 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.4657 loss: 1.4657 2022/09/07 09:54:55 - mmengine - INFO - Epoch(train) [59][3600/3757] lr: 1.0000e-03 eta: 6:47:27 time: 0.1600 data_time: 0.0112 memory: 7124 grad_norm: 5.8723 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1263 loss: 1.1263 2022/09/07 09:55:11 - mmengine - INFO - Epoch(train) [59][3700/3757] lr: 1.0000e-03 eta: 6:47:11 time: 0.1559 data_time: 0.0107 memory: 7124 grad_norm: 6.2292 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3942 loss: 1.3942 2022/09/07 09:55:20 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 09:55:20 - mmengine - INFO - Epoch(train) [59][3757/3757] lr: 1.0000e-03 eta: 6:47:05 time: 0.1361 data_time: 0.0073 memory: 7124 grad_norm: 6.1271 top1_acc: 0.8571 top5_acc: 1.0000 loss_cls: 1.0802 loss: 1.0802 2022/09/07 09:55:20 - mmengine - INFO - Saving checkpoint at 59 epochs 2022/09/07 09:57:39 - mmengine - INFO - Epoch(val) [59][100/310] eta: 0:04:15 time: 1.2149 data_time: 0.9128 memory: 7627 2022/09/07 09:59:55 - mmengine - INFO - Epoch(val) [59][200/310] eta: 0:02:25 time: 1.3216 data_time: 1.0195 memory: 7627 2022/09/07 10:01:59 - mmengine - INFO - Epoch(val) [59][300/310] eta: 0:00:11 time: 1.1251 data_time: 0.8259 memory: 7627 2022/09/07 10:02:18 - mmengine - INFO - Epoch(val) [59][310/310] acc/top1: 0.7387 acc/top5: 0.9140 acc/mean1: 0.7387 2022/09/07 10:02:36 - mmengine - INFO - Epoch(train) [60][100/3757] lr: 1.0000e-03 eta: 6:46:46 time: 0.1639 data_time: 0.0117 memory: 7627 grad_norm: 6.2201 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1996 loss: 1.1996 2022/09/07 10:02:52 - mmengine - INFO - Epoch(train) [60][200/3757] lr: 1.0000e-03 eta: 6:46:30 time: 0.1580 data_time: 0.0100 memory: 7124 grad_norm: 6.1317 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.4825 loss: 1.4825 2022/09/07 10:03:08 - mmengine - INFO - Epoch(train) [60][300/3757] lr: 1.0000e-03 eta: 6:46:14 time: 0.1565 data_time: 0.0096 memory: 7124 grad_norm: 6.0666 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2228 loss: 1.2228 2022/09/07 10:03:14 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:03:24 - mmengine - INFO - Epoch(train) [60][400/3757] lr: 1.0000e-03 eta: 6:45:58 time: 0.1538 data_time: 0.0102 memory: 7124 grad_norm: 6.3543 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1957 loss: 1.1957 2022/09/07 10:03:40 - mmengine - INFO - Epoch(train) [60][500/3757] lr: 1.0000e-03 eta: 6:45:42 time: 0.1574 data_time: 0.0101 memory: 7124 grad_norm: 6.0621 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9783 loss: 0.9783 2022/09/07 10:03:56 - mmengine - INFO - Epoch(train) [60][600/3757] lr: 1.0000e-03 eta: 6:45:27 time: 0.1605 data_time: 0.0110 memory: 7124 grad_norm: 6.0774 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1546 loss: 1.1546 2022/09/07 10:04:12 - mmengine - INFO - Epoch(train) [60][700/3757] lr: 1.0000e-03 eta: 6:45:11 time: 0.1552 data_time: 0.0089 memory: 7124 grad_norm: 6.1430 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.5545 loss: 1.5545 2022/09/07 10:04:28 - mmengine - INFO - Epoch(train) [60][800/3757] lr: 1.0000e-03 eta: 6:44:55 time: 0.1641 data_time: 0.0105 memory: 7124 grad_norm: 5.9333 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.1975 loss: 1.1975 2022/09/07 10:04:44 - mmengine - INFO - Epoch(train) [60][900/3757] lr: 1.0000e-03 eta: 6:44:39 time: 0.1557 data_time: 0.0106 memory: 7124 grad_norm: 6.1699 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2242 loss: 1.2242 2022/09/07 10:05:00 - mmengine - INFO - Epoch(train) [60][1000/3757] lr: 1.0000e-03 eta: 6:44:23 time: 0.1574 data_time: 0.0098 memory: 7124 grad_norm: 6.0911 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2334 loss: 1.2334 2022/09/07 10:05:16 - mmengine - INFO - Epoch(train) [60][1100/3757] lr: 1.0000e-03 eta: 6:44:07 time: 0.1571 data_time: 0.0098 memory: 7124 grad_norm: 6.1307 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2384 loss: 1.2384 2022/09/07 10:05:31 - mmengine - INFO - Epoch(train) [60][1200/3757] lr: 1.0000e-03 eta: 6:43:52 time: 0.1575 data_time: 0.0101 memory: 7124 grad_norm: 5.8957 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1010 loss: 1.1010 2022/09/07 10:05:48 - mmengine - INFO - Epoch(train) [60][1300/3757] lr: 1.0000e-03 eta: 6:43:36 time: 0.1563 data_time: 0.0095 memory: 7124 grad_norm: 6.1678 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1741 loss: 1.1741 2022/09/07 10:05:53 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:06:03 - mmengine - INFO - Epoch(train) [60][1400/3757] lr: 1.0000e-03 eta: 6:43:20 time: 0.1577 data_time: 0.0099 memory: 7124 grad_norm: 6.1473 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2807 loss: 1.2807 2022/09/07 10:06:19 - mmengine - INFO - Epoch(train) [60][1500/3757] lr: 1.0000e-03 eta: 6:43:04 time: 0.1585 data_time: 0.0112 memory: 7124 grad_norm: 6.1491 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1577 loss: 1.1577 2022/09/07 10:06:35 - mmengine - INFO - Epoch(train) [60][1600/3757] lr: 1.0000e-03 eta: 6:42:48 time: 0.1576 data_time: 0.0096 memory: 7124 grad_norm: 6.0252 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.2074 loss: 1.2074 2022/09/07 10:06:51 - mmengine - INFO - Epoch(train) [60][1700/3757] lr: 1.0000e-03 eta: 6:42:32 time: 0.1575 data_time: 0.0099 memory: 7124 grad_norm: 6.0854 top1_acc: 0.1250 top5_acc: 0.7500 loss_cls: 1.4038 loss: 1.4038 2022/09/07 10:07:07 - mmengine - INFO - Epoch(train) [60][1800/3757] lr: 1.0000e-03 eta: 6:42:16 time: 0.1551 data_time: 0.0108 memory: 7124 grad_norm: 6.0309 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2886 loss: 1.2886 2022/09/07 10:07:23 - mmengine - INFO - Epoch(train) [60][1900/3757] lr: 1.0000e-03 eta: 6:42:01 time: 0.1562 data_time: 0.0094 memory: 7124 grad_norm: 6.0651 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.3310 loss: 1.3310 2022/09/07 10:07:39 - mmengine - INFO - Epoch(train) [60][2000/3757] lr: 1.0000e-03 eta: 6:41:45 time: 0.1601 data_time: 0.0103 memory: 7124 grad_norm: 6.0016 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2740 loss: 1.2740 2022/09/07 10:07:55 - mmengine - INFO - Epoch(train) [60][2100/3757] lr: 1.0000e-03 eta: 6:41:29 time: 0.1684 data_time: 0.0098 memory: 7124 grad_norm: 6.1030 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.0938 loss: 1.0938 2022/09/07 10:08:11 - mmengine - INFO - Epoch(train) [60][2200/3757] lr: 1.0000e-03 eta: 6:41:13 time: 0.1604 data_time: 0.0123 memory: 7124 grad_norm: 6.1421 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0532 loss: 1.0532 2022/09/07 10:08:27 - mmengine - INFO - Epoch(train) [60][2300/3757] lr: 1.0000e-03 eta: 6:40:57 time: 0.1527 data_time: 0.0084 memory: 7124 grad_norm: 6.3144 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.4963 loss: 1.4963 2022/09/07 10:08:33 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:08:43 - mmengine - INFO - Epoch(train) [60][2400/3757] lr: 1.0000e-03 eta: 6:40:42 time: 0.1584 data_time: 0.0104 memory: 7124 grad_norm: 6.0424 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0329 loss: 1.0329 2022/09/07 10:08:59 - mmengine - INFO - Epoch(train) [60][2500/3757] lr: 1.0000e-03 eta: 6:40:26 time: 0.1577 data_time: 0.0100 memory: 7124 grad_norm: 6.4777 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1969 loss: 1.1969 2022/09/07 10:09:15 - mmengine - INFO - Epoch(train) [60][2600/3757] lr: 1.0000e-03 eta: 6:40:10 time: 0.1663 data_time: 0.0108 memory: 7124 grad_norm: 6.3420 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0946 loss: 1.0946 2022/09/07 10:09:31 - mmengine - INFO - Epoch(train) [60][2700/3757] lr: 1.0000e-03 eta: 6:39:54 time: 0.1559 data_time: 0.0094 memory: 7124 grad_norm: 6.4094 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.4813 loss: 1.4813 2022/09/07 10:09:47 - mmengine - INFO - Epoch(train) [60][2800/3757] lr: 1.0000e-03 eta: 6:39:38 time: 0.1609 data_time: 0.0102 memory: 7124 grad_norm: 6.4582 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4598 loss: 1.4598 2022/09/07 10:10:03 - mmengine - INFO - Epoch(train) [60][2900/3757] lr: 1.0000e-03 eta: 6:39:23 time: 0.1564 data_time: 0.0102 memory: 7124 grad_norm: 6.1607 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9601 loss: 0.9601 2022/09/07 10:10:19 - mmengine - INFO - Epoch(train) [60][3000/3757] lr: 1.0000e-03 eta: 6:39:07 time: 0.1554 data_time: 0.0109 memory: 7124 grad_norm: 6.1027 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.2712 loss: 1.2712 2022/09/07 10:10:35 - mmengine - INFO - Epoch(train) [60][3100/3757] lr: 1.0000e-03 eta: 6:38:51 time: 0.1631 data_time: 0.0095 memory: 7124 grad_norm: 6.1615 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9307 loss: 0.9307 2022/09/07 10:10:51 - mmengine - INFO - Epoch(train) [60][3200/3757] lr: 1.0000e-03 eta: 6:38:35 time: 0.1585 data_time: 0.0100 memory: 7124 grad_norm: 6.1212 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1697 loss: 1.1697 2022/09/07 10:11:07 - mmengine - INFO - Epoch(train) [60][3300/3757] lr: 1.0000e-03 eta: 6:38:19 time: 0.1606 data_time: 0.0100 memory: 7124 grad_norm: 6.0510 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2636 loss: 1.2636 2022/09/07 10:11:13 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:11:23 - mmengine - INFO - Epoch(train) [60][3400/3757] lr: 1.0000e-03 eta: 6:38:03 time: 0.1579 data_time: 0.0103 memory: 7124 grad_norm: 6.3098 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0638 loss: 1.0638 2022/09/07 10:11:39 - mmengine - INFO - Epoch(train) [60][3500/3757] lr: 1.0000e-03 eta: 6:37:47 time: 0.1587 data_time: 0.0102 memory: 7124 grad_norm: 6.2171 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1846 loss: 1.1846 2022/09/07 10:11:55 - mmengine - INFO - Epoch(train) [60][3600/3757] lr: 1.0000e-03 eta: 6:37:31 time: 0.1574 data_time: 0.0098 memory: 7124 grad_norm: 6.1647 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2661 loss: 1.2661 2022/09/07 10:12:10 - mmengine - INFO - Epoch(train) [60][3700/3757] lr: 1.0000e-03 eta: 6:37:16 time: 0.1577 data_time: 0.0102 memory: 7124 grad_norm: 5.8016 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2393 loss: 1.2393 2022/09/07 10:12:19 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:12:19 - mmengine - INFO - Epoch(train) [60][3757/3757] lr: 1.0000e-03 eta: 6:37:09 time: 0.1384 data_time: 0.0078 memory: 7124 grad_norm: 6.1720 top1_acc: 0.5714 top5_acc: 0.7143 loss_cls: 1.2341 loss: 1.2341 2022/09/07 10:12:19 - mmengine - INFO - Saving checkpoint at 60 epochs 2022/09/07 10:14:37 - mmengine - INFO - Epoch(val) [60][100/310] eta: 0:03:49 time: 1.0919 data_time: 0.7910 memory: 7627 2022/09/07 10:16:55 - mmengine - INFO - Epoch(val) [60][200/310] eta: 0:02:34 time: 1.4023 data_time: 1.0936 memory: 7627 2022/09/07 10:18:59 - mmengine - INFO - Epoch(val) [60][300/310] eta: 0:00:11 time: 1.1043 data_time: 0.8089 memory: 7627 2022/09/07 10:19:16 - mmengine - INFO - Epoch(val) [60][310/310] acc/top1: 0.7424 acc/top5: 0.9140 acc/mean1: 0.7423 2022/09/07 10:19:16 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_57.pth is removed 2022/09/07 10:19:18 - mmengine - INFO - The best checkpoint with 0.7424 acc/top1 at 60 epoch is saved to best_acc/top1_epoch_60.pth. 2022/09/07 10:19:35 - mmengine - INFO - Epoch(train) [61][100/3757] lr: 1.0000e-03 eta: 6:36:49 time: 0.1606 data_time: 0.0094 memory: 7627 grad_norm: 6.0380 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2446 loss: 1.2446 2022/09/07 10:19:51 - mmengine - INFO - Epoch(train) [61][200/3757] lr: 1.0000e-03 eta: 6:36:34 time: 0.1604 data_time: 0.0104 memory: 7124 grad_norm: 6.1904 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1118 loss: 1.1118 2022/09/07 10:20:07 - mmengine - INFO - Epoch(train) [61][300/3757] lr: 1.0000e-03 eta: 6:36:18 time: 0.1582 data_time: 0.0102 memory: 7124 grad_norm: 5.9836 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1247 loss: 1.1247 2022/09/07 10:20:23 - mmengine - INFO - Epoch(train) [61][400/3757] lr: 1.0000e-03 eta: 6:36:02 time: 0.1554 data_time: 0.0099 memory: 7124 grad_norm: 6.3489 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.4214 loss: 1.4214 2022/09/07 10:20:39 - mmengine - INFO - Epoch(train) [61][500/3757] lr: 1.0000e-03 eta: 6:35:46 time: 0.1702 data_time: 0.0104 memory: 7124 grad_norm: 6.0999 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0631 loss: 1.0631 2022/09/07 10:20:51 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:20:55 - mmengine - INFO - Epoch(train) [61][600/3757] lr: 1.0000e-03 eta: 6:35:30 time: 0.1601 data_time: 0.0114 memory: 7124 grad_norm: 6.0556 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0453 loss: 1.0453 2022/09/07 10:21:11 - mmengine - INFO - Epoch(train) [61][700/3757] lr: 1.0000e-03 eta: 6:35:14 time: 0.1577 data_time: 0.0096 memory: 7124 grad_norm: 6.3393 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2805 loss: 1.2805 2022/09/07 10:21:27 - mmengine - INFO - Epoch(train) [61][800/3757] lr: 1.0000e-03 eta: 6:34:59 time: 0.1589 data_time: 0.0108 memory: 7124 grad_norm: 6.0388 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1950 loss: 1.1950 2022/09/07 10:21:43 - mmengine - INFO - Epoch(train) [61][900/3757] lr: 1.0000e-03 eta: 6:34:43 time: 0.1603 data_time: 0.0098 memory: 7124 grad_norm: 6.3940 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3995 loss: 1.3995 2022/09/07 10:21:59 - mmengine - INFO - Epoch(train) [61][1000/3757] lr: 1.0000e-03 eta: 6:34:27 time: 0.1610 data_time: 0.0113 memory: 7124 grad_norm: 6.2958 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2891 loss: 1.2891 2022/09/07 10:22:15 - mmengine - INFO - Epoch(train) [61][1100/3757] lr: 1.0000e-03 eta: 6:34:11 time: 0.1575 data_time: 0.0092 memory: 7124 grad_norm: 6.3208 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1983 loss: 1.1983 2022/09/07 10:22:31 - mmengine - INFO - Epoch(train) [61][1200/3757] lr: 1.0000e-03 eta: 6:33:56 time: 0.1583 data_time: 0.0103 memory: 7124 grad_norm: 5.8853 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1713 loss: 1.1713 2022/09/07 10:22:47 - mmengine - INFO - Epoch(train) [61][1300/3757] lr: 1.0000e-03 eta: 6:33:40 time: 0.1581 data_time: 0.0106 memory: 7124 grad_norm: 6.1790 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1705 loss: 1.1705 2022/09/07 10:23:03 - mmengine - INFO - Epoch(train) [61][1400/3757] lr: 1.0000e-03 eta: 6:33:24 time: 0.1620 data_time: 0.0098 memory: 7124 grad_norm: 6.0289 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2036 loss: 1.2036 2022/09/07 10:23:19 - mmengine - INFO - Epoch(train) [61][1500/3757] lr: 1.0000e-03 eta: 6:33:08 time: 0.1571 data_time: 0.0101 memory: 7124 grad_norm: 6.2240 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.2684 loss: 1.2684 2022/09/07 10:23:32 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:23:35 - mmengine - INFO - Epoch(train) [61][1600/3757] lr: 1.0000e-03 eta: 6:32:52 time: 0.1579 data_time: 0.0101 memory: 7124 grad_norm: 6.1957 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2848 loss: 1.2848 2022/09/07 10:23:51 - mmengine - INFO - Epoch(train) [61][1700/3757] lr: 1.0000e-03 eta: 6:32:37 time: 0.1562 data_time: 0.0091 memory: 7124 grad_norm: 6.0737 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1033 loss: 1.1033 2022/09/07 10:24:07 - mmengine - INFO - Epoch(train) [61][1800/3757] lr: 1.0000e-03 eta: 6:32:21 time: 0.1594 data_time: 0.0113 memory: 7124 grad_norm: 6.1482 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1204 loss: 1.1204 2022/09/07 10:24:23 - mmengine - INFO - Epoch(train) [61][1900/3757] lr: 1.0000e-03 eta: 6:32:05 time: 0.1561 data_time: 0.0118 memory: 7124 grad_norm: 6.3299 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1251 loss: 1.1251 2022/09/07 10:24:39 - mmengine - INFO - Epoch(train) [61][2000/3757] lr: 1.0000e-03 eta: 6:31:49 time: 0.1618 data_time: 0.0102 memory: 7124 grad_norm: 6.0770 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2591 loss: 1.2591 2022/09/07 10:24:55 - mmengine - INFO - Epoch(train) [61][2100/3757] lr: 1.0000e-03 eta: 6:31:33 time: 0.1595 data_time: 0.0111 memory: 7124 grad_norm: 6.2709 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4473 loss: 1.4473 2022/09/07 10:25:11 - mmengine - INFO - Epoch(train) [61][2200/3757] lr: 1.0000e-03 eta: 6:31:18 time: 0.1656 data_time: 0.0088 memory: 7124 grad_norm: 6.0590 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.3163 loss: 1.3163 2022/09/07 10:25:27 - mmengine - INFO - Epoch(train) [61][2300/3757] lr: 1.0000e-03 eta: 6:31:02 time: 0.1650 data_time: 0.0103 memory: 7124 grad_norm: 5.8656 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0904 loss: 1.0904 2022/09/07 10:25:43 - mmengine - INFO - Epoch(train) [61][2400/3757] lr: 1.0000e-03 eta: 6:30:46 time: 0.1586 data_time: 0.0100 memory: 7124 grad_norm: 6.1941 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2051 loss: 1.2051 2022/09/07 10:25:59 - mmengine - INFO - Epoch(train) [61][2500/3757] lr: 1.0000e-03 eta: 6:30:30 time: 0.1516 data_time: 0.0090 memory: 7124 grad_norm: 5.9310 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1274 loss: 1.1274 2022/09/07 10:26:12 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:26:15 - mmengine - INFO - Epoch(train) [61][2600/3757] lr: 1.0000e-03 eta: 6:30:15 time: 0.1574 data_time: 0.0104 memory: 7124 grad_norm: 5.9809 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.2366 loss: 1.2366 2022/09/07 10:26:32 - mmengine - INFO - Epoch(train) [61][2700/3757] lr: 1.0000e-03 eta: 6:29:59 time: 0.1589 data_time: 0.0120 memory: 7124 grad_norm: 6.1705 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9855 loss: 0.9855 2022/09/07 10:26:48 - mmengine - INFO - Epoch(train) [61][2800/3757] lr: 1.0000e-03 eta: 6:29:43 time: 0.1645 data_time: 0.0093 memory: 7124 grad_norm: 5.9459 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0546 loss: 1.0546 2022/09/07 10:27:04 - mmengine - INFO - Epoch(train) [61][2900/3757] lr: 1.0000e-03 eta: 6:29:27 time: 0.1606 data_time: 0.0121 memory: 7124 grad_norm: 6.3864 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3563 loss: 1.3563 2022/09/07 10:27:20 - mmengine - INFO - Epoch(train) [61][3000/3757] lr: 1.0000e-03 eta: 6:29:11 time: 0.1573 data_time: 0.0098 memory: 7124 grad_norm: 6.4393 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1553 loss: 1.1553 2022/09/07 10:27:36 - mmengine - INFO - Epoch(train) [61][3100/3757] lr: 1.0000e-03 eta: 6:28:56 time: 0.1574 data_time: 0.0108 memory: 7124 grad_norm: 6.0599 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.3302 loss: 1.3302 2022/09/07 10:27:52 - mmengine - INFO - Epoch(train) [61][3200/3757] lr: 1.0000e-03 eta: 6:28:40 time: 0.1541 data_time: 0.0110 memory: 7124 grad_norm: 6.1748 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.3594 loss: 1.3594 2022/09/07 10:28:08 - mmengine - INFO - Epoch(train) [61][3300/3757] lr: 1.0000e-03 eta: 6:28:24 time: 0.1570 data_time: 0.0100 memory: 7124 grad_norm: 5.8801 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2861 loss: 1.2861 2022/09/07 10:28:24 - mmengine - INFO - Epoch(train) [61][3400/3757] lr: 1.0000e-03 eta: 6:28:08 time: 0.1565 data_time: 0.0094 memory: 7124 grad_norm: 6.1898 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2424 loss: 1.2424 2022/09/07 10:28:40 - mmengine - INFO - Epoch(train) [61][3500/3757] lr: 1.0000e-03 eta: 6:27:52 time: 0.1595 data_time: 0.0099 memory: 7124 grad_norm: 5.8502 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2490 loss: 1.2490 2022/09/07 10:28:53 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:28:56 - mmengine - INFO - Epoch(train) [61][3600/3757] lr: 1.0000e-03 eta: 6:27:37 time: 0.1579 data_time: 0.0096 memory: 7124 grad_norm: 6.1511 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.3466 loss: 1.3466 2022/09/07 10:29:12 - mmengine - INFO - Epoch(train) [61][3700/3757] lr: 1.0000e-03 eta: 6:27:21 time: 0.1578 data_time: 0.0102 memory: 7124 grad_norm: 6.2738 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0538 loss: 1.0538 2022/09/07 10:29:21 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:29:21 - mmengine - INFO - Epoch(train) [61][3757/3757] lr: 1.0000e-03 eta: 6:27:14 time: 0.1366 data_time: 0.0076 memory: 7124 grad_norm: 5.8308 top1_acc: 0.5714 top5_acc: 0.8571 loss_cls: 1.2581 loss: 1.2581 2022/09/07 10:29:21 - mmengine - INFO - Saving checkpoint at 61 epochs 2022/09/07 10:31:39 - mmengine - INFO - Epoch(val) [61][100/310] eta: 0:04:09 time: 1.1861 data_time: 0.8854 memory: 7627 2022/09/07 10:33:54 - mmengine - INFO - Epoch(val) [61][200/310] eta: 0:02:19 time: 1.2642 data_time: 0.9662 memory: 7627 2022/09/07 10:35:58 - mmengine - INFO - Epoch(val) [61][300/310] eta: 0:00:11 time: 1.1223 data_time: 0.8213 memory: 7627 2022/09/07 10:36:19 - mmengine - INFO - Epoch(val) [61][310/310] acc/top1: 0.7373 acc/top5: 0.9134 acc/mean1: 0.7372 2022/09/07 10:36:37 - mmengine - INFO - Epoch(train) [62][100/3757] lr: 1.0000e-03 eta: 6:26:55 time: 0.1577 data_time: 0.0100 memory: 7627 grad_norm: 6.1520 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1394 loss: 1.1394 2022/09/07 10:36:53 - mmengine - INFO - Epoch(train) [62][200/3757] lr: 1.0000e-03 eta: 6:26:40 time: 0.1595 data_time: 0.0097 memory: 7124 grad_norm: 5.8843 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0506 loss: 1.0506 2022/09/07 10:37:09 - mmengine - INFO - Epoch(train) [62][300/3757] lr: 1.0000e-03 eta: 6:26:24 time: 0.1571 data_time: 0.0097 memory: 7124 grad_norm: 6.0611 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9685 loss: 0.9685 2022/09/07 10:37:25 - mmengine - INFO - Epoch(train) [62][400/3757] lr: 1.0000e-03 eta: 6:26:08 time: 0.1571 data_time: 0.0101 memory: 7124 grad_norm: 6.0677 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.2484 loss: 1.2484 2022/09/07 10:37:41 - mmengine - INFO - Epoch(train) [62][500/3757] lr: 1.0000e-03 eta: 6:25:52 time: 0.1564 data_time: 0.0101 memory: 7124 grad_norm: 6.3380 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0048 loss: 1.0048 2022/09/07 10:37:57 - mmengine - INFO - Epoch(train) [62][600/3757] lr: 1.0000e-03 eta: 6:25:36 time: 0.1579 data_time: 0.0100 memory: 7124 grad_norm: 6.1918 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1760 loss: 1.1760 2022/09/07 10:38:13 - mmengine - INFO - Epoch(train) [62][700/3757] lr: 1.0000e-03 eta: 6:25:20 time: 0.1539 data_time: 0.0097 memory: 7124 grad_norm: 5.9306 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0905 loss: 1.0905 2022/09/07 10:38:29 - mmengine - INFO - Epoch(train) [62][800/3757] lr: 1.0000e-03 eta: 6:25:04 time: 0.1587 data_time: 0.0098 memory: 7124 grad_norm: 6.7205 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3926 loss: 1.3926 2022/09/07 10:38:33 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:38:45 - mmengine - INFO - Epoch(train) [62][900/3757] lr: 1.0000e-03 eta: 6:24:48 time: 0.1551 data_time: 0.0098 memory: 7124 grad_norm: 6.2133 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1806 loss: 1.1806 2022/09/07 10:39:01 - mmengine - INFO - Epoch(train) [62][1000/3757] lr: 1.0000e-03 eta: 6:24:33 time: 0.1581 data_time: 0.0112 memory: 7124 grad_norm: 6.1679 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0417 loss: 1.0417 2022/09/07 10:39:16 - mmengine - INFO - Epoch(train) [62][1100/3757] lr: 1.0000e-03 eta: 6:24:17 time: 0.1594 data_time: 0.0107 memory: 7124 grad_norm: 5.8724 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.3649 loss: 1.3649 2022/09/07 10:39:32 - mmengine - INFO - Epoch(train) [62][1200/3757] lr: 1.0000e-03 eta: 6:24:01 time: 0.1606 data_time: 0.0101 memory: 7124 grad_norm: 6.2036 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1695 loss: 1.1695 2022/09/07 10:39:48 - mmengine - INFO - Epoch(train) [62][1300/3757] lr: 1.0000e-03 eta: 6:23:45 time: 0.1574 data_time: 0.0098 memory: 7124 grad_norm: 6.1109 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3749 loss: 1.3749 2022/09/07 10:40:04 - mmengine - INFO - Epoch(train) [62][1400/3757] lr: 1.0000e-03 eta: 6:23:29 time: 0.1575 data_time: 0.0112 memory: 7124 grad_norm: 6.4782 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.5814 loss: 1.5814 2022/09/07 10:40:20 - mmengine - INFO - Epoch(train) [62][1500/3757] lr: 1.0000e-03 eta: 6:23:13 time: 0.1638 data_time: 0.0128 memory: 7124 grad_norm: 5.9819 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0122 loss: 1.0122 2022/09/07 10:40:36 - mmengine - INFO - Epoch(train) [62][1600/3757] lr: 1.0000e-03 eta: 6:22:57 time: 0.1589 data_time: 0.0113 memory: 7124 grad_norm: 5.6758 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.4628 loss: 1.4628 2022/09/07 10:40:52 - mmengine - INFO - Epoch(train) [62][1700/3757] lr: 1.0000e-03 eta: 6:22:42 time: 0.1569 data_time: 0.0103 memory: 7124 grad_norm: 6.3299 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.3891 loss: 1.3891 2022/09/07 10:41:08 - mmengine - INFO - Epoch(train) [62][1800/3757] lr: 1.0000e-03 eta: 6:22:26 time: 0.1594 data_time: 0.0115 memory: 7124 grad_norm: 6.1421 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1800 loss: 1.1800 2022/09/07 10:41:12 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:41:24 - mmengine - INFO - Epoch(train) [62][1900/3757] lr: 1.0000e-03 eta: 6:22:10 time: 0.1586 data_time: 0.0107 memory: 7124 grad_norm: 6.2593 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1024 loss: 1.1024 2022/09/07 10:41:40 - mmengine - INFO - Epoch(train) [62][2000/3757] lr: 1.0000e-03 eta: 6:21:54 time: 0.1617 data_time: 0.0121 memory: 7124 grad_norm: 6.1968 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2847 loss: 1.2847 2022/09/07 10:41:56 - mmengine - INFO - Epoch(train) [62][2100/3757] lr: 1.0000e-03 eta: 6:21:38 time: 0.1594 data_time: 0.0123 memory: 7124 grad_norm: 6.1428 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.2545 loss: 1.2545 2022/09/07 10:42:12 - mmengine - INFO - Epoch(train) [62][2200/3757] lr: 1.0000e-03 eta: 6:21:22 time: 0.1559 data_time: 0.0097 memory: 7124 grad_norm: 6.2202 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1265 loss: 1.1265 2022/09/07 10:42:28 - mmengine - INFO - Epoch(train) [62][2300/3757] lr: 1.0000e-03 eta: 6:21:07 time: 0.1647 data_time: 0.0092 memory: 7124 grad_norm: 6.5529 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1669 loss: 1.1669 2022/09/07 10:42:44 - mmengine - INFO - Epoch(train) [62][2400/3757] lr: 1.0000e-03 eta: 6:20:51 time: 0.1573 data_time: 0.0100 memory: 7124 grad_norm: 6.0403 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2213 loss: 1.2213 2022/09/07 10:43:00 - mmengine - INFO - Epoch(train) [62][2500/3757] lr: 1.0000e-03 eta: 6:20:35 time: 0.1589 data_time: 0.0105 memory: 7124 grad_norm: 6.2178 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1583 loss: 1.1583 2022/09/07 10:43:16 - mmengine - INFO - Epoch(train) [62][2600/3757] lr: 1.0000e-03 eta: 6:20:19 time: 0.1584 data_time: 0.0113 memory: 7124 grad_norm: 6.2054 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3315 loss: 1.3315 2022/09/07 10:43:32 - mmengine - INFO - Epoch(train) [62][2700/3757] lr: 1.0000e-03 eta: 6:20:03 time: 0.1548 data_time: 0.0098 memory: 7124 grad_norm: 6.2358 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.1007 loss: 1.1007 2022/09/07 10:43:47 - mmengine - INFO - Epoch(train) [62][2800/3757] lr: 1.0000e-03 eta: 6:19:47 time: 0.1557 data_time: 0.0092 memory: 7124 grad_norm: 6.2462 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.2623 loss: 1.2623 2022/09/07 10:43:51 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:44:03 - mmengine - INFO - Epoch(train) [62][2900/3757] lr: 1.0000e-03 eta: 6:19:31 time: 0.1585 data_time: 0.0101 memory: 7124 grad_norm: 6.0905 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1223 loss: 1.1223 2022/09/07 10:44:19 - mmengine - INFO - Epoch(train) [62][3000/3757] lr: 1.0000e-03 eta: 6:19:15 time: 0.1689 data_time: 0.0107 memory: 7124 grad_norm: 6.3519 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2597 loss: 1.2597 2022/09/07 10:44:35 - mmengine - INFO - Epoch(train) [62][3100/3757] lr: 1.0000e-03 eta: 6:19:00 time: 0.1586 data_time: 0.0107 memory: 7124 grad_norm: 6.1610 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0882 loss: 1.0882 2022/09/07 10:44:51 - mmengine - INFO - Epoch(train) [62][3200/3757] lr: 1.0000e-03 eta: 6:18:44 time: 0.1567 data_time: 0.0117 memory: 7124 grad_norm: 6.0527 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1910 loss: 1.1910 2022/09/07 10:45:07 - mmengine - INFO - Epoch(train) [62][3300/3757] lr: 1.0000e-03 eta: 6:18:28 time: 0.1587 data_time: 0.0106 memory: 7124 grad_norm: 6.0380 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1919 loss: 1.1919 2022/09/07 10:45:23 - mmengine - INFO - Epoch(train) [62][3400/3757] lr: 1.0000e-03 eta: 6:18:12 time: 0.1578 data_time: 0.0105 memory: 7124 grad_norm: 6.1917 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9938 loss: 0.9938 2022/09/07 10:45:39 - mmengine - INFO - Epoch(train) [62][3500/3757] lr: 1.0000e-03 eta: 6:17:56 time: 0.1574 data_time: 0.0100 memory: 7124 grad_norm: 6.3275 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.3230 loss: 1.3230 2022/09/07 10:45:55 - mmengine - INFO - Epoch(train) [62][3600/3757] lr: 1.0000e-03 eta: 6:17:40 time: 0.1575 data_time: 0.0103 memory: 7124 grad_norm: 6.2729 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2241 loss: 1.2241 2022/09/07 10:46:11 - mmengine - INFO - Epoch(train) [62][3700/3757] lr: 1.0000e-03 eta: 6:17:25 time: 0.1597 data_time: 0.0102 memory: 7124 grad_norm: 6.1306 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2855 loss: 1.2855 2022/09/07 10:46:20 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:46:20 - mmengine - INFO - Epoch(train) [62][3757/3757] lr: 1.0000e-03 eta: 6:17:18 time: 0.1382 data_time: 0.0074 memory: 7124 grad_norm: 5.9943 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1426 loss: 1.1426 2022/09/07 10:46:20 - mmengine - INFO - Saving checkpoint at 62 epochs 2022/09/07 10:48:39 - mmengine - INFO - Epoch(val) [62][100/310] eta: 0:04:33 time: 1.3004 data_time: 0.9980 memory: 7627 2022/09/07 10:50:52 - mmengine - INFO - Epoch(val) [62][200/310] eta: 0:02:03 time: 1.1271 data_time: 0.8258 memory: 7627 2022/09/07 10:53:00 - mmengine - INFO - Epoch(val) [62][300/310] eta: 0:00:13 time: 1.3022 data_time: 1.0013 memory: 7627 2022/09/07 10:53:18 - mmengine - INFO - Epoch(val) [62][310/310] acc/top1: 0.7421 acc/top5: 0.9136 acc/mean1: 0.7420 2022/09/07 10:53:30 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:53:36 - mmengine - INFO - Epoch(train) [63][100/3757] lr: 1.0000e-03 eta: 6:16:59 time: 0.1595 data_time: 0.0113 memory: 7627 grad_norm: 6.0278 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0877 loss: 1.0877 2022/09/07 10:53:52 - mmengine - INFO - Epoch(train) [63][200/3757] lr: 1.0000e-03 eta: 6:16:43 time: 0.1577 data_time: 0.0094 memory: 7124 grad_norm: 6.1743 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3604 loss: 1.3604 2022/09/07 10:54:08 - mmengine - INFO - Epoch(train) [63][300/3757] lr: 1.0000e-03 eta: 6:16:27 time: 0.1558 data_time: 0.0105 memory: 7124 grad_norm: 6.2577 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0702 loss: 1.0702 2022/09/07 10:54:24 - mmengine - INFO - Epoch(train) [63][400/3757] lr: 1.0000e-03 eta: 6:16:11 time: 0.1581 data_time: 0.0094 memory: 7124 grad_norm: 6.2505 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1270 loss: 1.1270 2022/09/07 10:54:39 - mmengine - INFO - Epoch(train) [63][500/3757] lr: 1.0000e-03 eta: 6:15:56 time: 0.1585 data_time: 0.0101 memory: 7124 grad_norm: 6.0953 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0289 loss: 1.0289 2022/09/07 10:54:55 - mmengine - INFO - Epoch(train) [63][600/3757] lr: 1.0000e-03 eta: 6:15:40 time: 0.1574 data_time: 0.0111 memory: 7124 grad_norm: 6.3374 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3797 loss: 1.3797 2022/09/07 10:55:12 - mmengine - INFO - Epoch(train) [63][700/3757] lr: 1.0000e-03 eta: 6:15:24 time: 0.1554 data_time: 0.0095 memory: 7124 grad_norm: 6.1087 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0637 loss: 1.0637 2022/09/07 10:55:28 - mmengine - INFO - Epoch(train) [63][800/3757] lr: 1.0000e-03 eta: 6:15:08 time: 0.1565 data_time: 0.0090 memory: 7124 grad_norm: 6.1376 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1704 loss: 1.1704 2022/09/07 10:55:43 - mmengine - INFO - Epoch(train) [63][900/3757] lr: 1.0000e-03 eta: 6:14:52 time: 0.1562 data_time: 0.0100 memory: 7124 grad_norm: 6.4049 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2794 loss: 1.2794 2022/09/07 10:55:59 - mmengine - INFO - Epoch(train) [63][1000/3757] lr: 1.0000e-03 eta: 6:14:36 time: 0.1603 data_time: 0.0105 memory: 7124 grad_norm: 6.3170 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1290 loss: 1.1290 2022/09/07 10:56:10 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:56:15 - mmengine - INFO - Epoch(train) [63][1100/3757] lr: 1.0000e-03 eta: 6:14:21 time: 0.1581 data_time: 0.0112 memory: 7124 grad_norm: 6.2620 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2426 loss: 1.2426 2022/09/07 10:56:32 - mmengine - INFO - Epoch(train) [63][1200/3757] lr: 1.0000e-03 eta: 6:14:05 time: 0.1566 data_time: 0.0102 memory: 7124 grad_norm: 6.0132 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.0640 loss: 1.0640 2022/09/07 10:56:47 - mmengine - INFO - Epoch(train) [63][1300/3757] lr: 1.0000e-03 eta: 6:13:49 time: 0.1574 data_time: 0.0099 memory: 7124 grad_norm: 6.2812 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1275 loss: 1.1275 2022/09/07 10:57:04 - mmengine - INFO - Epoch(train) [63][1400/3757] lr: 1.0000e-03 eta: 6:13:33 time: 0.1617 data_time: 0.0109 memory: 7124 grad_norm: 6.0650 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9509 loss: 0.9509 2022/09/07 10:57:20 - mmengine - INFO - Epoch(train) [63][1500/3757] lr: 1.0000e-03 eta: 6:13:17 time: 0.1627 data_time: 0.0096 memory: 7124 grad_norm: 6.4053 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.2211 loss: 1.2211 2022/09/07 10:57:35 - mmengine - INFO - Epoch(train) [63][1600/3757] lr: 1.0000e-03 eta: 6:13:02 time: 0.1602 data_time: 0.0128 memory: 7124 grad_norm: 6.1262 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2144 loss: 1.2144 2022/09/07 10:57:51 - mmengine - INFO - Epoch(train) [63][1700/3757] lr: 1.0000e-03 eta: 6:12:46 time: 0.1597 data_time: 0.0099 memory: 7124 grad_norm: 6.2664 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.1118 loss: 1.1118 2022/09/07 10:58:07 - mmengine - INFO - Epoch(train) [63][1800/3757] lr: 1.0000e-03 eta: 6:12:30 time: 0.1583 data_time: 0.0097 memory: 7124 grad_norm: 6.3098 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.3056 loss: 1.3056 2022/09/07 10:58:24 - mmengine - INFO - Epoch(train) [63][1900/3757] lr: 1.0000e-03 eta: 6:12:14 time: 0.1544 data_time: 0.0100 memory: 7124 grad_norm: 6.1575 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1880 loss: 1.1880 2022/09/07 10:58:40 - mmengine - INFO - Epoch(train) [63][2000/3757] lr: 1.0000e-03 eta: 6:11:58 time: 0.1618 data_time: 0.0105 memory: 7124 grad_norm: 6.3122 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2847 loss: 1.2847 2022/09/07 10:58:50 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 10:58:56 - mmengine - INFO - Epoch(train) [63][2100/3757] lr: 1.0000e-03 eta: 6:11:43 time: 0.1565 data_time: 0.0098 memory: 7124 grad_norm: 6.2043 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.9741 loss: 0.9741 2022/09/07 10:59:12 - mmengine - INFO - Epoch(train) [63][2200/3757] lr: 1.0000e-03 eta: 6:11:27 time: 0.1630 data_time: 0.0114 memory: 7124 grad_norm: 6.4305 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1765 loss: 1.1765 2022/09/07 10:59:27 - mmengine - INFO - Epoch(train) [63][2300/3757] lr: 1.0000e-03 eta: 6:11:11 time: 0.1570 data_time: 0.0112 memory: 7124 grad_norm: 6.1610 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.4627 loss: 1.4627 2022/09/07 10:59:44 - mmengine - INFO - Epoch(train) [63][2400/3757] lr: 1.0000e-03 eta: 6:10:55 time: 0.1595 data_time: 0.0131 memory: 7124 grad_norm: 6.4059 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.3255 loss: 1.3255 2022/09/07 10:59:59 - mmengine - INFO - Epoch(train) [63][2500/3757] lr: 1.0000e-03 eta: 6:10:39 time: 0.1565 data_time: 0.0106 memory: 7124 grad_norm: 6.2287 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2286 loss: 1.2286 2022/09/07 11:00:15 - mmengine - INFO - Epoch(train) [63][2600/3757] lr: 1.0000e-03 eta: 6:10:23 time: 0.1577 data_time: 0.0094 memory: 7124 grad_norm: 6.1632 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.3430 loss: 1.3430 2022/09/07 11:00:31 - mmengine - INFO - Epoch(train) [63][2700/3757] lr: 1.0000e-03 eta: 6:10:07 time: 0.1612 data_time: 0.0111 memory: 7124 grad_norm: 6.1803 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0007 loss: 1.0007 2022/09/07 11:00:47 - mmengine - INFO - Epoch(train) [63][2800/3757] lr: 1.0000e-03 eta: 6:09:52 time: 0.1579 data_time: 0.0098 memory: 7124 grad_norm: 6.3471 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1777 loss: 1.1777 2022/09/07 11:01:03 - mmengine - INFO - Epoch(train) [63][2900/3757] lr: 1.0000e-03 eta: 6:09:36 time: 0.1576 data_time: 0.0100 memory: 7124 grad_norm: 6.1890 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2634 loss: 1.2634 2022/09/07 11:01:19 - mmengine - INFO - Epoch(train) [63][3000/3757] lr: 1.0000e-03 eta: 6:09:20 time: 0.1566 data_time: 0.0096 memory: 7124 grad_norm: 6.4372 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.3257 loss: 1.3257 2022/09/07 11:01:30 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:01:35 - mmengine - INFO - Epoch(train) [63][3100/3757] lr: 1.0000e-03 eta: 6:09:04 time: 0.1596 data_time: 0.0099 memory: 7124 grad_norm: 6.2521 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.3040 loss: 1.3040 2022/09/07 11:01:51 - mmengine - INFO - Epoch(train) [63][3200/3757] lr: 1.0000e-03 eta: 6:08:48 time: 0.1615 data_time: 0.0110 memory: 7124 grad_norm: 6.1303 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0592 loss: 1.0592 2022/09/07 11:02:07 - mmengine - INFO - Epoch(train) [63][3300/3757] lr: 1.0000e-03 eta: 6:08:32 time: 0.1590 data_time: 0.0111 memory: 7124 grad_norm: 6.4536 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9138 loss: 0.9138 2022/09/07 11:02:23 - mmengine - INFO - Epoch(train) [63][3400/3757] lr: 1.0000e-03 eta: 6:08:17 time: 0.1626 data_time: 0.0108 memory: 7124 grad_norm: 6.2088 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.4871 loss: 1.4871 2022/09/07 11:02:39 - mmengine - INFO - Epoch(train) [63][3500/3757] lr: 1.0000e-03 eta: 6:08:01 time: 0.1568 data_time: 0.0106 memory: 7124 grad_norm: 6.2312 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.3126 loss: 1.3126 2022/09/07 11:02:55 - mmengine - INFO - Epoch(train) [63][3600/3757] lr: 1.0000e-03 eta: 6:07:45 time: 0.1586 data_time: 0.0101 memory: 7124 grad_norm: 6.5288 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.4094 loss: 1.4094 2022/09/07 11:03:11 - mmengine - INFO - Epoch(train) [63][3700/3757] lr: 1.0000e-03 eta: 6:07:29 time: 0.1603 data_time: 0.0096 memory: 7124 grad_norm: 6.1975 top1_acc: 0.1250 top5_acc: 0.6250 loss_cls: 1.5161 loss: 1.5161 2022/09/07 11:03:20 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:03:20 - mmengine - INFO - Epoch(train) [63][3757/3757] lr: 1.0000e-03 eta: 6:07:23 time: 0.1377 data_time: 0.0072 memory: 7124 grad_norm: 6.4407 top1_acc: 0.7143 top5_acc: 1.0000 loss_cls: 1.3268 loss: 1.3268 2022/09/07 11:03:20 - mmengine - INFO - Saving checkpoint at 63 epochs 2022/09/07 11:05:38 - mmengine - INFO - Epoch(val) [63][100/310] eta: 0:04:03 time: 1.1578 data_time: 0.8553 memory: 7627 2022/09/07 11:07:53 - mmengine - INFO - Epoch(val) [63][200/310] eta: 0:02:14 time: 1.2229 data_time: 0.9204 memory: 7627 2022/09/07 11:10:00 - mmengine - INFO - Epoch(val) [63][300/310] eta: 0:00:12 time: 1.2624 data_time: 0.9620 memory: 7627 2022/09/07 11:10:17 - mmengine - INFO - Epoch(val) [63][310/310] acc/top1: 0.7413 acc/top5: 0.9156 acc/mean1: 0.7413 2022/09/07 11:10:35 - mmengine - INFO - Epoch(train) [64][100/3757] lr: 1.0000e-03 eta: 6:07:04 time: 0.1585 data_time: 0.0107 memory: 7627 grad_norm: 6.3723 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1482 loss: 1.1482 2022/09/07 11:10:51 - mmengine - INFO - Epoch(train) [64][200/3757] lr: 1.0000e-03 eta: 6:06:48 time: 0.1614 data_time: 0.0106 memory: 7124 grad_norm: 6.0850 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9179 loss: 0.9179 2022/09/07 11:11:07 - mmengine - INFO - Epoch(train) [64][300/3757] lr: 1.0000e-03 eta: 6:06:32 time: 0.1582 data_time: 0.0093 memory: 7124 grad_norm: 6.2032 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0678 loss: 1.0678 2022/09/07 11:11:09 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:11:23 - mmengine - INFO - Epoch(train) [64][400/3757] lr: 1.0000e-03 eta: 6:06:16 time: 0.1608 data_time: 0.0116 memory: 7124 grad_norm: 6.1698 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2808 loss: 1.2808 2022/09/07 11:11:39 - mmengine - INFO - Epoch(train) [64][500/3757] lr: 1.0000e-03 eta: 6:06:00 time: 0.1592 data_time: 0.0096 memory: 7124 grad_norm: 6.3392 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.0334 loss: 1.0334 2022/09/07 11:11:55 - mmengine - INFO - Epoch(train) [64][600/3757] lr: 1.0000e-03 eta: 6:05:45 time: 0.1586 data_time: 0.0096 memory: 7124 grad_norm: 6.3954 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2294 loss: 1.2294 2022/09/07 11:12:12 - mmengine - INFO - Epoch(train) [64][700/3757] lr: 1.0000e-03 eta: 6:05:29 time: 0.1702 data_time: 0.0095 memory: 7124 grad_norm: 6.3805 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1126 loss: 1.1126 2022/09/07 11:12:28 - mmengine - INFO - Epoch(train) [64][800/3757] lr: 1.0000e-03 eta: 6:05:13 time: 0.1583 data_time: 0.0100 memory: 7124 grad_norm: 6.2958 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.4040 loss: 1.4040 2022/09/07 11:12:44 - mmengine - INFO - Epoch(train) [64][900/3757] lr: 1.0000e-03 eta: 6:04:58 time: 0.1632 data_time: 0.0107 memory: 7124 grad_norm: 6.2353 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1063 loss: 1.1063 2022/09/07 11:13:00 - mmengine - INFO - Epoch(train) [64][1000/3757] lr: 1.0000e-03 eta: 6:04:42 time: 0.1609 data_time: 0.0115 memory: 7124 grad_norm: 6.2564 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.4484 loss: 1.4484 2022/09/07 11:13:17 - mmengine - INFO - Epoch(train) [64][1100/3757] lr: 1.0000e-03 eta: 6:04:26 time: 0.1583 data_time: 0.0115 memory: 7124 grad_norm: 6.2074 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.9460 loss: 0.9460 2022/09/07 11:13:33 - mmengine - INFO - Epoch(train) [64][1200/3757] lr: 1.0000e-03 eta: 6:04:10 time: 0.1617 data_time: 0.0110 memory: 7124 grad_norm: 6.6619 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.1667 loss: 1.1667 2022/09/07 11:13:49 - mmengine - INFO - Epoch(train) [64][1300/3757] lr: 1.0000e-03 eta: 6:03:55 time: 0.1604 data_time: 0.0099 memory: 7124 grad_norm: 6.2848 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1285 loss: 1.1285 2022/09/07 11:13:50 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:14:05 - mmengine - INFO - Epoch(train) [64][1400/3757] lr: 1.0000e-03 eta: 6:03:39 time: 0.1606 data_time: 0.0108 memory: 7124 grad_norm: 6.0775 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.3041 loss: 1.3041 2022/09/07 11:14:21 - mmengine - INFO - Epoch(train) [64][1500/3757] lr: 1.0000e-03 eta: 6:03:23 time: 0.1561 data_time: 0.0102 memory: 7124 grad_norm: 6.3250 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0984 loss: 1.0984 2022/09/07 11:14:37 - mmengine - INFO - Epoch(train) [64][1600/3757] lr: 1.0000e-03 eta: 6:03:07 time: 0.1593 data_time: 0.0100 memory: 7124 grad_norm: 6.5411 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.3099 loss: 1.3099 2022/09/07 11:14:53 - mmengine - INFO - Epoch(train) [64][1700/3757] lr: 1.0000e-03 eta: 6:02:52 time: 0.1618 data_time: 0.0098 memory: 7124 grad_norm: 6.1798 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2191 loss: 1.2191 2022/09/07 11:15:09 - mmengine - INFO - Epoch(train) [64][1800/3757] lr: 1.0000e-03 eta: 6:02:36 time: 0.1559 data_time: 0.0097 memory: 7124 grad_norm: 6.6284 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2267 loss: 1.2267 2022/09/07 11:15:25 - mmengine - INFO - Epoch(train) [64][1900/3757] lr: 1.0000e-03 eta: 6:02:20 time: 0.1583 data_time: 0.0111 memory: 7124 grad_norm: 6.4136 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0765 loss: 1.0765 2022/09/07 11:15:41 - mmengine - INFO - Epoch(train) [64][2000/3757] lr: 1.0000e-03 eta: 6:02:04 time: 0.1587 data_time: 0.0100 memory: 7124 grad_norm: 6.1306 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2314 loss: 1.2314 2022/09/07 11:15:57 - mmengine - INFO - Epoch(train) [64][2100/3757] lr: 1.0000e-03 eta: 6:01:48 time: 0.1599 data_time: 0.0106 memory: 7124 grad_norm: 6.4680 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2360 loss: 1.2360 2022/09/07 11:16:13 - mmengine - INFO - Epoch(train) [64][2200/3757] lr: 1.0000e-03 eta: 6:01:32 time: 0.1629 data_time: 0.0095 memory: 7124 grad_norm: 6.3997 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.2134 loss: 1.2134 2022/09/07 11:16:29 - mmengine - INFO - Epoch(train) [64][2300/3757] lr: 1.0000e-03 eta: 6:01:17 time: 0.1552 data_time: 0.0098 memory: 7124 grad_norm: 6.3190 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3346 loss: 1.3346 2022/09/07 11:16:30 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:16:45 - mmengine - INFO - Epoch(train) [64][2400/3757] lr: 1.0000e-03 eta: 6:01:01 time: 0.1583 data_time: 0.0098 memory: 7124 grad_norm: 6.2054 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2387 loss: 1.2387 2022/09/07 11:17:01 - mmengine - INFO - Epoch(train) [64][2500/3757] lr: 1.0000e-03 eta: 6:00:45 time: 0.1594 data_time: 0.0111 memory: 7124 grad_norm: 6.2911 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2235 loss: 1.2235 2022/09/07 11:17:17 - mmengine - INFO - Epoch(train) [64][2600/3757] lr: 1.0000e-03 eta: 6:00:29 time: 0.1581 data_time: 0.0109 memory: 7124 grad_norm: 6.2245 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0962 loss: 1.0962 2022/09/07 11:17:34 - mmengine - INFO - Epoch(train) [64][2700/3757] lr: 1.0000e-03 eta: 6:00:14 time: 0.1596 data_time: 0.0109 memory: 7124 grad_norm: 6.2341 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1817 loss: 1.1817 2022/09/07 11:17:50 - mmengine - INFO - Epoch(train) [64][2800/3757] lr: 1.0000e-03 eta: 5:59:58 time: 0.1613 data_time: 0.0103 memory: 7124 grad_norm: 6.3981 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3959 loss: 1.3959 2022/09/07 11:18:06 - mmengine - INFO - Epoch(train) [64][2900/3757] lr: 1.0000e-03 eta: 5:59:42 time: 0.1571 data_time: 0.0104 memory: 7124 grad_norm: 6.3098 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1433 loss: 1.1433 2022/09/07 11:18:22 - mmengine - INFO - Epoch(train) [64][3000/3757] lr: 1.0000e-03 eta: 5:59:26 time: 0.1595 data_time: 0.0098 memory: 7124 grad_norm: 6.4236 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1162 loss: 1.1162 2022/09/07 11:18:38 - mmengine - INFO - Epoch(train) [64][3100/3757] lr: 1.0000e-03 eta: 5:59:11 time: 0.1582 data_time: 0.0105 memory: 7124 grad_norm: 6.4334 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0192 loss: 1.0192 2022/09/07 11:18:54 - mmengine - INFO - Epoch(train) [64][3200/3757] lr: 1.0000e-03 eta: 5:58:55 time: 0.1575 data_time: 0.0096 memory: 7124 grad_norm: 5.9352 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9481 loss: 0.9481 2022/09/07 11:19:11 - mmengine - INFO - Epoch(train) [64][3300/3757] lr: 1.0000e-03 eta: 5:58:39 time: 0.1579 data_time: 0.0101 memory: 7124 grad_norm: 6.2436 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.2454 loss: 1.2454 2022/09/07 11:19:12 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:19:27 - mmengine - INFO - Epoch(train) [64][3400/3757] lr: 1.0000e-03 eta: 5:58:24 time: 0.1572 data_time: 0.0090 memory: 7124 grad_norm: 6.5581 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 1.2947 loss: 1.2947 2022/09/07 11:19:43 - mmengine - INFO - Epoch(train) [64][3500/3757] lr: 1.0000e-03 eta: 5:58:08 time: 0.1611 data_time: 0.0105 memory: 7124 grad_norm: 6.3794 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.3730 loss: 1.3730 2022/09/07 11:19:59 - mmengine - INFO - Epoch(train) [64][3600/3757] lr: 1.0000e-03 eta: 5:57:52 time: 0.1565 data_time: 0.0101 memory: 7124 grad_norm: 6.6325 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2421 loss: 1.2421 2022/09/07 11:20:15 - mmengine - INFO - Epoch(train) [64][3700/3757] lr: 1.0000e-03 eta: 5:57:36 time: 0.1517 data_time: 0.0093 memory: 7124 grad_norm: 6.5352 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0373 loss: 1.0373 2022/09/07 11:20:24 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:20:24 - mmengine - INFO - Epoch(train) [64][3757/3757] lr: 1.0000e-03 eta: 5:57:30 time: 0.1391 data_time: 0.0083 memory: 7124 grad_norm: 6.3018 top1_acc: 0.7143 top5_acc: 0.7143 loss_cls: 1.3963 loss: 1.3963 2022/09/07 11:20:24 - mmengine - INFO - Saving checkpoint at 64 epochs 2022/09/07 11:22:41 - mmengine - INFO - Epoch(val) [64][100/310] eta: 0:03:33 time: 1.0144 data_time: 0.7139 memory: 7627 2022/09/07 11:24:58 - mmengine - INFO - Epoch(val) [64][200/310] eta: 0:02:28 time: 1.3522 data_time: 1.0526 memory: 7627 2022/09/07 11:27:01 - mmengine - INFO - Epoch(val) [64][300/310] eta: 0:00:11 time: 1.1119 data_time: 0.8126 memory: 7627 2022/09/07 11:27:21 - mmengine - INFO - Epoch(val) [64][310/310] acc/top1: 0.7394 acc/top5: 0.9133 acc/mean1: 0.7394 2022/09/07 11:27:39 - mmengine - INFO - Epoch(train) [65][100/3757] lr: 1.0000e-03 eta: 5:57:11 time: 0.1596 data_time: 0.0102 memory: 7627 grad_norm: 6.2444 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.4157 loss: 1.4157 2022/09/07 11:27:55 - mmengine - INFO - Epoch(train) [65][200/3757] lr: 1.0000e-03 eta: 5:56:55 time: 0.1622 data_time: 0.0088 memory: 7124 grad_norm: 6.2231 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2388 loss: 1.2388 2022/09/07 11:28:11 - mmengine - INFO - Epoch(train) [65][300/3757] lr: 1.0000e-03 eta: 5:56:40 time: 0.1570 data_time: 0.0099 memory: 7124 grad_norm: 6.3479 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1698 loss: 1.1698 2022/09/07 11:28:27 - mmengine - INFO - Epoch(train) [65][400/3757] lr: 1.0000e-03 eta: 5:56:24 time: 0.1584 data_time: 0.0102 memory: 7124 grad_norm: 6.1542 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1294 loss: 1.1294 2022/09/07 11:28:43 - mmengine - INFO - Epoch(train) [65][500/3757] lr: 1.0000e-03 eta: 5:56:08 time: 0.1622 data_time: 0.0097 memory: 7124 grad_norm: 6.0491 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0387 loss: 1.0387 2022/09/07 11:28:52 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:28:59 - mmengine - INFO - Epoch(train) [65][600/3757] lr: 1.0000e-03 eta: 5:55:52 time: 0.1585 data_time: 0.0109 memory: 7124 grad_norm: 6.3292 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.2093 loss: 1.2093 2022/09/07 11:29:16 - mmengine - INFO - Epoch(train) [65][700/3757] lr: 1.0000e-03 eta: 5:55:37 time: 0.1600 data_time: 0.0114 memory: 7124 grad_norm: 6.4010 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2080 loss: 1.2080 2022/09/07 11:29:32 - mmengine - INFO - Epoch(train) [65][800/3757] lr: 1.0000e-03 eta: 5:55:21 time: 0.1587 data_time: 0.0109 memory: 7124 grad_norm: 6.5530 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1274 loss: 1.1274 2022/09/07 11:29:48 - mmengine - INFO - Epoch(train) [65][900/3757] lr: 1.0000e-03 eta: 5:55:05 time: 0.1596 data_time: 0.0093 memory: 7124 grad_norm: 6.2148 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9212 loss: 0.9212 2022/09/07 11:30:04 - mmengine - INFO - Epoch(train) [65][1000/3757] lr: 1.0000e-03 eta: 5:54:49 time: 0.1612 data_time: 0.0104 memory: 7124 grad_norm: 6.4015 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.4683 loss: 1.4683 2022/09/07 11:30:20 - mmengine - INFO - Epoch(train) [65][1100/3757] lr: 1.0000e-03 eta: 5:54:34 time: 0.1597 data_time: 0.0104 memory: 7124 grad_norm: 6.1626 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 0.9614 loss: 0.9614 2022/09/07 11:30:36 - mmengine - INFO - Epoch(train) [65][1200/3757] lr: 1.0000e-03 eta: 5:54:18 time: 0.1553 data_time: 0.0092 memory: 7124 grad_norm: 6.4802 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1278 loss: 1.1278 2022/09/07 11:30:52 - mmengine - INFO - Epoch(train) [65][1300/3757] lr: 1.0000e-03 eta: 5:54:02 time: 0.1625 data_time: 0.0092 memory: 7124 grad_norm: 6.2111 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2333 loss: 1.2333 2022/09/07 11:31:08 - mmengine - INFO - Epoch(train) [65][1400/3757] lr: 1.0000e-03 eta: 5:53:46 time: 0.1585 data_time: 0.0102 memory: 7124 grad_norm: 6.3196 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1196 loss: 1.1196 2022/09/07 11:31:24 - mmengine - INFO - Epoch(train) [65][1500/3757] lr: 1.0000e-03 eta: 5:53:31 time: 0.1597 data_time: 0.0102 memory: 7124 grad_norm: 6.2425 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1350 loss: 1.1350 2022/09/07 11:31:33 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:31:40 - mmengine - INFO - Epoch(train) [65][1600/3757] lr: 1.0000e-03 eta: 5:53:15 time: 0.1599 data_time: 0.0102 memory: 7124 grad_norm: 6.0958 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0728 loss: 1.0728 2022/09/07 11:31:56 - mmengine - INFO - Epoch(train) [65][1700/3757] lr: 1.0000e-03 eta: 5:52:59 time: 0.1569 data_time: 0.0096 memory: 7124 grad_norm: 6.1780 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1803 loss: 1.1803 2022/09/07 11:32:13 - mmengine - INFO - Epoch(train) [65][1800/3757] lr: 1.0000e-03 eta: 5:52:43 time: 0.1582 data_time: 0.0097 memory: 7124 grad_norm: 6.1795 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1362 loss: 1.1362 2022/09/07 11:32:29 - mmengine - INFO - Epoch(train) [65][1900/3757] lr: 1.0000e-03 eta: 5:52:27 time: 0.1587 data_time: 0.0103 memory: 7124 grad_norm: 6.5024 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1535 loss: 1.1535 2022/09/07 11:32:45 - mmengine - INFO - Epoch(train) [65][2000/3757] lr: 1.0000e-03 eta: 5:52:12 time: 0.1571 data_time: 0.0118 memory: 7124 grad_norm: 6.4556 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3366 loss: 1.3366 2022/09/07 11:33:01 - mmengine - INFO - Epoch(train) [65][2100/3757] lr: 1.0000e-03 eta: 5:51:56 time: 0.1604 data_time: 0.0099 memory: 7124 grad_norm: 6.3281 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0718 loss: 1.0718 2022/09/07 11:33:17 - mmengine - INFO - Epoch(train) [65][2200/3757] lr: 1.0000e-03 eta: 5:51:40 time: 0.1598 data_time: 0.0103 memory: 7124 grad_norm: 6.4861 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8784 loss: 0.8784 2022/09/07 11:33:33 - mmengine - INFO - Epoch(train) [65][2300/3757] lr: 1.0000e-03 eta: 5:51:24 time: 0.1692 data_time: 0.0126 memory: 7124 grad_norm: 6.3338 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0854 loss: 1.0854 2022/09/07 11:33:49 - mmengine - INFO - Epoch(train) [65][2400/3757] lr: 1.0000e-03 eta: 5:51:09 time: 0.1569 data_time: 0.0102 memory: 7124 grad_norm: 6.4063 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1332 loss: 1.1332 2022/09/07 11:34:05 - mmengine - INFO - Epoch(train) [65][2500/3757] lr: 1.0000e-03 eta: 5:50:53 time: 0.1603 data_time: 0.0102 memory: 7124 grad_norm: 6.4233 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1471 loss: 1.1471 2022/09/07 11:34:14 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:34:22 - mmengine - INFO - Epoch(train) [65][2600/3757] lr: 1.0000e-03 eta: 5:50:37 time: 0.1639 data_time: 0.0100 memory: 7124 grad_norm: 6.5434 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.3938 loss: 1.3938 2022/09/07 11:34:38 - mmengine - INFO - Epoch(train) [65][2700/3757] lr: 1.0000e-03 eta: 5:50:21 time: 0.1590 data_time: 0.0106 memory: 7124 grad_norm: 6.3690 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4019 loss: 1.4019 2022/09/07 11:34:54 - mmengine - INFO - Epoch(train) [65][2800/3757] lr: 1.0000e-03 eta: 5:50:06 time: 0.1647 data_time: 0.0106 memory: 7124 grad_norm: 6.2857 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2475 loss: 1.2475 2022/09/07 11:35:10 - mmengine - INFO - Epoch(train) [65][2900/3757] lr: 1.0000e-03 eta: 5:49:50 time: 0.1606 data_time: 0.0111 memory: 7124 grad_norm: 6.5489 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3212 loss: 1.3212 2022/09/07 11:35:26 - mmengine - INFO - Epoch(train) [65][3000/3757] lr: 1.0000e-03 eta: 5:49:34 time: 0.1600 data_time: 0.0107 memory: 7124 grad_norm: 6.3506 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2030 loss: 1.2030 2022/09/07 11:35:42 - mmengine - INFO - Epoch(train) [65][3100/3757] lr: 1.0000e-03 eta: 5:49:18 time: 0.1568 data_time: 0.0099 memory: 7124 grad_norm: 6.3941 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3723 loss: 1.3723 2022/09/07 11:35:58 - mmengine - INFO - Epoch(train) [65][3200/3757] lr: 1.0000e-03 eta: 5:49:02 time: 0.1599 data_time: 0.0120 memory: 7124 grad_norm: 6.5416 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2569 loss: 1.2569 2022/09/07 11:36:14 - mmengine - INFO - Epoch(train) [65][3300/3757] lr: 1.0000e-03 eta: 5:48:47 time: 0.1578 data_time: 0.0095 memory: 7124 grad_norm: 6.3909 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0601 loss: 1.0601 2022/09/07 11:36:30 - mmengine - INFO - Epoch(train) [65][3400/3757] lr: 1.0000e-03 eta: 5:48:31 time: 0.1612 data_time: 0.0107 memory: 7124 grad_norm: 6.4756 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1668 loss: 1.1668 2022/09/07 11:36:46 - mmengine - INFO - Epoch(train) [65][3500/3757] lr: 1.0000e-03 eta: 5:48:15 time: 0.1564 data_time: 0.0098 memory: 7124 grad_norm: 6.8886 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1570 loss: 1.1570 2022/09/07 11:36:55 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:37:02 - mmengine - INFO - Epoch(train) [65][3600/3757] lr: 1.0000e-03 eta: 5:47:59 time: 0.1599 data_time: 0.0109 memory: 7124 grad_norm: 6.4851 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.3277 loss: 1.3277 2022/09/07 11:37:18 - mmengine - INFO - Epoch(train) [65][3700/3757] lr: 1.0000e-03 eta: 5:47:43 time: 0.1578 data_time: 0.0105 memory: 7124 grad_norm: 6.3811 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.2385 loss: 1.2385 2022/09/07 11:37:27 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:37:27 - mmengine - INFO - Epoch(train) [65][3757/3757] lr: 1.0000e-03 eta: 5:47:37 time: 0.1379 data_time: 0.0074 memory: 7124 grad_norm: 6.3921 top1_acc: 0.7143 top5_acc: 0.7143 loss_cls: 1.2990 loss: 1.2990 2022/09/07 11:37:27 - mmengine - INFO - Saving checkpoint at 65 epochs 2022/09/07 11:39:44 - mmengine - INFO - Epoch(val) [65][100/310] eta: 0:03:43 time: 1.0626 data_time: 0.7570 memory: 7627 2022/09/07 11:42:04 - mmengine - INFO - Epoch(val) [65][200/310] eta: 0:02:29 time: 1.3585 data_time: 1.0533 memory: 7627 2022/09/07 11:44:09 - mmengine - INFO - Epoch(val) [65][300/310] eta: 0:00:10 time: 1.0943 data_time: 0.7951 memory: 7627 2022/09/07 11:44:26 - mmengine - INFO - Epoch(val) [65][310/310] acc/top1: 0.7434 acc/top5: 0.9139 acc/mean1: 0.7433 2022/09/07 11:44:26 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_60.pth is removed 2022/09/07 11:44:28 - mmengine - INFO - The best checkpoint with 0.7434 acc/top1 at 65 epoch is saved to best_acc/top1_epoch_65.pth. 2022/09/07 11:44:45 - mmengine - INFO - Epoch(train) [66][100/3757] lr: 1.0000e-03 eta: 5:47:18 time: 0.1605 data_time: 0.0101 memory: 7627 grad_norm: 6.3021 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2872 loss: 1.2872 2022/09/07 11:45:01 - mmengine - INFO - Epoch(train) [66][200/3757] lr: 1.0000e-03 eta: 5:47:02 time: 0.1578 data_time: 0.0108 memory: 7124 grad_norm: 6.8075 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3671 loss: 1.3671 2022/09/07 11:45:17 - mmengine - INFO - Epoch(train) [66][300/3757] lr: 1.0000e-03 eta: 5:46:46 time: 0.1570 data_time: 0.0106 memory: 7124 grad_norm: 6.3407 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0693 loss: 1.0693 2022/09/07 11:45:33 - mmengine - INFO - Epoch(train) [66][400/3757] lr: 1.0000e-03 eta: 5:46:30 time: 0.1598 data_time: 0.0106 memory: 7124 grad_norm: 6.4500 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1703 loss: 1.1703 2022/09/07 11:45:49 - mmengine - INFO - Epoch(train) [66][500/3757] lr: 1.0000e-03 eta: 5:46:14 time: 0.1593 data_time: 0.0120 memory: 7124 grad_norm: 6.8015 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2103 loss: 1.2103 2022/09/07 11:46:05 - mmengine - INFO - Epoch(train) [66][600/3757] lr: 1.0000e-03 eta: 5:45:59 time: 0.1618 data_time: 0.0111 memory: 7124 grad_norm: 6.5493 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2863 loss: 1.2863 2022/09/07 11:46:21 - mmengine - INFO - Epoch(train) [66][700/3757] lr: 1.0000e-03 eta: 5:45:43 time: 0.1597 data_time: 0.0104 memory: 7124 grad_norm: 6.5251 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9291 loss: 0.9291 2022/09/07 11:46:36 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:46:37 - mmengine - INFO - Epoch(train) [66][800/3757] lr: 1.0000e-03 eta: 5:45:27 time: 0.1547 data_time: 0.0106 memory: 7124 grad_norm: 6.6823 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2288 loss: 1.2288 2022/09/07 11:46:53 - mmengine - INFO - Epoch(train) [66][900/3757] lr: 1.0000e-03 eta: 5:45:11 time: 0.1594 data_time: 0.0099 memory: 7124 grad_norm: 6.5974 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3407 loss: 1.3407 2022/09/07 11:47:09 - mmengine - INFO - Epoch(train) [66][1000/3757] lr: 1.0000e-03 eta: 5:44:55 time: 0.1606 data_time: 0.0121 memory: 7124 grad_norm: 6.3223 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1879 loss: 1.1879 2022/09/07 11:47:26 - mmengine - INFO - Epoch(train) [66][1100/3757] lr: 1.0000e-03 eta: 5:44:40 time: 0.1617 data_time: 0.0108 memory: 7124 grad_norm: 6.2857 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.2519 loss: 1.2519 2022/09/07 11:47:42 - mmengine - INFO - Epoch(train) [66][1200/3757] lr: 1.0000e-03 eta: 5:44:24 time: 0.1584 data_time: 0.0100 memory: 7124 grad_norm: 6.4811 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9193 loss: 0.9193 2022/09/07 11:47:58 - mmengine - INFO - Epoch(train) [66][1300/3757] lr: 1.0000e-03 eta: 5:44:08 time: 0.1617 data_time: 0.0103 memory: 7124 grad_norm: 6.6941 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4034 loss: 1.4034 2022/09/07 11:48:14 - mmengine - INFO - Epoch(train) [66][1400/3757] lr: 1.0000e-03 eta: 5:43:52 time: 0.1627 data_time: 0.0095 memory: 7124 grad_norm: 6.2841 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2973 loss: 1.2973 2022/09/07 11:48:30 - mmengine - INFO - Epoch(train) [66][1500/3757] lr: 1.0000e-03 eta: 5:43:37 time: 0.1570 data_time: 0.0094 memory: 7124 grad_norm: 6.3040 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.2753 loss: 1.2753 2022/09/07 11:48:46 - mmengine - INFO - Epoch(train) [66][1600/3757] lr: 1.0000e-03 eta: 5:43:21 time: 0.1606 data_time: 0.0093 memory: 7124 grad_norm: 6.4998 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1478 loss: 1.1478 2022/09/07 11:49:02 - mmengine - INFO - Epoch(train) [66][1700/3757] lr: 1.0000e-03 eta: 5:43:05 time: 0.1593 data_time: 0.0099 memory: 7124 grad_norm: 6.2684 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2504 loss: 1.2504 2022/09/07 11:49:17 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:49:18 - mmengine - INFO - Epoch(train) [66][1800/3757] lr: 1.0000e-03 eta: 5:42:49 time: 0.1601 data_time: 0.0111 memory: 7124 grad_norm: 6.4806 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0778 loss: 1.0778 2022/09/07 11:49:34 - mmengine - INFO - Epoch(train) [66][1900/3757] lr: 1.0000e-03 eta: 5:42:33 time: 0.1605 data_time: 0.0106 memory: 7124 grad_norm: 6.4697 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3633 loss: 1.3633 2022/09/07 11:49:51 - mmengine - INFO - Epoch(train) [66][2000/3757] lr: 1.0000e-03 eta: 5:42:18 time: 0.1644 data_time: 0.0098 memory: 7124 grad_norm: 6.3315 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2251 loss: 1.2251 2022/09/07 11:50:07 - mmengine - INFO - Epoch(train) [66][2100/3757] lr: 1.0000e-03 eta: 5:42:02 time: 0.1634 data_time: 0.0111 memory: 7124 grad_norm: 6.6797 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.4559 loss: 1.4559 2022/09/07 11:50:23 - mmengine - INFO - Epoch(train) [66][2200/3757] lr: 1.0000e-03 eta: 5:41:46 time: 0.1591 data_time: 0.0112 memory: 7124 grad_norm: 6.5855 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0345 loss: 1.0345 2022/09/07 11:50:39 - mmengine - INFO - Epoch(train) [66][2300/3757] lr: 1.0000e-03 eta: 5:41:30 time: 0.1601 data_time: 0.0105 memory: 7124 grad_norm: 6.4582 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.3190 loss: 1.3190 2022/09/07 11:50:55 - mmengine - INFO - Epoch(train) [66][2400/3757] lr: 1.0000e-03 eta: 5:41:15 time: 0.1691 data_time: 0.0097 memory: 7124 grad_norm: 6.2127 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1108 loss: 1.1108 2022/09/07 11:51:12 - mmengine - INFO - Epoch(train) [66][2500/3757] lr: 1.0000e-03 eta: 5:40:59 time: 0.1608 data_time: 0.0106 memory: 7124 grad_norm: 6.5679 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3148 loss: 1.3148 2022/09/07 11:51:28 - mmengine - INFO - Epoch(train) [66][2600/3757] lr: 1.0000e-03 eta: 5:40:43 time: 0.1584 data_time: 0.0109 memory: 7124 grad_norm: 6.4787 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.3192 loss: 1.3192 2022/09/07 11:51:44 - mmengine - INFO - Epoch(train) [66][2700/3757] lr: 1.0000e-03 eta: 5:40:28 time: 0.1594 data_time: 0.0096 memory: 7124 grad_norm: 6.2767 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.2859 loss: 1.2859 2022/09/07 11:51:59 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:52:00 - mmengine - INFO - Epoch(train) [66][2800/3757] lr: 1.0000e-03 eta: 5:40:12 time: 0.1571 data_time: 0.0113 memory: 7124 grad_norm: 6.7307 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.0647 loss: 1.0647 2022/09/07 11:52:16 - mmengine - INFO - Epoch(train) [66][2900/3757] lr: 1.0000e-03 eta: 5:39:56 time: 0.1604 data_time: 0.0107 memory: 7124 grad_norm: 6.4853 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0415 loss: 1.0415 2022/09/07 11:52:32 - mmengine - INFO - Epoch(train) [66][3000/3757] lr: 1.0000e-03 eta: 5:39:40 time: 0.1594 data_time: 0.0101 memory: 7124 grad_norm: 6.2211 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9523 loss: 0.9523 2022/09/07 11:52:49 - mmengine - INFO - Epoch(train) [66][3100/3757] lr: 1.0000e-03 eta: 5:39:25 time: 0.1587 data_time: 0.0109 memory: 7124 grad_norm: 6.4733 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2050 loss: 1.2050 2022/09/07 11:53:05 - mmengine - INFO - Epoch(train) [66][3200/3757] lr: 1.0000e-03 eta: 5:39:09 time: 0.1576 data_time: 0.0099 memory: 7124 grad_norm: 6.4596 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9804 loss: 0.9804 2022/09/07 11:53:21 - mmengine - INFO - Epoch(train) [66][3300/3757] lr: 1.0000e-03 eta: 5:38:53 time: 0.1589 data_time: 0.0097 memory: 7124 grad_norm: 6.2570 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2198 loss: 1.2198 2022/09/07 11:53:37 - mmengine - INFO - Epoch(train) [66][3400/3757] lr: 1.0000e-03 eta: 5:38:37 time: 0.1598 data_time: 0.0109 memory: 7124 grad_norm: 6.3039 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9139 loss: 0.9139 2022/09/07 11:53:53 - mmengine - INFO - Epoch(train) [66][3500/3757] lr: 1.0000e-03 eta: 5:38:21 time: 0.1576 data_time: 0.0100 memory: 7124 grad_norm: 6.2794 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9933 loss: 0.9933 2022/09/07 11:54:09 - mmengine - INFO - Epoch(train) [66][3600/3757] lr: 1.0000e-03 eta: 5:38:06 time: 0.1603 data_time: 0.0104 memory: 7124 grad_norm: 6.4792 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2707 loss: 1.2707 2022/09/07 11:54:25 - mmengine - INFO - Epoch(train) [66][3700/3757] lr: 1.0000e-03 eta: 5:37:50 time: 0.1595 data_time: 0.0106 memory: 7124 grad_norm: 6.7304 top1_acc: 0.2500 top5_acc: 0.7500 loss_cls: 1.2129 loss: 1.2129 2022/09/07 11:54:34 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 11:54:34 - mmengine - INFO - Epoch(train) [66][3757/3757] lr: 1.0000e-03 eta: 5:37:44 time: 0.1371 data_time: 0.0068 memory: 7124 grad_norm: 6.4592 top1_acc: 0.5714 top5_acc: 0.8571 loss_cls: 1.3252 loss: 1.3252 2022/09/07 11:54:34 - mmengine - INFO - Saving checkpoint at 66 epochs 2022/09/07 11:56:54 - mmengine - INFO - Epoch(val) [66][100/310] eta: 0:04:29 time: 1.2845 data_time: 0.9828 memory: 7627 2022/09/07 11:59:06 - mmengine - INFO - Epoch(val) [66][200/310] eta: 0:02:18 time: 1.2550 data_time: 0.9501 memory: 7627 2022/09/07 12:01:14 - mmengine - INFO - Epoch(val) [66][300/310] eta: 0:00:13 time: 1.3351 data_time: 1.0376 memory: 7627 2022/09/07 12:01:35 - mmengine - INFO - Epoch(val) [66][310/310] acc/top1: 0.7416 acc/top5: 0.9133 acc/mean1: 0.7416 2022/09/07 12:01:43 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:01:53 - mmengine - INFO - Epoch(train) [67][100/3757] lr: 1.0000e-03 eta: 5:37:24 time: 0.1584 data_time: 0.0115 memory: 7627 grad_norm: 6.5398 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3856 loss: 1.3856 2022/09/07 12:02:09 - mmengine - INFO - Epoch(train) [67][200/3757] lr: 1.0000e-03 eta: 5:37:09 time: 0.1603 data_time: 0.0102 memory: 7124 grad_norm: 6.6310 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8515 loss: 0.8515 2022/09/07 12:02:25 - mmengine - INFO - Epoch(train) [67][300/3757] lr: 1.0000e-03 eta: 5:36:53 time: 0.1574 data_time: 0.0098 memory: 7124 grad_norm: 6.5468 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.5903 loss: 1.5903 2022/09/07 12:02:41 - mmengine - INFO - Epoch(train) [67][400/3757] lr: 1.0000e-03 eta: 5:36:37 time: 0.1581 data_time: 0.0095 memory: 7124 grad_norm: 6.5655 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1266 loss: 1.1266 2022/09/07 12:02:57 - mmengine - INFO - Epoch(train) [67][500/3757] lr: 1.0000e-03 eta: 5:36:21 time: 0.1596 data_time: 0.0095 memory: 7124 grad_norm: 6.4091 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0635 loss: 1.0635 2022/09/07 12:03:13 - mmengine - INFO - Epoch(train) [67][600/3757] lr: 1.0000e-03 eta: 5:36:06 time: 0.1601 data_time: 0.0105 memory: 7124 grad_norm: 6.5406 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0866 loss: 1.0866 2022/09/07 12:03:29 - mmengine - INFO - Epoch(train) [67][700/3757] lr: 1.0000e-03 eta: 5:35:50 time: 0.1603 data_time: 0.0117 memory: 7124 grad_norm: 6.2726 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0742 loss: 1.0742 2022/09/07 12:03:45 - mmengine - INFO - Epoch(train) [67][800/3757] lr: 1.0000e-03 eta: 5:35:34 time: 0.1563 data_time: 0.0104 memory: 7124 grad_norm: 6.2738 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9612 loss: 0.9612 2022/09/07 12:04:01 - mmengine - INFO - Epoch(train) [67][900/3757] lr: 1.0000e-03 eta: 5:35:18 time: 0.1610 data_time: 0.0100 memory: 7124 grad_norm: 6.4971 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2592 loss: 1.2592 2022/09/07 12:04:18 - mmengine - INFO - Epoch(train) [67][1000/3757] lr: 1.0000e-03 eta: 5:35:02 time: 0.1602 data_time: 0.0127 memory: 7124 grad_norm: 6.6640 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.4496 loss: 1.4496 2022/09/07 12:04:24 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:04:34 - mmengine - INFO - Epoch(train) [67][1100/3757] lr: 1.0000e-03 eta: 5:34:47 time: 0.1602 data_time: 0.0107 memory: 7124 grad_norm: 6.5232 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1006 loss: 1.1006 2022/09/07 12:04:50 - mmengine - INFO - Epoch(train) [67][1200/3757] lr: 1.0000e-03 eta: 5:34:31 time: 0.1581 data_time: 0.0102 memory: 7124 grad_norm: 6.4203 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0949 loss: 1.0949 2022/09/07 12:05:06 - mmengine - INFO - Epoch(train) [67][1300/3757] lr: 1.0000e-03 eta: 5:34:15 time: 0.1667 data_time: 0.0118 memory: 7124 grad_norm: 6.6670 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.3219 loss: 1.3219 2022/09/07 12:05:22 - mmengine - INFO - Epoch(train) [67][1400/3757] lr: 1.0000e-03 eta: 5:33:59 time: 0.1580 data_time: 0.0102 memory: 7124 grad_norm: 6.2598 top1_acc: 0.3750 top5_acc: 0.3750 loss_cls: 1.0530 loss: 1.0530 2022/09/07 12:05:38 - mmengine - INFO - Epoch(train) [67][1500/3757] lr: 1.0000e-03 eta: 5:33:43 time: 0.1587 data_time: 0.0098 memory: 7124 grad_norm: 6.6311 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1708 loss: 1.1708 2022/09/07 12:05:54 - mmengine - INFO - Epoch(train) [67][1600/3757] lr: 1.0000e-03 eta: 5:33:28 time: 0.1567 data_time: 0.0092 memory: 7124 grad_norm: 6.4510 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0117 loss: 1.0117 2022/09/07 12:06:10 - mmengine - INFO - Epoch(train) [67][1700/3757] lr: 1.0000e-03 eta: 5:33:12 time: 0.1613 data_time: 0.0101 memory: 7124 grad_norm: 6.2578 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 0.9706 loss: 0.9706 2022/09/07 12:06:26 - mmengine - INFO - Epoch(train) [67][1800/3757] lr: 1.0000e-03 eta: 5:32:56 time: 0.1665 data_time: 0.0103 memory: 7124 grad_norm: 6.4830 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2189 loss: 1.2189 2022/09/07 12:06:42 - mmengine - INFO - Epoch(train) [67][1900/3757] lr: 1.0000e-03 eta: 5:32:40 time: 0.1583 data_time: 0.0103 memory: 7124 grad_norm: 6.2222 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.1090 loss: 1.1090 2022/09/07 12:06:58 - mmengine - INFO - Epoch(train) [67][2000/3757] lr: 1.0000e-03 eta: 5:32:25 time: 0.1597 data_time: 0.0087 memory: 7124 grad_norm: 6.4164 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.3155 loss: 1.3155 2022/09/07 12:07:05 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:07:15 - mmengine - INFO - Epoch(train) [67][2100/3757] lr: 1.0000e-03 eta: 5:32:09 time: 0.1590 data_time: 0.0104 memory: 7124 grad_norm: 6.3194 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1441 loss: 1.1441 2022/09/07 12:07:31 - mmengine - INFO - Epoch(train) [67][2200/3757] lr: 1.0000e-03 eta: 5:31:53 time: 0.1587 data_time: 0.0102 memory: 7124 grad_norm: 6.5763 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.4573 loss: 1.4573 2022/09/07 12:07:47 - mmengine - INFO - Epoch(train) [67][2300/3757] lr: 1.0000e-03 eta: 5:31:37 time: 0.1606 data_time: 0.0086 memory: 7124 grad_norm: 6.4596 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.3235 loss: 1.3235 2022/09/07 12:08:03 - mmengine - INFO - Epoch(train) [67][2400/3757] lr: 1.0000e-03 eta: 5:31:21 time: 0.1585 data_time: 0.0101 memory: 7124 grad_norm: 6.4403 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1254 loss: 1.1254 2022/09/07 12:08:19 - mmengine - INFO - Epoch(train) [67][2500/3757] lr: 1.0000e-03 eta: 5:31:06 time: 0.1608 data_time: 0.0114 memory: 7124 grad_norm: 6.5798 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0688 loss: 1.0688 2022/09/07 12:08:35 - mmengine - INFO - Epoch(train) [67][2600/3757] lr: 1.0000e-03 eta: 5:30:50 time: 0.1580 data_time: 0.0102 memory: 7124 grad_norm: 6.6221 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.4826 loss: 1.4826 2022/09/07 12:08:51 - mmengine - INFO - Epoch(train) [67][2700/3757] lr: 1.0000e-03 eta: 5:30:34 time: 0.1557 data_time: 0.0102 memory: 7124 grad_norm: 6.3071 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0702 loss: 1.0702 2022/09/07 12:09:07 - mmengine - INFO - Epoch(train) [67][2800/3757] lr: 1.0000e-03 eta: 5:30:18 time: 0.1597 data_time: 0.0105 memory: 7124 grad_norm: 6.2207 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 0.9710 loss: 0.9710 2022/09/07 12:09:23 - mmengine - INFO - Epoch(train) [67][2900/3757] lr: 1.0000e-03 eta: 5:30:02 time: 0.1606 data_time: 0.0103 memory: 7124 grad_norm: 6.1497 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1935 loss: 1.1935 2022/09/07 12:09:39 - mmengine - INFO - Epoch(train) [67][3000/3757] lr: 1.0000e-03 eta: 5:29:47 time: 0.1590 data_time: 0.0115 memory: 7124 grad_norm: 6.4805 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1834 loss: 1.1834 2022/09/07 12:09:45 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:09:55 - mmengine - INFO - Epoch(train) [67][3100/3757] lr: 1.0000e-03 eta: 5:29:31 time: 0.1612 data_time: 0.0092 memory: 7124 grad_norm: 6.6129 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.2281 loss: 1.2281 2022/09/07 12:10:12 - mmengine - INFO - Epoch(train) [67][3200/3757] lr: 1.0000e-03 eta: 5:29:15 time: 0.1787 data_time: 0.0103 memory: 7124 grad_norm: 6.5479 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5127 loss: 1.5127 2022/09/07 12:10:28 - mmengine - INFO - Epoch(train) [67][3300/3757] lr: 1.0000e-03 eta: 5:28:59 time: 0.1592 data_time: 0.0102 memory: 7124 grad_norm: 6.6873 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0879 loss: 1.0879 2022/09/07 12:10:44 - mmengine - INFO - Epoch(train) [67][3400/3757] lr: 1.0000e-03 eta: 5:28:44 time: 0.1603 data_time: 0.0096 memory: 7124 grad_norm: 6.6879 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.4119 loss: 1.4119 2022/09/07 12:11:00 - mmengine - INFO - Epoch(train) [67][3500/3757] lr: 1.0000e-03 eta: 5:28:28 time: 0.1578 data_time: 0.0106 memory: 7124 grad_norm: 6.5651 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2615 loss: 1.2615 2022/09/07 12:11:16 - mmengine - INFO - Epoch(train) [67][3600/3757] lr: 1.0000e-03 eta: 5:28:12 time: 0.1582 data_time: 0.0097 memory: 7124 grad_norm: 6.5844 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1344 loss: 1.1344 2022/09/07 12:11:32 - mmengine - INFO - Epoch(train) [67][3700/3757] lr: 1.0000e-03 eta: 5:27:56 time: 0.1583 data_time: 0.0099 memory: 7124 grad_norm: 6.3903 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0511 loss: 1.0511 2022/09/07 12:11:41 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:11:41 - mmengine - INFO - Epoch(train) [67][3757/3757] lr: 1.0000e-03 eta: 5:27:50 time: 0.1364 data_time: 0.0072 memory: 7124 grad_norm: 6.5779 top1_acc: 0.8571 top5_acc: 0.8571 loss_cls: 1.1197 loss: 1.1197 2022/09/07 12:11:41 - mmengine - INFO - Saving checkpoint at 67 epochs 2022/09/07 12:14:01 - mmengine - INFO - Epoch(val) [67][100/310] eta: 0:04:27 time: 1.2717 data_time: 0.9675 memory: 7627 2022/09/07 12:16:16 - mmengine - INFO - Epoch(val) [67][200/310] eta: 0:02:11 time: 1.1940 data_time: 0.8878 memory: 7627 2022/09/07 12:18:22 - mmengine - INFO - Epoch(val) [67][300/310] eta: 0:00:12 time: 1.2522 data_time: 0.9482 memory: 7627 2022/09/07 12:18:39 - mmengine - INFO - Epoch(val) [67][310/310] acc/top1: 0.7455 acc/top5: 0.9151 acc/mean1: 0.7455 2022/09/07 12:18:39 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_65.pth is removed 2022/09/07 12:18:41 - mmengine - INFO - The best checkpoint with 0.7455 acc/top1 at 67 epoch is saved to best_acc/top1_epoch_67.pth. 2022/09/07 12:18:58 - mmengine - INFO - Epoch(train) [68][100/3757] lr: 1.0000e-03 eta: 5:27:31 time: 0.1580 data_time: 0.0095 memory: 7627 grad_norm: 6.3599 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2707 loss: 1.2707 2022/09/07 12:19:15 - mmengine - INFO - Epoch(train) [68][200/3757] lr: 1.0000e-03 eta: 5:27:15 time: 0.1591 data_time: 0.0102 memory: 7124 grad_norm: 6.7212 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2482 loss: 1.2482 2022/09/07 12:19:28 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:19:31 - mmengine - INFO - Epoch(train) [68][300/3757] lr: 1.0000e-03 eta: 5:26:59 time: 0.1614 data_time: 0.0098 memory: 7124 grad_norm: 6.4184 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1270 loss: 1.1270 2022/09/07 12:19:47 - mmengine - INFO - Epoch(train) [68][400/3757] lr: 1.0000e-03 eta: 5:26:43 time: 0.1651 data_time: 0.0105 memory: 7124 grad_norm: 6.5014 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9195 loss: 0.9195 2022/09/07 12:20:03 - mmengine - INFO - Epoch(train) [68][500/3757] lr: 1.0000e-03 eta: 5:26:28 time: 0.1589 data_time: 0.0104 memory: 7124 grad_norm: 6.5036 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1841 loss: 1.1841 2022/09/07 12:20:20 - mmengine - INFO - Epoch(train) [68][600/3757] lr: 1.0000e-03 eta: 5:26:12 time: 0.1655 data_time: 0.0105 memory: 7124 grad_norm: 6.7330 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2100 loss: 1.2100 2022/09/07 12:20:36 - mmengine - INFO - Epoch(train) [68][700/3757] lr: 1.0000e-03 eta: 5:25:56 time: 0.1614 data_time: 0.0100 memory: 7124 grad_norm: 6.6367 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0953 loss: 1.0953 2022/09/07 12:20:52 - mmengine - INFO - Epoch(train) [68][800/3757] lr: 1.0000e-03 eta: 5:25:41 time: 0.1588 data_time: 0.0109 memory: 7124 grad_norm: 6.5616 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2914 loss: 1.2914 2022/09/07 12:21:08 - mmengine - INFO - Epoch(train) [68][900/3757] lr: 1.0000e-03 eta: 5:25:25 time: 0.1628 data_time: 0.0113 memory: 7124 grad_norm: 6.5058 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.9154 loss: 0.9154 2022/09/07 12:21:24 - mmengine - INFO - Epoch(train) [68][1000/3757] lr: 1.0000e-03 eta: 5:25:09 time: 0.1582 data_time: 0.0097 memory: 7124 grad_norm: 6.5750 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1483 loss: 1.1483 2022/09/07 12:21:41 - mmengine - INFO - Epoch(train) [68][1100/3757] lr: 1.0000e-03 eta: 5:24:53 time: 0.1582 data_time: 0.0111 memory: 7124 grad_norm: 6.5463 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.2000 loss: 1.2000 2022/09/07 12:21:57 - mmengine - INFO - Epoch(train) [68][1200/3757] lr: 1.0000e-03 eta: 5:24:37 time: 0.1602 data_time: 0.0108 memory: 7124 grad_norm: 6.7238 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0361 loss: 1.0361 2022/09/07 12:22:10 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:22:13 - mmengine - INFO - Epoch(train) [68][1300/3757] lr: 1.0000e-03 eta: 5:24:22 time: 0.1578 data_time: 0.0106 memory: 7124 grad_norm: 6.7938 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2931 loss: 1.2931 2022/09/07 12:22:29 - mmengine - INFO - Epoch(train) [68][1400/3757] lr: 1.0000e-03 eta: 5:24:06 time: 0.1625 data_time: 0.0108 memory: 7124 grad_norm: 6.7142 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3438 loss: 1.3438 2022/09/07 12:22:45 - mmengine - INFO - Epoch(train) [68][1500/3757] lr: 1.0000e-03 eta: 5:23:50 time: 0.1629 data_time: 0.0094 memory: 7124 grad_norm: 6.4629 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.1582 loss: 1.1582 2022/09/07 12:23:01 - mmengine - INFO - Epoch(train) [68][1600/3757] lr: 1.0000e-03 eta: 5:23:34 time: 0.1577 data_time: 0.0092 memory: 7124 grad_norm: 6.6220 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.9828 loss: 0.9828 2022/09/07 12:23:17 - mmengine - INFO - Epoch(train) [68][1700/3757] lr: 1.0000e-03 eta: 5:23:19 time: 0.1603 data_time: 0.0099 memory: 7124 grad_norm: 6.4116 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1374 loss: 1.1374 2022/09/07 12:23:33 - mmengine - INFO - Epoch(train) [68][1800/3757] lr: 1.0000e-03 eta: 5:23:03 time: 0.1601 data_time: 0.0106 memory: 7124 grad_norm: 6.4951 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 0.9385 loss: 0.9385 2022/09/07 12:23:50 - mmengine - INFO - Epoch(train) [68][1900/3757] lr: 1.0000e-03 eta: 5:22:47 time: 0.1629 data_time: 0.0100 memory: 7124 grad_norm: 6.2497 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8555 loss: 0.8555 2022/09/07 12:24:06 - mmengine - INFO - Epoch(train) [68][2000/3757] lr: 1.0000e-03 eta: 5:22:31 time: 0.1605 data_time: 0.0098 memory: 7124 grad_norm: 6.6313 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2006 loss: 1.2006 2022/09/07 12:24:22 - mmengine - INFO - Epoch(train) [68][2100/3757] lr: 1.0000e-03 eta: 5:22:16 time: 0.1580 data_time: 0.0098 memory: 7124 grad_norm: 6.8800 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.1614 loss: 1.1614 2022/09/07 12:24:39 - mmengine - INFO - Epoch(train) [68][2200/3757] lr: 1.0000e-03 eta: 5:22:00 time: 0.1725 data_time: 0.0112 memory: 7124 grad_norm: 6.3482 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1146 loss: 1.1146 2022/09/07 12:24:52 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:24:55 - mmengine - INFO - Epoch(train) [68][2300/3757] lr: 1.0000e-03 eta: 5:21:44 time: 0.1598 data_time: 0.0111 memory: 7124 grad_norm: 6.8611 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.3850 loss: 1.3850 2022/09/07 12:25:11 - mmengine - INFO - Epoch(train) [68][2400/3757] lr: 1.0000e-03 eta: 5:21:28 time: 0.1592 data_time: 0.0103 memory: 7124 grad_norm: 6.6642 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.3404 loss: 1.3404 2022/09/07 12:25:27 - mmengine - INFO - Epoch(train) [68][2500/3757] lr: 1.0000e-03 eta: 5:21:13 time: 0.1592 data_time: 0.0097 memory: 7124 grad_norm: 6.6600 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0993 loss: 1.0993 2022/09/07 12:25:43 - mmengine - INFO - Epoch(train) [68][2600/3757] lr: 1.0000e-03 eta: 5:20:57 time: 0.1593 data_time: 0.0110 memory: 7124 grad_norm: 6.7682 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.3372 loss: 1.3372 2022/09/07 12:25:59 - mmengine - INFO - Epoch(train) [68][2700/3757] lr: 1.0000e-03 eta: 5:20:41 time: 0.1640 data_time: 0.0096 memory: 7124 grad_norm: 6.5028 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.1876 loss: 1.1876 2022/09/07 12:26:15 - mmengine - INFO - Epoch(train) [68][2800/3757] lr: 1.0000e-03 eta: 5:20:25 time: 0.1578 data_time: 0.0093 memory: 7124 grad_norm: 6.5412 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1350 loss: 1.1350 2022/09/07 12:26:31 - mmengine - INFO - Epoch(train) [68][2900/3757] lr: 1.0000e-03 eta: 5:20:09 time: 0.1639 data_time: 0.0111 memory: 7124 grad_norm: 7.0412 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.1717 loss: 1.1717 2022/09/07 12:26:48 - mmengine - INFO - Epoch(train) [68][3000/3757] lr: 1.0000e-03 eta: 5:19:54 time: 0.1610 data_time: 0.0095 memory: 7124 grad_norm: 6.8633 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2957 loss: 1.2957 2022/09/07 12:27:04 - mmengine - INFO - Epoch(train) [68][3100/3757] lr: 1.0000e-03 eta: 5:19:38 time: 0.1587 data_time: 0.0106 memory: 7124 grad_norm: 6.5131 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0243 loss: 1.0243 2022/09/07 12:27:20 - mmengine - INFO - Epoch(train) [68][3200/3757] lr: 1.0000e-03 eta: 5:19:22 time: 0.1632 data_time: 0.0117 memory: 7124 grad_norm: 6.3098 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.1846 loss: 1.1846 2022/09/07 12:27:33 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:27:36 - mmengine - INFO - Epoch(train) [68][3300/3757] lr: 1.0000e-03 eta: 5:19:06 time: 0.1590 data_time: 0.0105 memory: 7124 grad_norm: 6.6222 top1_acc: 0.2500 top5_acc: 0.3750 loss_cls: 1.2918 loss: 1.2918 2022/09/07 12:27:52 - mmengine - INFO - Epoch(train) [68][3400/3757] lr: 1.0000e-03 eta: 5:18:50 time: 0.1575 data_time: 0.0099 memory: 7124 grad_norm: 6.4930 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0524 loss: 1.0524 2022/09/07 12:28:08 - mmengine - INFO - Epoch(train) [68][3500/3757] lr: 1.0000e-03 eta: 5:18:35 time: 0.1631 data_time: 0.0121 memory: 7124 grad_norm: 6.1849 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9712 loss: 0.9712 2022/09/07 12:28:24 - mmengine - INFO - Epoch(train) [68][3600/3757] lr: 1.0000e-03 eta: 5:18:19 time: 0.1562 data_time: 0.0105 memory: 7124 grad_norm: 6.6296 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2208 loss: 1.2208 2022/09/07 12:28:40 - mmengine - INFO - Epoch(train) [68][3700/3757] lr: 1.0000e-03 eta: 5:18:03 time: 0.1550 data_time: 0.0101 memory: 7124 grad_norm: 6.5391 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2780 loss: 1.2780 2022/09/07 12:28:49 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:28:49 - mmengine - INFO - Epoch(train) [68][3757/3757] lr: 1.0000e-03 eta: 5:17:57 time: 0.1483 data_time: 0.0080 memory: 7124 grad_norm: 6.4114 top1_acc: 0.8571 top5_acc: 1.0000 loss_cls: 1.2725 loss: 1.2725 2022/09/07 12:28:49 - mmengine - INFO - Saving checkpoint at 68 epochs 2022/09/07 12:31:06 - mmengine - INFO - Epoch(val) [68][100/310] eta: 0:03:40 time: 1.0509 data_time: 0.7496 memory: 7627 2022/09/07 12:33:26 - mmengine - INFO - Epoch(val) [68][200/310] eta: 0:02:30 time: 1.3700 data_time: 1.0636 memory: 7627 2022/09/07 12:35:30 - mmengine - INFO - Epoch(val) [68][300/310] eta: 0:00:10 time: 1.0992 data_time: 0.8002 memory: 7627 2022/09/07 12:35:46 - mmengine - INFO - Epoch(val) [68][310/310] acc/top1: 0.7447 acc/top5: 0.9156 acc/mean1: 0.7446 2022/09/07 12:36:04 - mmengine - INFO - Epoch(train) [69][100/3757] lr: 1.0000e-03 eta: 5:17:38 time: 0.1635 data_time: 0.0106 memory: 7627 grad_norm: 6.5986 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1186 loss: 1.1186 2022/09/07 12:36:20 - mmengine - INFO - Epoch(train) [69][200/3757] lr: 1.0000e-03 eta: 5:17:22 time: 0.1655 data_time: 0.0107 memory: 7124 grad_norm: 6.5027 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0501 loss: 1.0501 2022/09/07 12:36:36 - mmengine - INFO - Epoch(train) [69][300/3757] lr: 1.0000e-03 eta: 5:17:06 time: 0.1602 data_time: 0.0096 memory: 7124 grad_norm: 6.7007 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0062 loss: 1.0062 2022/09/07 12:36:52 - mmengine - INFO - Epoch(train) [69][400/3757] lr: 1.0000e-03 eta: 5:16:51 time: 0.1608 data_time: 0.0106 memory: 7124 grad_norm: 6.6452 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.0695 loss: 1.0695 2022/09/07 12:37:08 - mmengine - INFO - Epoch(train) [69][500/3757] lr: 1.0000e-03 eta: 5:16:35 time: 0.1615 data_time: 0.0101 memory: 7124 grad_norm: 6.5694 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2213 loss: 1.2213 2022/09/07 12:37:12 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:37:24 - mmengine - INFO - Epoch(train) [69][600/3757] lr: 1.0000e-03 eta: 5:16:19 time: 0.1559 data_time: 0.0094 memory: 7124 grad_norm: 6.6995 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2625 loss: 1.2625 2022/09/07 12:37:41 - mmengine - INFO - Epoch(train) [69][700/3757] lr: 1.0000e-03 eta: 5:16:03 time: 0.1644 data_time: 0.0105 memory: 7124 grad_norm: 6.2867 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.9917 loss: 0.9917 2022/09/07 12:37:57 - mmengine - INFO - Epoch(train) [69][800/3757] lr: 1.0000e-03 eta: 5:15:47 time: 0.1599 data_time: 0.0093 memory: 7124 grad_norm: 6.7346 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1486 loss: 1.1486 2022/09/07 12:38:13 - mmengine - INFO - Epoch(train) [69][900/3757] lr: 1.0000e-03 eta: 5:15:32 time: 0.1573 data_time: 0.0107 memory: 7124 grad_norm: 6.3093 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9302 loss: 0.9302 2022/09/07 12:38:29 - mmengine - INFO - Epoch(train) [69][1000/3757] lr: 1.0000e-03 eta: 5:15:16 time: 0.1620 data_time: 0.0108 memory: 7124 grad_norm: 6.4835 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.5330 loss: 1.5330 2022/09/07 12:38:45 - mmengine - INFO - Epoch(train) [69][1100/3757] lr: 1.0000e-03 eta: 5:15:00 time: 0.1591 data_time: 0.0094 memory: 7124 grad_norm: 6.6697 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0474 loss: 1.0474 2022/09/07 12:39:02 - mmengine - INFO - Epoch(train) [69][1200/3757] lr: 1.0000e-03 eta: 5:14:44 time: 0.1606 data_time: 0.0097 memory: 7124 grad_norm: 6.4633 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2803 loss: 1.2803 2022/09/07 12:39:18 - mmengine - INFO - Epoch(train) [69][1300/3757] lr: 1.0000e-03 eta: 5:14:29 time: 0.1614 data_time: 0.0116 memory: 7124 grad_norm: 6.6620 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0897 loss: 1.0897 2022/09/07 12:39:34 - mmengine - INFO - Epoch(train) [69][1400/3757] lr: 1.0000e-03 eta: 5:14:13 time: 0.1601 data_time: 0.0104 memory: 7124 grad_norm: 6.7126 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1318 loss: 1.1318 2022/09/07 12:39:50 - mmengine - INFO - Epoch(train) [69][1500/3757] lr: 1.0000e-03 eta: 5:13:57 time: 0.1599 data_time: 0.0114 memory: 7124 grad_norm: 6.8473 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2987 loss: 1.2987 2022/09/07 12:39:54 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:40:06 - mmengine - INFO - Epoch(train) [69][1600/3757] lr: 1.0000e-03 eta: 5:13:41 time: 0.1618 data_time: 0.0102 memory: 7124 grad_norm: 6.6677 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0654 loss: 1.0654 2022/09/07 12:40:22 - mmengine - INFO - Epoch(train) [69][1700/3757] lr: 1.0000e-03 eta: 5:13:25 time: 0.1637 data_time: 0.0150 memory: 7124 grad_norm: 6.7405 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2139 loss: 1.2139 2022/09/07 12:40:38 - mmengine - INFO - Epoch(train) [69][1800/3757] lr: 1.0000e-03 eta: 5:13:10 time: 0.1589 data_time: 0.0086 memory: 7124 grad_norm: 6.6945 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0458 loss: 1.0458 2022/09/07 12:40:54 - mmengine - INFO - Epoch(train) [69][1900/3757] lr: 1.0000e-03 eta: 5:12:54 time: 0.1597 data_time: 0.0093 memory: 7124 grad_norm: 6.6214 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0961 loss: 1.0961 2022/09/07 12:41:11 - mmengine - INFO - Epoch(train) [69][2000/3757] lr: 1.0000e-03 eta: 5:12:38 time: 0.1590 data_time: 0.0107 memory: 7124 grad_norm: 6.7335 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2484 loss: 1.2484 2022/09/07 12:41:27 - mmengine - INFO - Epoch(train) [69][2100/3757] lr: 1.0000e-03 eta: 5:12:22 time: 0.1586 data_time: 0.0101 memory: 7124 grad_norm: 6.4117 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2332 loss: 1.2332 2022/09/07 12:41:43 - mmengine - INFO - Epoch(train) [69][2200/3757] lr: 1.0000e-03 eta: 5:12:07 time: 0.1712 data_time: 0.0104 memory: 7124 grad_norm: 6.7069 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0038 loss: 1.0038 2022/09/07 12:41:59 - mmengine - INFO - Epoch(train) [69][2300/3757] lr: 1.0000e-03 eta: 5:11:51 time: 0.1600 data_time: 0.0116 memory: 7124 grad_norm: 6.6596 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.4316 loss: 1.4316 2022/09/07 12:42:15 - mmengine - INFO - Epoch(train) [69][2400/3757] lr: 1.0000e-03 eta: 5:11:35 time: 0.1597 data_time: 0.0106 memory: 7124 grad_norm: 6.6583 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9592 loss: 0.9592 2022/09/07 12:42:31 - mmengine - INFO - Epoch(train) [69][2500/3757] lr: 1.0000e-03 eta: 5:11:19 time: 0.1613 data_time: 0.0097 memory: 7124 grad_norm: 6.5824 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2197 loss: 1.2197 2022/09/07 12:42:35 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:42:47 - mmengine - INFO - Epoch(train) [69][2600/3757] lr: 1.0000e-03 eta: 5:11:03 time: 0.1568 data_time: 0.0106 memory: 7124 grad_norm: 6.5116 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9576 loss: 0.9576 2022/09/07 12:43:03 - mmengine - INFO - Epoch(train) [69][2700/3757] lr: 1.0000e-03 eta: 5:10:48 time: 0.1597 data_time: 0.0099 memory: 7124 grad_norm: 6.6408 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.0289 loss: 1.0289 2022/09/07 12:43:20 - mmengine - INFO - Epoch(train) [69][2800/3757] lr: 1.0000e-03 eta: 5:10:32 time: 0.1648 data_time: 0.0119 memory: 7124 grad_norm: 6.8536 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2047 loss: 1.2047 2022/09/07 12:43:36 - mmengine - INFO - Epoch(train) [69][2900/3757] lr: 1.0000e-03 eta: 5:10:16 time: 0.1603 data_time: 0.0102 memory: 7124 grad_norm: 6.7173 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1582 loss: 1.1582 2022/09/07 12:43:52 - mmengine - INFO - Epoch(train) [69][3000/3757] lr: 1.0000e-03 eta: 5:10:00 time: 0.1611 data_time: 0.0097 memory: 7124 grad_norm: 7.0148 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.4251 loss: 1.4251 2022/09/07 12:44:08 - mmengine - INFO - Epoch(train) [69][3100/3757] lr: 1.0000e-03 eta: 5:09:44 time: 0.1560 data_time: 0.0090 memory: 7124 grad_norm: 6.2813 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9935 loss: 0.9935 2022/09/07 12:44:24 - mmengine - INFO - Epoch(train) [69][3200/3757] lr: 1.0000e-03 eta: 5:09:29 time: 0.1560 data_time: 0.0093 memory: 7124 grad_norm: 6.6596 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1409 loss: 1.1409 2022/09/07 12:44:40 - mmengine - INFO - Epoch(train) [69][3300/3757] lr: 1.0000e-03 eta: 5:09:13 time: 0.1642 data_time: 0.0116 memory: 7124 grad_norm: 6.4752 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9737 loss: 0.9737 2022/09/07 12:44:57 - mmengine - INFO - Epoch(train) [69][3400/3757] lr: 1.0000e-03 eta: 5:08:57 time: 0.1580 data_time: 0.0096 memory: 7124 grad_norm: 6.5720 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1229 loss: 1.1229 2022/09/07 12:45:13 - mmengine - INFO - Epoch(train) [69][3500/3757] lr: 1.0000e-03 eta: 5:08:41 time: 0.1614 data_time: 0.0111 memory: 7124 grad_norm: 6.5428 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0074 loss: 1.0074 2022/09/07 12:45:17 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:45:29 - mmengine - INFO - Epoch(train) [69][3600/3757] lr: 1.0000e-03 eta: 5:08:26 time: 0.1657 data_time: 0.0102 memory: 7124 grad_norm: 6.5765 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3711 loss: 1.3711 2022/09/07 12:45:45 - mmengine - INFO - Epoch(train) [69][3700/3757] lr: 1.0000e-03 eta: 5:08:10 time: 0.1578 data_time: 0.0099 memory: 7124 grad_norm: 6.8585 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1771 loss: 1.1771 2022/09/07 12:45:54 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:45:54 - mmengine - INFO - Epoch(train) [69][3757/3757] lr: 1.0000e-03 eta: 5:08:03 time: 0.1386 data_time: 0.0076 memory: 7124 grad_norm: 6.7137 top1_acc: 0.7143 top5_acc: 0.8571 loss_cls: 1.1715 loss: 1.1715 2022/09/07 12:45:54 - mmengine - INFO - Saving checkpoint at 69 epochs 2022/09/07 12:48:12 - mmengine - INFO - Epoch(val) [69][100/310] eta: 0:03:44 time: 1.0702 data_time: 0.7708 memory: 7627 2022/09/07 12:50:33 - mmengine - INFO - Epoch(val) [69][200/310] eta: 0:02:33 time: 1.3963 data_time: 1.0962 memory: 7627 2022/09/07 12:52:37 - mmengine - INFO - Epoch(val) [69][300/310] eta: 0:00:10 time: 1.0587 data_time: 0.7610 memory: 7627 2022/09/07 12:52:50 - mmengine - INFO - Epoch(val) [69][310/310] acc/top1: 0.7443 acc/top5: 0.9145 acc/mean1: 0.7443 2022/09/07 12:53:08 - mmengine - INFO - Epoch(train) [70][100/3757] lr: 1.0000e-03 eta: 5:07:45 time: 0.1612 data_time: 0.0106 memory: 7627 grad_norm: 6.7762 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9078 loss: 0.9078 2022/09/07 12:53:24 - mmengine - INFO - Epoch(train) [70][200/3757] lr: 1.0000e-03 eta: 5:07:29 time: 0.1587 data_time: 0.0095 memory: 7124 grad_norm: 6.6564 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1807 loss: 1.1807 2022/09/07 12:53:40 - mmengine - INFO - Epoch(train) [70][300/3757] lr: 1.0000e-03 eta: 5:07:13 time: 0.1576 data_time: 0.0102 memory: 7124 grad_norm: 6.4926 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2561 loss: 1.2561 2022/09/07 12:53:57 - mmengine - INFO - Epoch(train) [70][400/3757] lr: 1.0000e-03 eta: 5:06:57 time: 0.1611 data_time: 0.0100 memory: 7124 grad_norm: 6.8561 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2840 loss: 1.2840 2022/09/07 12:54:13 - mmengine - INFO - Epoch(train) [70][500/3757] lr: 1.0000e-03 eta: 5:06:41 time: 0.1619 data_time: 0.0093 memory: 7124 grad_norm: 6.7406 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1955 loss: 1.1955 2022/09/07 12:54:29 - mmengine - INFO - Epoch(train) [70][600/3757] lr: 1.0000e-03 eta: 5:06:26 time: 0.1578 data_time: 0.0103 memory: 7124 grad_norm: 6.4953 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.0177 loss: 1.0177 2022/09/07 12:54:45 - mmengine - INFO - Epoch(train) [70][700/3757] lr: 1.0000e-03 eta: 5:06:10 time: 0.1580 data_time: 0.0107 memory: 7124 grad_norm: 6.5810 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1982 loss: 1.1982 2022/09/07 12:54:56 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:55:01 - mmengine - INFO - Epoch(train) [70][800/3757] lr: 1.0000e-03 eta: 5:05:54 time: 0.1598 data_time: 0.0102 memory: 7124 grad_norm: 6.4613 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2425 loss: 1.2425 2022/09/07 12:55:17 - mmengine - INFO - Epoch(train) [70][900/3757] lr: 1.0000e-03 eta: 5:05:38 time: 0.1575 data_time: 0.0106 memory: 7124 grad_norm: 6.6485 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1235 loss: 1.1235 2022/09/07 12:55:33 - mmengine - INFO - Epoch(train) [70][1000/3757] lr: 1.0000e-03 eta: 5:05:23 time: 0.1621 data_time: 0.0109 memory: 7124 grad_norm: 6.7586 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1291 loss: 1.1291 2022/09/07 12:55:49 - mmengine - INFO - Epoch(train) [70][1100/3757] lr: 1.0000e-03 eta: 5:05:07 time: 0.1614 data_time: 0.0114 memory: 7124 grad_norm: 6.8593 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1831 loss: 1.1831 2022/09/07 12:56:06 - mmengine - INFO - Epoch(train) [70][1200/3757] lr: 1.0000e-03 eta: 5:04:51 time: 0.1594 data_time: 0.0113 memory: 7124 grad_norm: 6.7655 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1779 loss: 1.1779 2022/09/07 12:56:22 - mmengine - INFO - Epoch(train) [70][1300/3757] lr: 1.0000e-03 eta: 5:04:35 time: 0.1614 data_time: 0.0101 memory: 7124 grad_norm: 6.6037 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0233 loss: 1.0233 2022/09/07 12:56:38 - mmengine - INFO - Epoch(train) [70][1400/3757] lr: 1.0000e-03 eta: 5:04:19 time: 0.1590 data_time: 0.0111 memory: 7124 grad_norm: 6.5473 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1183 loss: 1.1183 2022/09/07 12:56:54 - mmengine - INFO - Epoch(train) [70][1500/3757] lr: 1.0000e-03 eta: 5:04:04 time: 0.1642 data_time: 0.0169 memory: 7124 grad_norm: 6.5388 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9670 loss: 0.9670 2022/09/07 12:57:10 - mmengine - INFO - Epoch(train) [70][1600/3757] lr: 1.0000e-03 eta: 5:03:48 time: 0.1603 data_time: 0.0090 memory: 7124 grad_norm: 6.3117 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2917 loss: 1.2917 2022/09/07 12:57:26 - mmengine - INFO - Epoch(train) [70][1700/3757] lr: 1.0000e-03 eta: 5:03:32 time: 0.1579 data_time: 0.0108 memory: 7124 grad_norm: 6.7320 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1109 loss: 1.1109 2022/09/07 12:57:37 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 12:57:43 - mmengine - INFO - Epoch(train) [70][1800/3757] lr: 1.0000e-03 eta: 5:03:16 time: 0.1674 data_time: 0.0104 memory: 7124 grad_norm: 6.8148 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3515 loss: 1.3515 2022/09/07 12:57:59 - mmengine - INFO - Epoch(train) [70][1900/3757] lr: 1.0000e-03 eta: 5:03:00 time: 0.1590 data_time: 0.0091 memory: 7124 grad_norm: 6.7095 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1413 loss: 1.1413 2022/09/07 12:58:15 - mmengine - INFO - Epoch(train) [70][2000/3757] lr: 1.0000e-03 eta: 5:02:45 time: 0.1603 data_time: 0.0121 memory: 7124 grad_norm: 6.3911 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1049 loss: 1.1049 2022/09/07 12:58:31 - mmengine - INFO - Epoch(train) [70][2100/3757] lr: 1.0000e-03 eta: 5:02:29 time: 0.1556 data_time: 0.0097 memory: 7124 grad_norm: 6.8546 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2435 loss: 1.2435 2022/09/07 12:58:47 - mmengine - INFO - Epoch(train) [70][2200/3757] lr: 1.0000e-03 eta: 5:02:13 time: 0.1583 data_time: 0.0109 memory: 7124 grad_norm: 6.7742 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8372 loss: 0.8372 2022/09/07 12:59:03 - mmengine - INFO - Epoch(train) [70][2300/3757] lr: 1.0000e-03 eta: 5:01:57 time: 0.1781 data_time: 0.0094 memory: 7124 grad_norm: 6.8545 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2435 loss: 1.2435 2022/09/07 12:59:20 - mmengine - INFO - Epoch(train) [70][2400/3757] lr: 1.0000e-03 eta: 5:01:42 time: 0.1611 data_time: 0.0108 memory: 7124 grad_norm: 6.5752 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0826 loss: 1.0826 2022/09/07 12:59:36 - mmengine - INFO - Epoch(train) [70][2500/3757] lr: 1.0000e-03 eta: 5:01:26 time: 0.1596 data_time: 0.0106 memory: 7124 grad_norm: 6.6863 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3648 loss: 1.3648 2022/09/07 12:59:52 - mmengine - INFO - Epoch(train) [70][2600/3757] lr: 1.0000e-03 eta: 5:01:10 time: 0.1631 data_time: 0.0122 memory: 7124 grad_norm: 6.8116 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.0842 loss: 1.0842 2022/09/07 13:00:08 - mmengine - INFO - Epoch(train) [70][2700/3757] lr: 1.0000e-03 eta: 5:00:54 time: 0.1594 data_time: 0.0106 memory: 7124 grad_norm: 6.9547 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0099 loss: 1.0099 2022/09/07 13:00:19 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 13:00:24 - mmengine - INFO - Epoch(train) [70][2800/3757] lr: 1.0000e-03 eta: 5:00:38 time: 0.1601 data_time: 0.0090 memory: 7124 grad_norm: 6.4283 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0243 loss: 1.0243 2022/09/07 13:00:40 - mmengine - INFO - Epoch(train) [70][2900/3757] lr: 1.0000e-03 eta: 5:00:23 time: 0.1574 data_time: 0.0097 memory: 7124 grad_norm: 6.4842 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8174 loss: 0.8174 2022/09/07 13:00:56 - mmengine - INFO - Epoch(train) [70][3000/3757] lr: 1.0000e-03 eta: 5:00:07 time: 0.1625 data_time: 0.0096 memory: 7124 grad_norm: 6.7339 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9951 loss: 0.9951 2022/09/07 13:01:12 - mmengine - INFO - Epoch(train) [70][3100/3757] lr: 1.0000e-03 eta: 4:59:51 time: 0.1612 data_time: 0.0120 memory: 7124 grad_norm: 7.1215 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.3032 loss: 1.3032 2022/09/07 13:01:28 - mmengine - INFO - Epoch(train) [70][3200/3757] lr: 1.0000e-03 eta: 4:59:35 time: 0.1589 data_time: 0.0111 memory: 7124 grad_norm: 7.0182 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2719 loss: 1.2719 2022/09/07 13:01:45 - mmengine - INFO - Epoch(train) [70][3300/3757] lr: 1.0000e-03 eta: 4:59:19 time: 0.1580 data_time: 0.0096 memory: 7124 grad_norm: 6.7426 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2527 loss: 1.2527 2022/09/07 13:02:01 - mmengine - INFO - Epoch(train) [70][3400/3757] lr: 1.0000e-03 eta: 4:59:03 time: 0.1553 data_time: 0.0101 memory: 7124 grad_norm: 6.7029 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1731 loss: 1.1731 2022/09/07 13:02:17 - mmengine - INFO - Epoch(train) [70][3500/3757] lr: 1.0000e-03 eta: 4:58:48 time: 0.1590 data_time: 0.0111 memory: 7124 grad_norm: 6.7999 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9384 loss: 0.9384 2022/09/07 13:02:33 - mmengine - INFO - Epoch(train) [70][3600/3757] lr: 1.0000e-03 eta: 4:58:32 time: 0.1576 data_time: 0.0111 memory: 7124 grad_norm: 6.7518 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1525 loss: 1.1525 2022/09/07 13:02:49 - mmengine - INFO - Epoch(train) [70][3700/3757] lr: 1.0000e-03 eta: 4:58:16 time: 0.1590 data_time: 0.0111 memory: 7124 grad_norm: 6.8143 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8943 loss: 0.8943 2022/09/07 13:02:58 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 13:02:58 - mmengine - INFO - Epoch(train) [70][3757/3757] lr: 1.0000e-03 eta: 4:58:10 time: 0.1376 data_time: 0.0072 memory: 7124 grad_norm: 6.6791 top1_acc: 0.8571 top5_acc: 1.0000 loss_cls: 1.1978 loss: 1.1978 2022/09/07 13:02:58 - mmengine - INFO - Saving checkpoint at 70 epochs 2022/09/07 13:05:17 - mmengine - INFO - Epoch(val) [70][100/310] eta: 0:04:10 time: 1.1935 data_time: 0.8915 memory: 7627 2022/09/07 13:07:32 - mmengine - INFO - Epoch(val) [70][200/310] eta: 0:02:15 time: 1.2333 data_time: 0.9332 memory: 7627 2022/09/07 13:09:37 - mmengine - INFO - Epoch(val) [70][300/310] eta: 0:00:12 time: 1.2119 data_time: 0.9139 memory: 7627 2022/09/07 13:09:56 - mmengine - INFO - Epoch(val) [70][310/310] acc/top1: 0.7448 acc/top5: 0.9147 acc/mean1: 0.7447 2022/09/07 13:09:59 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 13:10:14 - mmengine - INFO - Epoch(train) [71][100/3757] lr: 1.0000e-03 eta: 4:57:51 time: 0.1617 data_time: 0.0111 memory: 7627 grad_norm: 6.5749 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0984 loss: 1.0984 2022/09/07 13:10:31 - mmengine - INFO - Epoch(train) [71][200/3757] lr: 1.0000e-03 eta: 4:57:35 time: 0.1625 data_time: 0.0098 memory: 7124 grad_norm: 6.4579 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0954 loss: 1.0954 2022/09/07 13:10:49 - mmengine - INFO - Epoch(train) [71][300/3757] lr: 1.0000e-03 eta: 4:57:21 time: 0.2050 data_time: 0.0144 memory: 7124 grad_norm: 6.6390 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2628 loss: 1.2628 2022/09/07 13:11:10 - mmengine - INFO - Epoch(train) [71][400/3757] lr: 1.0000e-03 eta: 4:57:07 time: 0.2053 data_time: 0.0157 memory: 7124 grad_norm: 6.4486 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1188 loss: 1.1188 2022/09/07 13:11:31 - mmengine - INFO - Epoch(train) [71][500/3757] lr: 1.0000e-03 eta: 4:56:53 time: 0.2012 data_time: 0.0149 memory: 7124 grad_norm: 6.5008 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0607 loss: 1.0607 2022/09/07 13:11:52 - mmengine - INFO - Epoch(train) [71][600/3757] lr: 1.0000e-03 eta: 4:56:39 time: 0.2116 data_time: 0.0156 memory: 7124 grad_norm: 6.7838 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0850 loss: 1.0850 2022/09/07 13:12:12 - mmengine - INFO - Epoch(train) [71][700/3757] lr: 1.0000e-03 eta: 4:56:25 time: 0.2111 data_time: 0.0141 memory: 7124 grad_norm: 6.7022 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1181 loss: 1.1181 2022/09/07 13:12:34 - mmengine - INFO - Epoch(train) [71][800/3757] lr: 1.0000e-03 eta: 4:56:12 time: 0.2128 data_time: 0.0210 memory: 7124 grad_norm: 6.8739 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2324 loss: 1.2324 2022/09/07 13:12:55 - mmengine - INFO - Epoch(train) [71][900/3757] lr: 1.0000e-03 eta: 4:55:58 time: 0.2164 data_time: 0.0147 memory: 7124 grad_norm: 6.4700 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0749 loss: 1.0749 2022/09/07 13:13:16 - mmengine - INFO - Epoch(train) [71][1000/3757] lr: 1.0000e-03 eta: 4:55:44 time: 0.2201 data_time: 0.0129 memory: 7124 grad_norm: 6.7524 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2126 loss: 1.2126 2022/09/07 13:13:19 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 13:13:38 - mmengine - INFO - Epoch(train) [71][1100/3757] lr: 1.0000e-03 eta: 4:55:31 time: 0.2111 data_time: 0.0151 memory: 7124 grad_norm: 6.5337 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0707 loss: 1.0707 2022/09/07 13:13:59 - mmengine - INFO - Epoch(train) [71][1200/3757] lr: 1.0000e-03 eta: 4:55:17 time: 0.2156 data_time: 0.0136 memory: 7124 grad_norm: 6.3757 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1460 loss: 1.1460 2022/09/07 13:14:21 - mmengine - INFO - Epoch(train) [71][1300/3757] lr: 1.0000e-03 eta: 4:55:04 time: 0.2172 data_time: 0.0147 memory: 7124 grad_norm: 6.7861 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1284 loss: 1.1284 2022/09/07 13:14:43 - mmengine - INFO - Epoch(train) [71][1400/3757] lr: 1.0000e-03 eta: 4:54:50 time: 0.2191 data_time: 0.0140 memory: 7124 grad_norm: 6.9185 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.2059 loss: 1.2059 2022/09/07 13:15:04 - mmengine - INFO - Epoch(train) [71][1500/3757] lr: 1.0000e-03 eta: 4:54:37 time: 0.2055 data_time: 0.0146 memory: 7124 grad_norm: 6.9662 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.3578 loss: 1.3578 2022/09/07 13:15:26 - mmengine - INFO - Epoch(train) [71][1600/3757] lr: 1.0000e-03 eta: 4:54:23 time: 0.2047 data_time: 0.0141 memory: 7124 grad_norm: 6.5663 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1815 loss: 1.1815 2022/09/07 13:15:47 - mmengine - INFO - Epoch(train) [71][1700/3757] lr: 1.0000e-03 eta: 4:54:09 time: 0.2126 data_time: 0.0148 memory: 7124 grad_norm: 6.8136 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0163 loss: 1.0163 2022/09/07 13:16:08 - mmengine - INFO - Epoch(train) [71][1800/3757] lr: 1.0000e-03 eta: 4:53:55 time: 0.2078 data_time: 0.0162 memory: 7124 grad_norm: 6.5662 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9899 loss: 0.9899 2022/09/07 13:16:30 - mmengine - INFO - Epoch(train) [71][1900/3757] lr: 1.0000e-03 eta: 4:53:42 time: 0.2100 data_time: 0.0154 memory: 7124 grad_norm: 6.7773 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0948 loss: 1.0948 2022/09/07 13:16:51 - mmengine - INFO - Epoch(train) [71][2000/3757] lr: 1.0000e-03 eta: 4:53:28 time: 0.2134 data_time: 0.0167 memory: 7124 grad_norm: 7.0915 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2541 loss: 1.2541 2022/09/07 13:16:54 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 13:17:13 - mmengine - INFO - Epoch(train) [71][2100/3757] lr: 1.0000e-03 eta: 4:53:15 time: 0.2087 data_time: 0.0151 memory: 7124 grad_norm: 6.7647 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2644 loss: 1.2644 2022/09/07 13:17:34 - mmengine - INFO - Epoch(train) [71][2200/3757] lr: 1.0000e-03 eta: 4:53:01 time: 0.1986 data_time: 0.0160 memory: 7124 grad_norm: 6.6878 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9120 loss: 0.9120 2022/09/07 13:17:55 - mmengine - INFO - Epoch(train) [71][2300/3757] lr: 1.0000e-03 eta: 4:52:47 time: 0.2087 data_time: 0.0151 memory: 7124 grad_norm: 7.0668 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1095 loss: 1.1095 2022/09/07 13:18:17 - mmengine - INFO - Epoch(train) [71][2400/3757] lr: 1.0000e-03 eta: 4:52:33 time: 0.2073 data_time: 0.0149 memory: 7124 grad_norm: 6.5576 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9733 loss: 0.9733 2022/09/07 13:18:38 - mmengine - INFO - Epoch(train) [71][2500/3757] lr: 1.0000e-03 eta: 4:52:19 time: 0.2052 data_time: 0.0147 memory: 7124 grad_norm: 7.0268 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1738 loss: 1.1738 2022/09/07 13:18:58 - mmengine - INFO - Epoch(train) [71][2600/3757] lr: 1.0000e-03 eta: 4:52:06 time: 0.2035 data_time: 0.0129 memory: 7124 grad_norm: 7.0131 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.4328 loss: 1.4328 2022/09/07 13:19:20 - mmengine - INFO - Epoch(train) [71][2700/3757] lr: 1.0000e-03 eta: 4:51:52 time: 0.2033 data_time: 0.0134 memory: 7124 grad_norm: 6.7405 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.2364 loss: 1.2364 2022/09/07 13:19:41 - mmengine - INFO - Epoch(train) [71][2800/3757] lr: 1.0000e-03 eta: 4:51:38 time: 0.2062 data_time: 0.0142 memory: 7124 grad_norm: 7.0150 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.2477 loss: 1.2477 2022/09/07 13:20:02 - mmengine - INFO - Epoch(train) [71][2900/3757] lr: 1.0000e-03 eta: 4:51:24 time: 0.2059 data_time: 0.0188 memory: 7124 grad_norm: 6.6352 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1184 loss: 1.1184 2022/09/07 13:20:22 - mmengine - INFO - Epoch(train) [71][3000/3757] lr: 1.0000e-03 eta: 4:51:10 time: 0.2064 data_time: 0.0135 memory: 7124 grad_norm: 7.0406 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.1673 loss: 1.1673 2022/09/07 13:20:24 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 13:20:43 - mmengine - INFO - Epoch(train) [71][3100/3757] lr: 1.0000e-03 eta: 4:50:56 time: 0.2035 data_time: 0.0153 memory: 7124 grad_norm: 6.9219 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1030 loss: 1.1030 2022/09/07 13:21:04 - mmengine - INFO - Epoch(train) [71][3200/3757] lr: 1.0000e-03 eta: 4:50:42 time: 0.2016 data_time: 0.0132 memory: 7124 grad_norm: 6.8737 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2725 loss: 1.2725 2022/09/07 13:21:25 - mmengine - INFO - Epoch(train) [71][3300/3757] lr: 1.0000e-03 eta: 4:50:28 time: 0.2023 data_time: 0.0169 memory: 7124 grad_norm: 6.8811 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1520 loss: 1.1520 2022/09/07 13:21:46 - mmengine - INFO - Epoch(train) [71][3400/3757] lr: 1.0000e-03 eta: 4:50:14 time: 0.2061 data_time: 0.0153 memory: 7124 grad_norm: 6.6279 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0451 loss: 1.0451 2022/09/07 13:22:08 - mmengine - INFO - Epoch(train) [71][3500/3757] lr: 1.0000e-03 eta: 4:50:01 time: 0.2070 data_time: 0.0155 memory: 7124 grad_norm: 6.5520 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1028 loss: 1.1028 2022/09/07 13:22:29 - mmengine - INFO - Epoch(train) [71][3600/3757] lr: 1.0000e-03 eta: 4:49:47 time: 0.2006 data_time: 0.0147 memory: 7124 grad_norm: 6.8220 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.0898 loss: 1.0898 2022/09/07 13:22:49 - mmengine - INFO - Epoch(train) [71][3700/3757] lr: 1.0000e-03 eta: 4:49:33 time: 0.2026 data_time: 0.0134 memory: 7124 grad_norm: 7.0852 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.2681 loss: 1.2681 2022/09/07 13:23:00 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 13:23:00 - mmengine - INFO - Epoch(train) [71][3757/3757] lr: 1.0000e-03 eta: 4:49:27 time: 0.1759 data_time: 0.0110 memory: 7124 grad_norm: 6.7486 top1_acc: 0.8571 top5_acc: 1.0000 loss_cls: 1.1300 loss: 1.1300 2022/09/07 13:23:01 - mmengine - INFO - Saving checkpoint at 71 epochs 2022/09/07 13:26:07 - mmengine - INFO - Epoch(val) [71][100/310] eta: 0:05:45 time: 1.6468 data_time: 1.3206 memory: 7627 2022/09/07 13:29:01 - mmengine - INFO - Epoch(val) [71][200/310] eta: 0:02:49 time: 1.5429 data_time: 1.2324 memory: 7627 2022/09/07 13:31:40 - mmengine - INFO - Epoch(val) [71][300/310] eta: 0:00:13 time: 1.3929 data_time: 1.0798 memory: 7627 2022/09/07 13:31:57 - mmengine - INFO - Epoch(val) [71][310/310] acc/top1: 0.7473 acc/top5: 0.9147 acc/mean1: 0.7472 2022/09/07 13:31:57 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_67.pth is removed 2022/09/07 13:32:00 - mmengine - INFO - The best checkpoint with 0.7473 acc/top1 at 71 epoch is saved to best_acc/top1_epoch_71.pth. 2022/09/07 13:32:21 - mmengine - INFO - Epoch(train) [72][100/3757] lr: 1.0000e-03 eta: 4:49:09 time: 0.1988 data_time: 0.0146 memory: 7627 grad_norm: 7.2395 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2740 loss: 1.2740 2022/09/07 13:32:40 - mmengine - INFO - Epoch(train) [72][200/3757] lr: 1.0000e-03 eta: 4:48:55 time: 0.1944 data_time: 0.0150 memory: 7124 grad_norm: 6.8740 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9942 loss: 0.9942 2022/09/07 13:32:51 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 13:33:00 - mmengine - INFO - Epoch(train) [72][300/3757] lr: 1.0000e-03 eta: 4:48:40 time: 0.1878 data_time: 0.0150 memory: 7124 grad_norm: 6.4203 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3359 loss: 1.3359 2022/09/07 13:33:21 - mmengine - INFO - Epoch(train) [72][400/3757] lr: 1.0000e-03 eta: 4:48:26 time: 0.1957 data_time: 0.0125 memory: 7124 grad_norm: 6.6015 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2248 loss: 1.2248 2022/09/07 13:33:41 - mmengine - INFO - Epoch(train) [72][500/3757] lr: 1.0000e-03 eta: 4:48:12 time: 0.1915 data_time: 0.0146 memory: 7124 grad_norm: 6.6447 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2433 loss: 1.2433 2022/09/07 13:34:00 - mmengine - INFO - Epoch(train) [72][600/3757] lr: 1.0000e-03 eta: 4:47:57 time: 0.1990 data_time: 0.0142 memory: 7124 grad_norm: 6.7365 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2758 loss: 1.2758 2022/09/07 13:34:21 - mmengine - INFO - Epoch(train) [72][700/3757] lr: 1.0000e-03 eta: 4:47:43 time: 0.1927 data_time: 0.0140 memory: 7124 grad_norm: 6.6900 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.5085 loss: 1.5085 2022/09/07 13:34:40 - mmengine - INFO - Epoch(train) [72][800/3757] lr: 1.0000e-03 eta: 4:47:29 time: 0.1878 data_time: 0.0144 memory: 7124 grad_norm: 6.5146 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0380 loss: 1.0380 2022/09/07 13:35:00 - mmengine - INFO - Epoch(train) [72][900/3757] lr: 1.0000e-03 eta: 4:47:14 time: 0.1852 data_time: 0.0145 memory: 7124 grad_norm: 6.8923 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1436 loss: 1.1436 2022/09/07 13:35:20 - mmengine - INFO - Epoch(train) [72][1000/3757] lr: 1.0000e-03 eta: 4:47:00 time: 0.1933 data_time: 0.0162 memory: 7124 grad_norm: 6.6260 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1012 loss: 1.1012 2022/09/07 13:35:39 - mmengine - INFO - Epoch(train) [72][1100/3757] lr: 1.0000e-03 eta: 4:46:45 time: 0.1831 data_time: 0.0139 memory: 7124 grad_norm: 6.9591 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1067 loss: 1.1067 2022/09/07 13:35:58 - mmengine - INFO - Epoch(train) [72][1200/3757] lr: 1.0000e-03 eta: 4:46:30 time: 0.1896 data_time: 0.0173 memory: 7124 grad_norm: 6.7904 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1778 loss: 1.1778 2022/09/07 13:36:08 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 13:36:17 - mmengine - INFO - Epoch(train) [72][1300/3757] lr: 1.0000e-03 eta: 4:46:15 time: 0.1982 data_time: 0.0123 memory: 7124 grad_norm: 6.8890 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3630 loss: 1.3630 2022/09/07 13:36:36 - mmengine - INFO - Epoch(train) [72][1400/3757] lr: 1.0000e-03 eta: 4:46:01 time: 0.1998 data_time: 0.0132 memory: 7124 grad_norm: 6.5944 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0012 loss: 1.0012 2022/09/07 13:36:55 - mmengine - INFO - Epoch(train) [72][1500/3757] lr: 1.0000e-03 eta: 4:45:46 time: 0.1928 data_time: 0.0140 memory: 7124 grad_norm: 6.8411 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.2655 loss: 1.2655 2022/09/07 13:37:14 - mmengine - INFO - Epoch(train) [72][1600/3757] lr: 1.0000e-03 eta: 4:45:31 time: 0.1817 data_time: 0.0116 memory: 7124 grad_norm: 6.9684 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1472 loss: 1.1472 2022/09/07 13:37:32 - mmengine - INFO - Epoch(train) [72][1700/3757] lr: 1.0000e-03 eta: 4:45:16 time: 0.1847 data_time: 0.0139 memory: 7124 grad_norm: 6.9029 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.0996 loss: 1.0996 2022/09/07 13:37:51 - mmengine - INFO - Epoch(train) [72][1800/3757] lr: 1.0000e-03 eta: 4:45:01 time: 0.1878 data_time: 0.0132 memory: 7124 grad_norm: 6.7402 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.1115 loss: 1.1115 2022/09/07 13:38:11 - mmengine - INFO - Epoch(train) [72][1900/3757] lr: 1.0000e-03 eta: 4:44:47 time: 0.1823 data_time: 0.0121 memory: 7124 grad_norm: 6.7723 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1903 loss: 1.1903 2022/09/07 13:38:29 - mmengine - INFO - Epoch(train) [72][2000/3757] lr: 1.0000e-03 eta: 4:44:32 time: 0.1912 data_time: 0.0132 memory: 7124 grad_norm: 6.7939 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0764 loss: 1.0764 2022/09/07 13:38:48 - mmengine - INFO - Epoch(train) [72][2100/3757] lr: 1.0000e-03 eta: 4:44:17 time: 0.1855 data_time: 0.0127 memory: 7124 grad_norm: 6.6960 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.3058 loss: 1.3058 2022/09/07 13:39:07 - mmengine - INFO - Epoch(train) [72][2200/3757] lr: 1.0000e-03 eta: 4:44:02 time: 0.1802 data_time: 0.0117 memory: 7124 grad_norm: 6.8551 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2035 loss: 1.2035 2022/09/07 13:39:17 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 13:39:26 - mmengine - INFO - Epoch(train) [72][2300/3757] lr: 1.0000e-03 eta: 4:43:47 time: 0.2215 data_time: 0.0132 memory: 7124 grad_norm: 6.6322 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0632 loss: 1.0632 2022/09/07 13:39:44 - mmengine - INFO - Epoch(train) [72][2400/3757] lr: 1.0000e-03 eta: 4:43:32 time: 0.1855 data_time: 0.0120 memory: 7124 grad_norm: 6.8818 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0852 loss: 1.0852 2022/09/07 13:40:03 - mmengine - INFO - Epoch(train) [72][2500/3757] lr: 1.0000e-03 eta: 4:43:17 time: 0.1921 data_time: 0.0130 memory: 7124 grad_norm: 6.7926 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1814 loss: 1.1814 2022/09/07 13:40:22 - mmengine - INFO - Epoch(train) [72][2600/3757] lr: 1.0000e-03 eta: 4:43:02 time: 0.2266 data_time: 0.0118 memory: 7124 grad_norm: 6.8457 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.9041 loss: 0.9041 2022/09/07 13:40:41 - mmengine - INFO - Epoch(train) [72][2700/3757] lr: 1.0000e-03 eta: 4:42:48 time: 0.1774 data_time: 0.0108 memory: 7124 grad_norm: 6.7078 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0693 loss: 1.0693 2022/09/07 13:41:00 - mmengine - INFO - Epoch(train) [72][2800/3757] lr: 1.0000e-03 eta: 4:42:33 time: 0.1835 data_time: 0.0131 memory: 7124 grad_norm: 6.5200 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9158 loss: 0.9158 2022/09/07 13:41:19 - mmengine - INFO - Epoch(train) [72][2900/3757] lr: 1.0000e-03 eta: 4:42:18 time: 0.1755 data_time: 0.0119 memory: 7124 grad_norm: 7.1062 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1481 loss: 1.1481 2022/09/07 13:41:39 - mmengine - INFO - Epoch(train) [72][3000/3757] lr: 1.0000e-03 eta: 4:42:03 time: 0.1880 data_time: 0.0133 memory: 7124 grad_norm: 6.8784 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9329 loss: 0.9329 2022/09/07 13:41:58 - mmengine - INFO - Epoch(train) [72][3100/3757] lr: 1.0000e-03 eta: 4:41:49 time: 0.1834 data_time: 0.0120 memory: 7124 grad_norm: 6.8597 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9783 loss: 0.9783 2022/09/07 13:42:17 - mmengine - INFO - Epoch(train) [72][3200/3757] lr: 1.0000e-03 eta: 4:41:34 time: 0.2057 data_time: 0.0111 memory: 7124 grad_norm: 6.9067 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.2794 loss: 1.2794 2022/09/07 13:42:27 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 13:42:36 - mmengine - INFO - Epoch(train) [72][3300/3757] lr: 1.0000e-03 eta: 4:41:19 time: 0.1864 data_time: 0.0132 memory: 7124 grad_norm: 7.1897 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2662 loss: 1.2662 2022/09/07 13:42:55 - mmengine - INFO - Epoch(train) [72][3400/3757] lr: 1.0000e-03 eta: 4:41:04 time: 0.1886 data_time: 0.0125 memory: 7124 grad_norm: 7.1020 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0565 loss: 1.0565 2022/09/07 13:43:13 - mmengine - INFO - Epoch(train) [72][3500/3757] lr: 1.0000e-03 eta: 4:40:49 time: 0.1859 data_time: 0.0128 memory: 7124 grad_norm: 6.8837 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2203 loss: 1.2203 2022/09/07 13:43:33 - mmengine - INFO - Epoch(train) [72][3600/3757] lr: 1.0000e-03 eta: 4:40:34 time: 0.1801 data_time: 0.0109 memory: 7124 grad_norm: 6.7908 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1867 loss: 1.1867 2022/09/07 13:43:52 - mmengine - INFO - Epoch(train) [72][3700/3757] lr: 1.0000e-03 eta: 4:40:19 time: 0.1771 data_time: 0.0111 memory: 7124 grad_norm: 6.8663 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9152 loss: 0.9152 2022/09/07 13:44:01 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 13:44:01 - mmengine - INFO - Epoch(train) [72][3757/3757] lr: 1.0000e-03 eta: 4:40:13 time: 0.1498 data_time: 0.0086 memory: 7124 grad_norm: 6.7785 top1_acc: 0.5714 top5_acc: 1.0000 loss_cls: 1.1655 loss: 1.1655 2022/09/07 13:44:01 - mmengine - INFO - Saving checkpoint at 72 epochs 2022/09/07 13:46:42 - mmengine - INFO - Epoch(val) [72][100/310] eta: 0:04:43 time: 1.3516 data_time: 1.0414 memory: 7627 2022/09/07 13:49:23 - mmengine - INFO - Epoch(val) [72][200/310] eta: 0:02:57 time: 1.6098 data_time: 1.2981 memory: 7627 2022/09/07 13:51:48 - mmengine - INFO - Epoch(val) [72][300/310] eta: 0:00:13 time: 1.3034 data_time: 0.9987 memory: 7627 2022/09/07 13:52:07 - mmengine - INFO - Epoch(val) [72][310/310] acc/top1: 0.7454 acc/top5: 0.9170 acc/mean1: 0.7452 2022/09/07 13:52:28 - mmengine - INFO - Epoch(train) [73][100/3757] lr: 1.0000e-03 eta: 4:39:55 time: 0.2006 data_time: 0.0146 memory: 7627 grad_norm: 6.9445 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.1174 loss: 1.1174 2022/09/07 13:52:47 - mmengine - INFO - Epoch(train) [73][200/3757] lr: 1.0000e-03 eta: 4:39:40 time: 0.2018 data_time: 0.0270 memory: 7124 grad_norm: 6.9364 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9315 loss: 0.9315 2022/09/07 13:53:06 - mmengine - INFO - Epoch(train) [73][300/3757] lr: 1.0000e-03 eta: 4:39:25 time: 0.1952 data_time: 0.0140 memory: 7124 grad_norm: 6.8511 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1537 loss: 1.1537 2022/09/07 13:53:25 - mmengine - INFO - Epoch(train) [73][400/3757] lr: 1.0000e-03 eta: 4:39:10 time: 0.1817 data_time: 0.0135 memory: 7124 grad_norm: 6.9643 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2777 loss: 1.2777 2022/09/07 13:53:42 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 13:53:43 - mmengine - INFO - Epoch(train) [73][500/3757] lr: 1.0000e-03 eta: 4:38:55 time: 0.1790 data_time: 0.0139 memory: 7124 grad_norm: 7.0453 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0813 loss: 1.0813 2022/09/07 13:54:02 - mmengine - INFO - Epoch(train) [73][600/3757] lr: 1.0000e-03 eta: 4:38:41 time: 0.1822 data_time: 0.0154 memory: 7124 grad_norm: 6.8268 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0890 loss: 1.0890 2022/09/07 13:54:21 - mmengine - INFO - Epoch(train) [73][700/3757] lr: 1.0000e-03 eta: 4:38:26 time: 0.1844 data_time: 0.0141 memory: 7124 grad_norm: 7.0042 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0450 loss: 1.0450 2022/09/07 13:54:40 - mmengine - INFO - Epoch(train) [73][800/3757] lr: 1.0000e-03 eta: 4:38:11 time: 0.1866 data_time: 0.0143 memory: 7124 grad_norm: 7.0071 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1572 loss: 1.1572 2022/09/07 13:54:58 - mmengine - INFO - Epoch(train) [73][900/3757] lr: 1.0000e-03 eta: 4:37:55 time: 0.1708 data_time: 0.0104 memory: 7124 grad_norm: 7.1713 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1321 loss: 1.1321 2022/09/07 13:55:16 - mmengine - INFO - Epoch(train) [73][1000/3757] lr: 1.0000e-03 eta: 4:37:40 time: 0.1910 data_time: 0.0114 memory: 7124 grad_norm: 7.1183 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9679 loss: 0.9679 2022/09/07 13:55:34 - mmengine - INFO - Epoch(train) [73][1100/3757] lr: 1.0000e-03 eta: 4:37:25 time: 0.1789 data_time: 0.0129 memory: 7124 grad_norm: 6.8983 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1211 loss: 1.1211 2022/09/07 13:55:53 - mmengine - INFO - Epoch(train) [73][1200/3757] lr: 1.0000e-03 eta: 4:37:10 time: 0.1740 data_time: 0.0123 memory: 7124 grad_norm: 6.9856 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1282 loss: 1.1282 2022/09/07 13:56:11 - mmengine - INFO - Epoch(train) [73][1300/3757] lr: 1.0000e-03 eta: 4:36:55 time: 0.1769 data_time: 0.0113 memory: 7124 grad_norm: 6.6267 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0803 loss: 1.0803 2022/09/07 13:56:29 - mmengine - INFO - Epoch(train) [73][1400/3757] lr: 1.0000e-03 eta: 4:36:40 time: 0.1942 data_time: 0.0107 memory: 7124 grad_norm: 6.6827 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.0097 loss: 1.0097 2022/09/07 13:56:47 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 13:56:48 - mmengine - INFO - Epoch(train) [73][1500/3757] lr: 1.0000e-03 eta: 4:36:25 time: 0.2045 data_time: 0.0156 memory: 7124 grad_norm: 6.8418 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9950 loss: 0.9950 2022/09/07 13:57:06 - mmengine - INFO - Epoch(train) [73][1600/3757] lr: 1.0000e-03 eta: 4:36:09 time: 0.1733 data_time: 0.0127 memory: 7124 grad_norm: 6.9436 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.3666 loss: 1.3666 2022/09/07 13:57:24 - mmengine - INFO - Epoch(train) [73][1700/3757] lr: 1.0000e-03 eta: 4:35:54 time: 0.1791 data_time: 0.0140 memory: 7124 grad_norm: 6.8419 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9316 loss: 0.9316 2022/09/07 13:57:43 - mmengine - INFO - Epoch(train) [73][1800/3757] lr: 1.0000e-03 eta: 4:35:39 time: 0.1828 data_time: 0.0119 memory: 7124 grad_norm: 6.9744 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9235 loss: 0.9235 2022/09/07 13:58:01 - mmengine - INFO - Epoch(train) [73][1900/3757] lr: 1.0000e-03 eta: 4:35:24 time: 0.1736 data_time: 0.0109 memory: 7124 grad_norm: 6.9287 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0816 loss: 1.0816 2022/09/07 13:58:19 - mmengine - INFO - Epoch(train) [73][2000/3757] lr: 1.0000e-03 eta: 4:35:09 time: 0.1809 data_time: 0.0112 memory: 7124 grad_norm: 6.7373 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2473 loss: 1.2473 2022/09/07 13:58:37 - mmengine - INFO - Epoch(train) [73][2100/3757] lr: 1.0000e-03 eta: 4:34:53 time: 0.1789 data_time: 0.0121 memory: 7124 grad_norm: 7.1179 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1170 loss: 1.1170 2022/09/07 13:58:55 - mmengine - INFO - Epoch(train) [73][2200/3757] lr: 1.0000e-03 eta: 4:34:38 time: 0.1777 data_time: 0.0138 memory: 7124 grad_norm: 7.0941 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0114 loss: 1.0114 2022/09/07 13:59:13 - mmengine - INFO - Epoch(train) [73][2300/3757] lr: 1.0000e-03 eta: 4:34:23 time: 0.1724 data_time: 0.0108 memory: 7124 grad_norm: 6.8116 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.2746 loss: 1.2746 2022/09/07 13:59:30 - mmengine - INFO - Epoch(train) [73][2400/3757] lr: 1.0000e-03 eta: 4:34:07 time: 0.1787 data_time: 0.0120 memory: 7124 grad_norm: 6.5129 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9208 loss: 0.9208 2022/09/07 13:59:48 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 13:59:49 - mmengine - INFO - Epoch(train) [73][2500/3757] lr: 1.0000e-03 eta: 4:33:52 time: 0.1776 data_time: 0.0110 memory: 7124 grad_norm: 7.2218 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.2171 loss: 1.2171 2022/09/07 14:00:06 - mmengine - INFO - Epoch(train) [73][2600/3757] lr: 1.0000e-03 eta: 4:33:37 time: 0.1737 data_time: 0.0113 memory: 7124 grad_norm: 6.7823 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.9426 loss: 0.9426 2022/09/07 14:00:25 - mmengine - INFO - Epoch(train) [73][2700/3757] lr: 1.0000e-03 eta: 4:33:21 time: 0.1758 data_time: 0.0118 memory: 7124 grad_norm: 6.8489 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0527 loss: 1.0527 2022/09/07 14:00:42 - mmengine - INFO - Epoch(train) [73][2800/3757] lr: 1.0000e-03 eta: 4:33:06 time: 0.1709 data_time: 0.0105 memory: 7124 grad_norm: 6.5418 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1105 loss: 1.1105 2022/09/07 14:00:59 - mmengine - INFO - Epoch(train) [73][2900/3757] lr: 1.0000e-03 eta: 4:32:50 time: 0.1681 data_time: 0.0115 memory: 7124 grad_norm: 6.8034 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1652 loss: 1.1652 2022/09/07 14:01:16 - mmengine - INFO - Epoch(train) [73][3000/3757] lr: 1.0000e-03 eta: 4:32:35 time: 0.1683 data_time: 0.0100 memory: 7124 grad_norm: 6.9742 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1929 loss: 1.1929 2022/09/07 14:01:34 - mmengine - INFO - Epoch(train) [73][3100/3757] lr: 1.0000e-03 eta: 4:32:19 time: 0.1725 data_time: 0.0121 memory: 7124 grad_norm: 6.8879 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1721 loss: 1.1721 2022/09/07 14:01:52 - mmengine - INFO - Epoch(train) [73][3200/3757] lr: 1.0000e-03 eta: 4:32:04 time: 0.1687 data_time: 0.0121 memory: 7124 grad_norm: 7.0632 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9287 loss: 0.9287 2022/09/07 14:02:09 - mmengine - INFO - Epoch(train) [73][3300/3757] lr: 1.0000e-03 eta: 4:31:48 time: 0.1784 data_time: 0.0131 memory: 7124 grad_norm: 6.8881 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3136 loss: 1.3136 2022/09/07 14:02:27 - mmengine - INFO - Epoch(train) [73][3400/3757] lr: 1.0000e-03 eta: 4:31:33 time: 0.1760 data_time: 0.0111 memory: 7124 grad_norm: 7.0295 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1659 loss: 1.1659 2022/09/07 14:02:44 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 14:02:45 - mmengine - INFO - Epoch(train) [73][3500/3757] lr: 1.0000e-03 eta: 4:31:18 time: 0.1775 data_time: 0.0117 memory: 7124 grad_norm: 6.7942 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1886 loss: 1.1886 2022/09/07 14:03:03 - mmengine - INFO - Epoch(train) [73][3600/3757] lr: 1.0000e-03 eta: 4:31:02 time: 0.1682 data_time: 0.0104 memory: 7124 grad_norm: 6.7778 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1117 loss: 1.1117 2022/09/07 14:03:20 - mmengine - INFO - Epoch(train) [73][3700/3757] lr: 1.0000e-03 eta: 4:30:47 time: 0.1685 data_time: 0.0106 memory: 7124 grad_norm: 6.9799 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.2153 loss: 1.2153 2022/09/07 14:03:29 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 14:03:29 - mmengine - INFO - Epoch(train) [73][3757/3757] lr: 1.0000e-03 eta: 4:30:40 time: 0.1488 data_time: 0.0093 memory: 7124 grad_norm: 6.9596 top1_acc: 0.7143 top5_acc: 0.8571 loss_cls: 1.0609 loss: 1.0609 2022/09/07 14:03:29 - mmengine - INFO - Saving checkpoint at 73 epochs 2022/09/07 14:06:02 - mmengine - INFO - Epoch(val) [73][100/310] eta: 0:04:25 time: 1.2620 data_time: 0.9471 memory: 7627 2022/09/07 14:08:26 - mmengine - INFO - Epoch(val) [73][200/310] eta: 0:02:14 time: 1.2257 data_time: 0.9210 memory: 7627 2022/09/07 14:16:40 - mmengine - INFO - Epoch(val) [73][300/310] eta: 0:03:10 time: 19.0088 data_time: 17.9394 memory: 7627 2022/09/07 14:21:47 - mmengine - INFO - Epoch(val) [73][310/310] acc/top1: 0.7413 acc/top5: 0.9135 acc/mean1: 0.7412 2022/09/07 14:25:44 - mmengine - INFO - Epoch(train) [74][100/3757] lr: 1.0000e-03 eta: 4:31:42 time: 2.3333 data_time: 0.1231 memory: 7627 grad_norm: 6.7932 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.7928 loss: 0.7928 2022/09/07 14:28:47 - mmengine - INFO - Epoch(train) [74][200/3757] lr: 1.0000e-03 eta: 4:32:27 time: 1.8239 data_time: 0.0955 memory: 7124 grad_norm: 6.9166 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0336 loss: 1.0336 2022/09/07 14:31:32 - mmengine - INFO - Epoch(train) [74][300/3757] lr: 1.0000e-03 eta: 4:33:05 time: 1.4246 data_time: 0.0989 memory: 7124 grad_norm: 6.7708 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9722 loss: 0.9722 2022/09/07 14:33:57 - mmengine - INFO - Epoch(train) [74][400/3757] lr: 1.0000e-03 eta: 4:33:36 time: 1.1555 data_time: 0.0761 memory: 7124 grad_norm: 6.6810 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2457 loss: 1.2457 2022/09/07 14:35:58 - mmengine - INFO - Epoch(train) [74][500/3757] lr: 1.0000e-03 eta: 4:33:58 time: 0.9433 data_time: 0.0574 memory: 7124 grad_norm: 6.9748 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0991 loss: 1.0991 2022/09/07 14:37:33 - mmengine - INFO - Epoch(train) [74][600/3757] lr: 1.0000e-03 eta: 4:34:10 time: 0.9800 data_time: 0.0592 memory: 7124 grad_norm: 6.9410 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1159 loss: 1.1159 2022/09/07 14:38:57 - mmengine - INFO - Epoch(train) [74][700/3757] lr: 1.0000e-03 eta: 4:34:18 time: 0.8491 data_time: 0.0531 memory: 7124 grad_norm: 6.8218 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1610 loss: 1.1610 2022/09/07 14:39:21 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 14:39:57 - mmengine - INFO - Epoch(train) [74][800/3757] lr: 1.0000e-03 eta: 4:34:18 time: 0.5349 data_time: 0.0454 memory: 7124 grad_norm: 7.0139 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1941 loss: 1.1941 2022/09/07 14:41:00 - mmengine - INFO - Epoch(train) [74][900/3757] lr: 1.0000e-03 eta: 4:34:18 time: 0.5907 data_time: 0.0387 memory: 7124 grad_norm: 6.7037 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9373 loss: 0.9373 2022/09/07 14:42:01 - mmengine - INFO - Epoch(train) [74][1000/3757] lr: 1.0000e-03 eta: 4:34:18 time: 0.5859 data_time: 0.0360 memory: 7124 grad_norm: 7.1007 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1352 loss: 1.1352 2022/09/07 14:42:54 - mmengine - INFO - Epoch(train) [74][1100/3757] lr: 1.0000e-03 eta: 4:34:15 time: 0.5878 data_time: 0.0322 memory: 7124 grad_norm: 6.4353 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0502 loss: 1.0502 2022/09/07 14:43:50 - mmengine - INFO - Epoch(train) [74][1200/3757] lr: 1.0000e-03 eta: 4:34:13 time: 0.5664 data_time: 0.0369 memory: 7124 grad_norm: 6.8432 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1320 loss: 1.1320 2022/09/07 14:44:41 - mmengine - INFO - Epoch(train) [74][1300/3757] lr: 1.0000e-03 eta: 4:34:09 time: 0.3859 data_time: 0.0307 memory: 7124 grad_norm: 7.0600 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1463 loss: 1.1463 2022/09/07 14:45:21 - mmengine - INFO - Epoch(train) [74][1400/3757] lr: 1.0000e-03 eta: 4:34:01 time: 0.3862 data_time: 0.0283 memory: 7124 grad_norm: 7.1250 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2483 loss: 1.2483 2022/09/07 14:46:01 - mmengine - INFO - Epoch(train) [74][1500/3757] lr: 1.0000e-03 eta: 4:33:53 time: 0.4416 data_time: 0.0265 memory: 7124 grad_norm: 6.8844 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2347 loss: 1.2347 2022/09/07 14:46:41 - mmengine - INFO - Epoch(train) [74][1600/3757] lr: 1.0000e-03 eta: 4:33:45 time: 0.4553 data_time: 0.0283 memory: 7124 grad_norm: 7.2788 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.2538 loss: 1.2538 2022/09/07 14:47:23 - mmengine - INFO - Epoch(train) [74][1700/3757] lr: 1.0000e-03 eta: 4:33:38 time: 0.4421 data_time: 0.0292 memory: 7124 grad_norm: 7.1352 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1009 loss: 1.1009 2022/09/07 14:47:40 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 14:48:06 - mmengine - INFO - Epoch(train) [74][1800/3757] lr: 1.0000e-03 eta: 4:33:31 time: 0.4367 data_time: 0.0270 memory: 7124 grad_norm: 7.0245 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0628 loss: 1.0628 2022/09/07 14:48:47 - mmengine - INFO - Epoch(train) [74][1900/3757] lr: 1.0000e-03 eta: 4:33:23 time: 0.4288 data_time: 0.0255 memory: 7124 grad_norm: 7.1499 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1971 loss: 1.1971 2022/09/07 14:49:29 - mmengine - INFO - Epoch(train) [74][2000/3757] lr: 1.0000e-03 eta: 4:33:15 time: 0.4124 data_time: 0.0322 memory: 7124 grad_norm: 6.6492 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1394 loss: 1.1394 2022/09/07 14:50:11 - mmengine - INFO - Epoch(train) [74][2100/3757] lr: 1.0000e-03 eta: 4:33:08 time: 0.4043 data_time: 0.0229 memory: 7124 grad_norm: 6.8382 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1163 loss: 1.1163 2022/09/07 14:50:54 - mmengine - INFO - Epoch(train) [74][2200/3757] lr: 1.0000e-03 eta: 4:33:01 time: 0.4252 data_time: 0.0234 memory: 7124 grad_norm: 6.8559 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9982 loss: 0.9982 2022/09/07 14:51:38 - mmengine - INFO - Epoch(train) [74][2300/3757] lr: 1.0000e-03 eta: 4:32:54 time: 0.5035 data_time: 0.0220 memory: 7124 grad_norm: 6.7141 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1274 loss: 1.1274 2022/09/07 14:52:23 - mmengine - INFO - Epoch(train) [74][2400/3757] lr: 1.0000e-03 eta: 4:32:48 time: 0.4363 data_time: 0.0247 memory: 7124 grad_norm: 6.7966 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1286 loss: 1.1286 2022/09/07 14:53:06 - mmengine - INFO - Epoch(train) [74][2500/3757] lr: 1.0000e-03 eta: 4:32:41 time: 0.4417 data_time: 0.0270 memory: 7124 grad_norm: 7.0696 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9716 loss: 0.9716 2022/09/07 14:53:49 - mmengine - INFO - Epoch(train) [74][2600/3757] lr: 1.0000e-03 eta: 4:32:34 time: 0.4267 data_time: 0.0251 memory: 7124 grad_norm: 6.9474 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1761 loss: 1.1761 2022/09/07 14:54:26 - mmengine - INFO - Epoch(train) [74][2700/3757] lr: 1.0000e-03 eta: 4:32:24 time: 0.3712 data_time: 0.0154 memory: 7124 grad_norm: 7.0398 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1743 loss: 1.1743 2022/09/07 14:54:40 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 14:54:58 - mmengine - INFO - Epoch(train) [74][2800/3757] lr: 1.0000e-03 eta: 4:32:13 time: 0.2661 data_time: 0.0146 memory: 7124 grad_norm: 7.0795 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0177 loss: 1.0177 2022/09/07 14:55:23 - mmengine - INFO - Epoch(train) [74][2900/3757] lr: 1.0000e-03 eta: 4:32:00 time: 0.2506 data_time: 0.0186 memory: 7124 grad_norm: 7.0116 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1076 loss: 1.1076 2022/09/07 14:55:46 - mmengine - INFO - Epoch(train) [74][3000/3757] lr: 1.0000e-03 eta: 4:31:45 time: 0.1930 data_time: 0.0156 memory: 7124 grad_norm: 7.1008 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.2757 loss: 1.2757 2022/09/07 14:56:10 - mmengine - INFO - Epoch(train) [74][3100/3757] lr: 1.0000e-03 eta: 4:31:31 time: 0.2166 data_time: 0.0140 memory: 7124 grad_norm: 7.1339 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1332 loss: 1.1332 2022/09/07 14:56:33 - mmengine - INFO - Epoch(train) [74][3200/3757] lr: 1.0000e-03 eta: 4:31:17 time: 0.2665 data_time: 0.0132 memory: 7124 grad_norm: 6.8599 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2680 loss: 1.2680 2022/09/07 14:56:57 - mmengine - INFO - Epoch(train) [74][3300/3757] lr: 1.0000e-03 eta: 4:31:03 time: 0.2511 data_time: 0.0167 memory: 7124 grad_norm: 7.0124 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.0857 loss: 1.0857 2022/09/07 14:57:21 - mmengine - INFO - Epoch(train) [74][3400/3757] lr: 1.0000e-03 eta: 4:30:49 time: 0.2558 data_time: 0.0132 memory: 7124 grad_norm: 6.8450 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9774 loss: 0.9774 2022/09/07 14:57:45 - mmengine - INFO - Epoch(train) [74][3500/3757] lr: 1.0000e-03 eta: 4:30:35 time: 0.2104 data_time: 0.0130 memory: 7124 grad_norm: 6.8405 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9939 loss: 0.9939 2022/09/07 14:58:08 - mmengine - INFO - Epoch(train) [74][3600/3757] lr: 1.0000e-03 eta: 4:30:20 time: 0.2110 data_time: 0.0129 memory: 7124 grad_norm: 6.8727 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0216 loss: 1.0216 2022/09/07 14:58:31 - mmengine - INFO - Epoch(train) [74][3700/3757] lr: 1.0000e-03 eta: 4:30:06 time: 0.2128 data_time: 0.0225 memory: 7124 grad_norm: 6.8237 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2000 loss: 1.2000 2022/09/07 14:58:40 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 14:58:44 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 14:58:44 - mmengine - INFO - Epoch(train) [74][3757/3757] lr: 1.0000e-03 eta: 4:30:00 time: 0.1976 data_time: 0.0122 memory: 7124 grad_norm: 6.9668 top1_acc: 0.8571 top5_acc: 1.0000 loss_cls: 0.9001 loss: 0.9001 2022/09/07 14:58:44 - mmengine - INFO - Saving checkpoint at 74 epochs 2022/09/07 15:01:09 - mmengine - INFO - Epoch(val) [74][100/310] eta: 0:04:18 time: 1.2333 data_time: 0.9287 memory: 7627 2022/09/07 15:03:29 - mmengine - INFO - Epoch(val) [74][200/310] eta: 0:02:22 time: 1.2940 data_time: 0.9878 memory: 7627 2022/09/07 15:05:50 - mmengine - INFO - Epoch(val) [74][300/310] eta: 0:00:17 time: 1.7930 data_time: 1.4806 memory: 7627 2022/09/07 15:06:31 - mmengine - INFO - Epoch(val) [74][310/310] acc/top1: 0.7457 acc/top5: 0.9152 acc/mean1: 0.7456 2022/09/07 15:07:05 - mmengine - INFO - Epoch(train) [75][100/3757] lr: 1.0000e-03 eta: 4:29:46 time: 0.3687 data_time: 0.0130 memory: 7627 grad_norm: 6.9842 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 0.9565 loss: 0.9565 2022/09/07 15:07:35 - mmengine - INFO - Epoch(train) [75][200/3757] lr: 1.0000e-03 eta: 4:29:34 time: 0.3247 data_time: 0.0173 memory: 7124 grad_norm: 6.9896 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.3527 loss: 1.3527 2022/09/07 15:08:05 - mmengine - INFO - Epoch(train) [75][300/3757] lr: 1.0000e-03 eta: 4:29:22 time: 0.3000 data_time: 0.0152 memory: 7124 grad_norm: 6.5501 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0386 loss: 1.0386 2022/09/07 15:08:35 - mmengine - INFO - Epoch(train) [75][400/3757] lr: 1.0000e-03 eta: 4:29:11 time: 0.3072 data_time: 0.0305 memory: 7124 grad_norm: 6.5501 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0879 loss: 1.0879 2022/09/07 15:09:07 - mmengine - INFO - Epoch(train) [75][500/3757] lr: 1.0000e-03 eta: 4:28:59 time: 0.2898 data_time: 0.0205 memory: 7124 grad_norm: 7.0112 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0569 loss: 1.0569 2022/09/07 15:09:37 - mmengine - INFO - Epoch(train) [75][600/3757] lr: 1.0000e-03 eta: 4:28:47 time: 0.3307 data_time: 0.0614 memory: 7124 grad_norm: 6.7559 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0796 loss: 1.0796 2022/09/07 15:10:06 - mmengine - INFO - Epoch(train) [75][700/3757] lr: 1.0000e-03 eta: 4:28:35 time: 0.3007 data_time: 0.0258 memory: 7124 grad_norm: 6.9651 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0457 loss: 1.0457 2022/09/07 15:10:37 - mmengine - INFO - Epoch(train) [75][800/3757] lr: 1.0000e-03 eta: 4:28:23 time: 0.3195 data_time: 0.0562 memory: 7124 grad_norm: 6.9686 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1207 loss: 1.1207 2022/09/07 15:11:08 - mmengine - INFO - Epoch(train) [75][900/3757] lr: 1.0000e-03 eta: 4:28:11 time: 0.2696 data_time: 0.0170 memory: 7124 grad_norm: 7.0108 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0412 loss: 1.0412 2022/09/07 15:11:33 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 15:11:38 - mmengine - INFO - Epoch(train) [75][1000/3757] lr: 1.0000e-03 eta: 4:27:59 time: 0.3044 data_time: 0.0142 memory: 7124 grad_norm: 7.1622 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.2594 loss: 1.2594 2022/09/07 15:12:08 - mmengine - INFO - Epoch(train) [75][1100/3757] lr: 1.0000e-03 eta: 4:27:47 time: 0.2743 data_time: 0.0164 memory: 7124 grad_norm: 6.7905 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2453 loss: 1.2453 2022/09/07 15:12:39 - mmengine - INFO - Epoch(train) [75][1200/3757] lr: 1.0000e-03 eta: 4:27:36 time: 0.2719 data_time: 0.0177 memory: 7124 grad_norm: 6.9453 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1299 loss: 1.1299 2022/09/07 15:13:09 - mmengine - INFO - Epoch(train) [75][1300/3757] lr: 1.0000e-03 eta: 4:27:24 time: 0.2790 data_time: 0.0200 memory: 7124 grad_norm: 7.0655 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.4863 loss: 1.4863 2022/09/07 15:13:40 - mmengine - INFO - Epoch(train) [75][1400/3757] lr: 1.0000e-03 eta: 4:27:12 time: 0.3300 data_time: 0.0190 memory: 7124 grad_norm: 6.8325 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0120 loss: 1.0120 2022/09/07 15:14:10 - mmengine - INFO - Epoch(train) [75][1500/3757] lr: 1.0000e-03 eta: 4:27:00 time: 0.2784 data_time: 0.0485 memory: 7124 grad_norm: 6.9990 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0553 loss: 1.0553 2022/09/07 15:14:41 - mmengine - INFO - Epoch(train) [75][1600/3757] lr: 1.0000e-03 eta: 4:26:48 time: 0.2953 data_time: 0.0169 memory: 7124 grad_norm: 6.9533 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.1841 loss: 1.1841 2022/09/07 15:15:12 - mmengine - INFO - Epoch(train) [75][1700/3757] lr: 1.0000e-03 eta: 4:26:36 time: 0.3033 data_time: 0.0159 memory: 7124 grad_norm: 7.1843 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2191 loss: 1.2191 2022/09/07 15:15:44 - mmengine - INFO - Epoch(train) [75][1800/3757] lr: 1.0000e-03 eta: 4:26:24 time: 0.3345 data_time: 0.0126 memory: 7124 grad_norm: 6.9217 top1_acc: 0.1250 top5_acc: 0.7500 loss_cls: 1.0116 loss: 1.0116 2022/09/07 15:16:13 - mmengine - INFO - Epoch(train) [75][1900/3757] lr: 1.0000e-03 eta: 4:26:12 time: 0.2842 data_time: 0.0185 memory: 7124 grad_norm: 6.9420 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1966 loss: 1.1966 2022/09/07 15:16:39 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 15:16:44 - mmengine - INFO - Epoch(train) [75][2000/3757] lr: 1.0000e-03 eta: 4:26:00 time: 0.3049 data_time: 0.0180 memory: 7124 grad_norm: 7.2448 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2730 loss: 1.2730 2022/09/07 15:17:16 - mmengine - INFO - Epoch(train) [75][2100/3757] lr: 1.0000e-03 eta: 4:25:49 time: 0.3437 data_time: 0.0195 memory: 7124 grad_norm: 6.9686 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2587 loss: 1.2587 2022/09/07 15:17:45 - mmengine - INFO - Epoch(train) [75][2200/3757] lr: 1.0000e-03 eta: 4:25:36 time: 0.2557 data_time: 0.0128 memory: 7124 grad_norm: 6.8303 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8886 loss: 0.8886 2022/09/07 15:18:16 - mmengine - INFO - Epoch(train) [75][2300/3757] lr: 1.0000e-03 eta: 4:25:24 time: 0.3112 data_time: 0.0156 memory: 7124 grad_norm: 7.3642 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.1459 loss: 1.1459 2022/09/07 15:18:48 - mmengine - INFO - Epoch(train) [75][2400/3757] lr: 1.0000e-03 eta: 4:25:13 time: 0.2941 data_time: 0.0191 memory: 7124 grad_norm: 7.0382 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1997 loss: 1.1997 2022/09/07 15:19:18 - mmengine - INFO - Epoch(train) [75][2500/3757] lr: 1.0000e-03 eta: 4:25:01 time: 0.3345 data_time: 0.0198 memory: 7124 grad_norm: 6.9146 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.8824 loss: 0.8824 2022/09/07 15:19:49 - mmengine - INFO - Epoch(train) [75][2600/3757] lr: 1.0000e-03 eta: 4:24:49 time: 0.3385 data_time: 0.0171 memory: 7124 grad_norm: 7.3296 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2024 loss: 1.2024 2022/09/07 15:20:18 - mmengine - INFO - Epoch(train) [75][2700/3757] lr: 1.0000e-03 eta: 4:24:36 time: 0.3021 data_time: 0.0162 memory: 7124 grad_norm: 6.9103 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9228 loss: 0.9228 2022/09/07 15:20:48 - mmengine - INFO - Epoch(train) [75][2800/3757] lr: 1.0000e-03 eta: 4:24:24 time: 0.3154 data_time: 0.0167 memory: 7124 grad_norm: 7.2325 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0987 loss: 1.0987 2022/09/07 15:21:19 - mmengine - INFO - Epoch(train) [75][2900/3757] lr: 1.0000e-03 eta: 4:24:12 time: 0.3071 data_time: 0.0162 memory: 7124 grad_norm: 7.0289 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8393 loss: 0.8393 2022/09/07 15:21:44 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 15:21:49 - mmengine - INFO - Epoch(train) [75][3000/3757] lr: 1.0000e-03 eta: 4:23:59 time: 0.2904 data_time: 0.0231 memory: 7124 grad_norm: 7.0136 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0322 loss: 1.0322 2022/09/07 15:22:21 - mmengine - INFO - Epoch(train) [75][3100/3757] lr: 1.0000e-03 eta: 4:23:48 time: 0.3702 data_time: 0.0163 memory: 7124 grad_norm: 7.3807 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2643 loss: 1.2643 2022/09/07 15:22:51 - mmengine - INFO - Epoch(train) [75][3200/3757] lr: 1.0000e-03 eta: 4:23:36 time: 0.3471 data_time: 0.0155 memory: 7124 grad_norm: 7.0779 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0609 loss: 1.0609 2022/09/07 15:23:21 - mmengine - INFO - Epoch(train) [75][3300/3757] lr: 1.0000e-03 eta: 4:23:23 time: 0.2839 data_time: 0.0153 memory: 7124 grad_norm: 7.2948 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0496 loss: 1.0496 2022/09/07 15:23:51 - mmengine - INFO - Epoch(train) [75][3400/3757] lr: 1.0000e-03 eta: 4:23:11 time: 0.3012 data_time: 0.0604 memory: 7124 grad_norm: 7.2025 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9812 loss: 0.9812 2022/09/07 15:24:22 - mmengine - INFO - Epoch(train) [75][3500/3757] lr: 1.0000e-03 eta: 4:22:59 time: 0.2807 data_time: 0.0127 memory: 7124 grad_norm: 6.7903 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2970 loss: 1.2970 2022/09/07 15:24:53 - mmengine - INFO - Epoch(train) [75][3600/3757] lr: 1.0000e-03 eta: 4:22:47 time: 0.3017 data_time: 0.0161 memory: 7124 grad_norm: 7.0824 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.4431 loss: 1.4431 2022/09/07 15:25:23 - mmengine - INFO - Epoch(train) [75][3700/3757] lr: 1.0000e-03 eta: 4:22:34 time: 0.3214 data_time: 0.0160 memory: 7124 grad_norm: 6.9578 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2823 loss: 1.2823 2022/09/07 15:25:39 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 15:25:39 - mmengine - INFO - Epoch(train) [75][3757/3757] lr: 1.0000e-03 eta: 4:22:29 time: 0.2302 data_time: 0.0136 memory: 7124 grad_norm: 6.8234 top1_acc: 0.7143 top5_acc: 1.0000 loss_cls: 1.2205 loss: 1.2205 2022/09/07 15:25:39 - mmengine - INFO - Saving checkpoint at 75 epochs 2022/09/07 15:30:10 - mmengine - INFO - Epoch(val) [75][100/310] eta: 0:08:27 time: 2.4167 data_time: 2.0940 memory: 7627 2022/09/07 15:34:37 - mmengine - INFO - Epoch(val) [75][200/310] eta: 0:04:15 time: 2.3266 data_time: 1.9968 memory: 7627 2022/09/07 15:38:50 - mmengine - INFO - Epoch(val) [75][300/310] eta: 0:00:26 time: 2.6593 data_time: 2.3223 memory: 7627 2022/09/07 15:39:17 - mmengine - INFO - Epoch(val) [75][310/310] acc/top1: 0.7415 acc/top5: 0.9153 acc/mean1: 0.7414 2022/09/07 15:39:42 - mmengine - INFO - Epoch(train) [76][100/3757] lr: 1.0000e-04 eta: 4:22:12 time: 0.2030 data_time: 0.0164 memory: 7627 grad_norm: 6.9235 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0775 loss: 1.0775 2022/09/07 15:40:02 - mmengine - INFO - Epoch(train) [76][200/3757] lr: 1.0000e-04 eta: 4:21:56 time: 0.1583 data_time: 0.0118 memory: 7124 grad_norm: 7.1651 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0670 loss: 1.0670 2022/09/07 15:40:06 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 15:40:18 - mmengine - INFO - Epoch(train) [76][300/3757] lr: 1.0000e-04 eta: 4:21:39 time: 0.1632 data_time: 0.0113 memory: 7124 grad_norm: 6.9246 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8918 loss: 0.8918 2022/09/07 15:40:34 - mmengine - INFO - Epoch(train) [76][400/3757] lr: 1.0000e-04 eta: 4:21:22 time: 0.1606 data_time: 0.0109 memory: 7124 grad_norm: 6.9225 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.7645 loss: 0.7645 2022/09/07 15:40:50 - mmengine - INFO - Epoch(train) [76][500/3757] lr: 1.0000e-04 eta: 4:21:05 time: 0.1588 data_time: 0.0097 memory: 7124 grad_norm: 6.9510 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1087 loss: 1.1087 2022/09/07 15:41:06 - mmengine - INFO - Epoch(train) [76][600/3757] lr: 1.0000e-04 eta: 4:20:48 time: 0.1549 data_time: 0.0095 memory: 7124 grad_norm: 7.0745 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2132 loss: 1.2132 2022/09/07 15:41:22 - mmengine - INFO - Epoch(train) [76][700/3757] lr: 1.0000e-04 eta: 4:20:31 time: 0.1535 data_time: 0.0106 memory: 7124 grad_norm: 6.9093 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2504 loss: 1.2504 2022/09/07 15:41:38 - mmengine - INFO - Epoch(train) [76][800/3757] lr: 1.0000e-04 eta: 4:20:14 time: 0.1609 data_time: 0.0098 memory: 7124 grad_norm: 6.8834 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0698 loss: 1.0698 2022/09/07 15:41:54 - mmengine - INFO - Epoch(train) [76][900/3757] lr: 1.0000e-04 eta: 4:19:57 time: 0.1568 data_time: 0.0110 memory: 7124 grad_norm: 6.8071 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9013 loss: 0.9013 2022/09/07 15:42:10 - mmengine - INFO - Epoch(train) [76][1000/3757] lr: 1.0000e-04 eta: 4:19:40 time: 0.1542 data_time: 0.0115 memory: 7124 grad_norm: 7.2944 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0551 loss: 1.0551 2022/09/07 15:42:26 - mmengine - INFO - Epoch(train) [76][1100/3757] lr: 1.0000e-04 eta: 4:19:22 time: 0.1577 data_time: 0.0106 memory: 7124 grad_norm: 7.0135 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0452 loss: 1.0452 2022/09/07 15:42:42 - mmengine - INFO - Epoch(train) [76][1200/3757] lr: 1.0000e-04 eta: 4:19:05 time: 0.1576 data_time: 0.0108 memory: 7124 grad_norm: 6.8036 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0505 loss: 1.0505 2022/09/07 15:42:46 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 15:42:58 - mmengine - INFO - Epoch(train) [76][1300/3757] lr: 1.0000e-04 eta: 4:18:48 time: 0.1606 data_time: 0.0110 memory: 7124 grad_norm: 6.9892 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9925 loss: 0.9925 2022/09/07 15:43:14 - mmengine - INFO - Epoch(train) [76][1400/3757] lr: 1.0000e-04 eta: 4:18:31 time: 0.1638 data_time: 0.0097 memory: 7124 grad_norm: 7.2164 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0490 loss: 1.0490 2022/09/07 15:43:30 - mmengine - INFO - Epoch(train) [76][1500/3757] lr: 1.0000e-04 eta: 4:18:14 time: 0.1570 data_time: 0.0097 memory: 7124 grad_norm: 6.7270 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2211 loss: 1.2211 2022/09/07 15:43:46 - mmengine - INFO - Epoch(train) [76][1600/3757] lr: 1.0000e-04 eta: 4:17:57 time: 0.1560 data_time: 0.0096 memory: 7124 grad_norm: 6.6978 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.6887 loss: 0.6887 2022/09/07 15:44:02 - mmengine - INFO - Epoch(train) [76][1700/3757] lr: 1.0000e-04 eta: 4:17:40 time: 0.1561 data_time: 0.0105 memory: 7124 grad_norm: 7.0837 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1527 loss: 1.1527 2022/09/07 15:44:18 - mmengine - INFO - Epoch(train) [76][1800/3757] lr: 1.0000e-04 eta: 4:17:23 time: 0.1637 data_time: 0.0108 memory: 7124 grad_norm: 6.9238 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.0818 loss: 1.0818 2022/09/07 15:44:34 - mmengine - INFO - Epoch(train) [76][1900/3757] lr: 1.0000e-04 eta: 4:17:06 time: 0.1604 data_time: 0.0103 memory: 7124 grad_norm: 6.9515 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1430 loss: 1.1430 2022/09/07 15:44:50 - mmengine - INFO - Epoch(train) [76][2000/3757] lr: 1.0000e-04 eta: 4:16:49 time: 0.1594 data_time: 0.0098 memory: 7124 grad_norm: 6.7756 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9477 loss: 0.9477 2022/09/07 15:45:06 - mmengine - INFO - Epoch(train) [76][2100/3757] lr: 1.0000e-04 eta: 4:16:32 time: 0.1612 data_time: 0.0099 memory: 7124 grad_norm: 6.9609 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0999 loss: 1.0999 2022/09/07 15:45:22 - mmengine - INFO - Epoch(train) [76][2200/3757] lr: 1.0000e-04 eta: 4:16:15 time: 0.1576 data_time: 0.0108 memory: 7124 grad_norm: 7.1999 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0517 loss: 1.0517 2022/09/07 15:45:26 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 15:45:38 - mmengine - INFO - Epoch(train) [76][2300/3757] lr: 1.0000e-04 eta: 4:15:58 time: 0.1610 data_time: 0.0115 memory: 7124 grad_norm: 6.9109 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0691 loss: 1.0691 2022/09/07 15:45:54 - mmengine - INFO - Epoch(train) [76][2400/3757] lr: 1.0000e-04 eta: 4:15:41 time: 0.1567 data_time: 0.0112 memory: 7124 grad_norm: 7.0187 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0578 loss: 1.0578 2022/09/07 15:46:10 - mmengine - INFO - Epoch(train) [76][2500/3757] lr: 1.0000e-04 eta: 4:15:24 time: 0.1536 data_time: 0.0110 memory: 7124 grad_norm: 6.9468 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0005 loss: 1.0005 2022/09/07 15:46:26 - mmengine - INFO - Epoch(train) [76][2600/3757] lr: 1.0000e-04 eta: 4:15:07 time: 0.1550 data_time: 0.0106 memory: 7124 grad_norm: 6.8295 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1535 loss: 1.1535 2022/09/07 15:46:42 - mmengine - INFO - Epoch(train) [76][2700/3757] lr: 1.0000e-04 eta: 4:14:50 time: 0.1568 data_time: 0.0120 memory: 7124 grad_norm: 7.3387 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0635 loss: 1.0635 2022/09/07 15:46:58 - mmengine - INFO - Epoch(train) [76][2800/3757] lr: 1.0000e-04 eta: 4:14:33 time: 0.1607 data_time: 0.0119 memory: 7124 grad_norm: 6.8708 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0770 loss: 1.0770 2022/09/07 15:47:14 - mmengine - INFO - Epoch(train) [76][2900/3757] lr: 1.0000e-04 eta: 4:14:16 time: 0.1566 data_time: 0.0101 memory: 7124 grad_norm: 7.0832 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.2197 loss: 1.2197 2022/09/07 15:47:30 - mmengine - INFO - Epoch(train) [76][3000/3757] lr: 1.0000e-04 eta: 4:13:59 time: 0.1576 data_time: 0.0116 memory: 7124 grad_norm: 6.8391 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.3266 loss: 1.3266 2022/09/07 15:47:46 - mmengine - INFO - Epoch(train) [76][3100/3757] lr: 1.0000e-04 eta: 4:13:42 time: 0.1556 data_time: 0.0114 memory: 7124 grad_norm: 6.7908 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1249 loss: 1.1249 2022/09/07 15:48:02 - mmengine - INFO - Epoch(train) [76][3200/3757] lr: 1.0000e-04 eta: 4:13:25 time: 0.1572 data_time: 0.0106 memory: 7124 grad_norm: 6.8272 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1172 loss: 1.1172 2022/09/07 15:48:06 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 15:48:18 - mmengine - INFO - Epoch(train) [76][3300/3757] lr: 1.0000e-04 eta: 4:13:08 time: 0.1588 data_time: 0.0114 memory: 7124 grad_norm: 6.9067 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9972 loss: 0.9972 2022/09/07 15:48:34 - mmengine - INFO - Epoch(train) [76][3400/3757] lr: 1.0000e-04 eta: 4:12:51 time: 0.1601 data_time: 0.0106 memory: 7124 grad_norm: 6.9843 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0065 loss: 1.0065 2022/09/07 15:48:50 - mmengine - INFO - Epoch(train) [76][3500/3757] lr: 1.0000e-04 eta: 4:12:34 time: 0.1671 data_time: 0.0123 memory: 7124 grad_norm: 7.0907 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.3683 loss: 1.3683 2022/09/07 15:49:06 - mmengine - INFO - Epoch(train) [76][3600/3757] lr: 1.0000e-04 eta: 4:12:17 time: 0.1560 data_time: 0.0104 memory: 7124 grad_norm: 7.0186 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1392 loss: 1.1392 2022/09/07 15:49:22 - mmengine - INFO - Epoch(train) [76][3700/3757] lr: 1.0000e-04 eta: 4:12:00 time: 0.1591 data_time: 0.0108 memory: 7124 grad_norm: 6.7115 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9422 loss: 0.9422 2022/09/07 15:49:31 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 15:49:31 - mmengine - INFO - Epoch(train) [76][3757/3757] lr: 1.0000e-04 eta: 4:11:53 time: 0.1371 data_time: 0.0076 memory: 7124 grad_norm: 6.8191 top1_acc: 0.7143 top5_acc: 0.8571 loss_cls: 0.8862 loss: 0.8862 2022/09/07 15:49:31 - mmengine - INFO - Saving checkpoint at 76 epochs 2022/09/07 15:51:47 - mmengine - INFO - Epoch(val) [76][100/310] eta: 0:03:40 time: 1.0479 data_time: 0.7410 memory: 7627 2022/09/07 15:54:07 - mmengine - INFO - Epoch(val) [76][200/310] eta: 0:02:31 time: 1.3798 data_time: 1.0686 memory: 7627 2022/09/07 15:56:11 - mmengine - INFO - Epoch(val) [76][300/310] eta: 0:00:11 time: 1.1151 data_time: 0.8157 memory: 7627 2022/09/07 15:56:27 - mmengine - INFO - Epoch(val) [76][310/310] acc/top1: 0.7505 acc/top5: 0.9185 acc/mean1: 0.7504 2022/09/07 15:56:27 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_71.pth is removed 2022/09/07 15:56:29 - mmengine - INFO - The best checkpoint with 0.7505 acc/top1 at 76 epoch is saved to best_acc/top1_epoch_76.pth. 2022/09/07 15:56:46 - mmengine - INFO - Epoch(train) [77][100/3757] lr: 1.0000e-04 eta: 4:11:32 time: 0.1613 data_time: 0.0117 memory: 7627 grad_norm: 6.8775 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0398 loss: 1.0398 2022/09/07 15:57:02 - mmengine - INFO - Epoch(train) [77][200/3757] lr: 1.0000e-04 eta: 4:11:15 time: 0.1607 data_time: 0.0112 memory: 7124 grad_norm: 7.0707 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9508 loss: 0.9508 2022/09/07 15:57:18 - mmengine - INFO - Epoch(train) [77][300/3757] lr: 1.0000e-04 eta: 4:10:58 time: 0.1688 data_time: 0.0113 memory: 7124 grad_norm: 7.0310 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9292 loss: 0.9292 2022/09/07 15:57:35 - mmengine - INFO - Epoch(train) [77][400/3757] lr: 1.0000e-04 eta: 4:10:41 time: 0.1582 data_time: 0.0120 memory: 7124 grad_norm: 7.0961 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1364 loss: 1.1364 2022/09/07 15:57:46 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 15:57:51 - mmengine - INFO - Epoch(train) [77][500/3757] lr: 1.0000e-04 eta: 4:10:24 time: 0.1592 data_time: 0.0113 memory: 7124 grad_norm: 6.8799 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1634 loss: 1.1634 2022/09/07 15:58:07 - mmengine - INFO - Epoch(train) [77][600/3757] lr: 1.0000e-04 eta: 4:10:07 time: 0.1597 data_time: 0.0113 memory: 7124 grad_norm: 6.8818 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1144 loss: 1.1144 2022/09/07 15:58:23 - mmengine - INFO - Epoch(train) [77][700/3757] lr: 1.0000e-04 eta: 4:09:50 time: 0.1572 data_time: 0.0112 memory: 7124 grad_norm: 6.9941 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0596 loss: 1.0596 2022/09/07 15:58:39 - mmengine - INFO - Epoch(train) [77][800/3757] lr: 1.0000e-04 eta: 4:09:33 time: 0.1580 data_time: 0.0113 memory: 7124 grad_norm: 6.8885 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1034 loss: 1.1034 2022/09/07 15:58:55 - mmengine - INFO - Epoch(train) [77][900/3757] lr: 1.0000e-04 eta: 4:09:16 time: 0.1584 data_time: 0.0120 memory: 7124 grad_norm: 7.0417 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0329 loss: 1.0329 2022/09/07 15:59:11 - mmengine - INFO - Epoch(train) [77][1000/3757] lr: 1.0000e-04 eta: 4:08:59 time: 0.1562 data_time: 0.0107 memory: 7124 grad_norm: 6.9818 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9516 loss: 0.9516 2022/09/07 15:59:28 - mmengine - INFO - Epoch(train) [77][1100/3757] lr: 1.0000e-04 eta: 4:08:42 time: 0.1563 data_time: 0.0104 memory: 7124 grad_norm: 6.6848 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.0823 loss: 1.0823 2022/09/07 15:59:44 - mmengine - INFO - Epoch(train) [77][1200/3757] lr: 1.0000e-04 eta: 4:08:25 time: 0.1567 data_time: 0.0112 memory: 7124 grad_norm: 7.3920 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0290 loss: 1.0290 2022/09/07 16:00:00 - mmengine - INFO - Epoch(train) [77][1300/3757] lr: 1.0000e-04 eta: 4:08:08 time: 0.1603 data_time: 0.0109 memory: 7124 grad_norm: 6.9635 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0558 loss: 1.0558 2022/09/07 16:00:16 - mmengine - INFO - Epoch(train) [77][1400/3757] lr: 1.0000e-04 eta: 4:07:51 time: 0.1673 data_time: 0.0108 memory: 7124 grad_norm: 6.9553 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0135 loss: 1.0135 2022/09/07 16:00:27 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:00:32 - mmengine - INFO - Epoch(train) [77][1500/3757] lr: 1.0000e-04 eta: 4:07:34 time: 0.1582 data_time: 0.0120 memory: 7124 grad_norm: 6.8793 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.8648 loss: 0.8648 2022/09/07 16:00:48 - mmengine - INFO - Epoch(train) [77][1600/3757] lr: 1.0000e-04 eta: 4:07:17 time: 0.1592 data_time: 0.0107 memory: 7124 grad_norm: 7.0250 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0645 loss: 1.0645 2022/09/07 16:01:04 - mmengine - INFO - Epoch(train) [77][1700/3757] lr: 1.0000e-04 eta: 4:07:00 time: 0.1614 data_time: 0.0110 memory: 7124 grad_norm: 6.9447 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0990 loss: 1.0990 2022/09/07 16:01:21 - mmengine - INFO - Epoch(train) [77][1800/3757] lr: 1.0000e-04 eta: 4:06:44 time: 0.1601 data_time: 0.0116 memory: 7124 grad_norm: 6.8282 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8001 loss: 0.8001 2022/09/07 16:01:37 - mmengine - INFO - Epoch(train) [77][1900/3757] lr: 1.0000e-04 eta: 4:06:27 time: 0.1565 data_time: 0.0109 memory: 7124 grad_norm: 7.0188 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0802 loss: 1.0802 2022/09/07 16:01:53 - mmengine - INFO - Epoch(train) [77][2000/3757] lr: 1.0000e-04 eta: 4:06:10 time: 0.1565 data_time: 0.0116 memory: 7124 grad_norm: 6.9275 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9794 loss: 0.9794 2022/09/07 16:02:09 - mmengine - INFO - Epoch(train) [77][2100/3757] lr: 1.0000e-04 eta: 4:05:53 time: 0.1650 data_time: 0.0116 memory: 7124 grad_norm: 6.8796 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8865 loss: 0.8865 2022/09/07 16:02:25 - mmengine - INFO - Epoch(train) [77][2200/3757] lr: 1.0000e-04 eta: 4:05:36 time: 0.1607 data_time: 0.0119 memory: 7124 grad_norm: 6.8692 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2164 loss: 1.2164 2022/09/07 16:02:41 - mmengine - INFO - Epoch(train) [77][2300/3757] lr: 1.0000e-04 eta: 4:05:19 time: 0.1582 data_time: 0.0109 memory: 7124 grad_norm: 7.2072 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1519 loss: 1.1519 2022/09/07 16:02:58 - mmengine - INFO - Epoch(train) [77][2400/3757] lr: 1.0000e-04 eta: 4:05:02 time: 0.1587 data_time: 0.0109 memory: 7124 grad_norm: 7.0441 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0012 loss: 1.0012 2022/09/07 16:03:08 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:03:14 - mmengine - INFO - Epoch(train) [77][2500/3757] lr: 1.0000e-04 eta: 4:04:45 time: 0.1589 data_time: 0.0108 memory: 7124 grad_norm: 7.2985 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9650 loss: 0.9650 2022/09/07 16:03:30 - mmengine - INFO - Epoch(train) [77][2600/3757] lr: 1.0000e-04 eta: 4:04:28 time: 0.1709 data_time: 0.0111 memory: 7124 grad_norm: 7.0903 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.1607 loss: 1.1607 2022/09/07 16:03:46 - mmengine - INFO - Epoch(train) [77][2700/3757] lr: 1.0000e-04 eta: 4:04:11 time: 0.1574 data_time: 0.0108 memory: 7124 grad_norm: 6.9785 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.9919 loss: 0.9919 2022/09/07 16:04:03 - mmengine - INFO - Epoch(train) [77][2800/3757] lr: 1.0000e-04 eta: 4:03:54 time: 0.1611 data_time: 0.0109 memory: 7124 grad_norm: 7.1311 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1649 loss: 1.1649 2022/09/07 16:04:19 - mmengine - INFO - Epoch(train) [77][2900/3757] lr: 1.0000e-04 eta: 4:03:37 time: 0.1639 data_time: 0.0107 memory: 7124 grad_norm: 6.8877 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.7842 loss: 0.7842 2022/09/07 16:04:35 - mmengine - INFO - Epoch(train) [77][3000/3757] lr: 1.0000e-04 eta: 4:03:20 time: 0.1636 data_time: 0.0122 memory: 7124 grad_norm: 7.0609 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0277 loss: 1.0277 2022/09/07 16:04:52 - mmengine - INFO - Epoch(train) [77][3100/3757] lr: 1.0000e-04 eta: 4:03:03 time: 0.1585 data_time: 0.0126 memory: 7124 grad_norm: 7.1931 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2746 loss: 1.2746 2022/09/07 16:05:08 - mmengine - INFO - Epoch(train) [77][3200/3757] lr: 1.0000e-04 eta: 4:02:46 time: 0.1582 data_time: 0.0116 memory: 7124 grad_norm: 6.7733 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1552 loss: 1.1552 2022/09/07 16:05:24 - mmengine - INFO - Epoch(train) [77][3300/3757] lr: 1.0000e-04 eta: 4:02:29 time: 0.1576 data_time: 0.0108 memory: 7124 grad_norm: 6.7891 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9105 loss: 0.9105 2022/09/07 16:05:40 - mmengine - INFO - Epoch(train) [77][3400/3757] lr: 1.0000e-04 eta: 4:02:12 time: 0.1647 data_time: 0.0107 memory: 7124 grad_norm: 7.4183 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9892 loss: 0.9892 2022/09/07 16:05:51 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:05:56 - mmengine - INFO - Epoch(train) [77][3500/3757] lr: 1.0000e-04 eta: 4:01:55 time: 0.1609 data_time: 0.0112 memory: 7124 grad_norm: 7.1601 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1189 loss: 1.1189 2022/09/07 16:06:12 - mmengine - INFO - Epoch(train) [77][3600/3757] lr: 1.0000e-04 eta: 4:01:38 time: 0.1568 data_time: 0.0111 memory: 7124 grad_norm: 6.9505 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9834 loss: 0.9834 2022/09/07 16:06:29 - mmengine - INFO - Epoch(train) [77][3700/3757] lr: 1.0000e-04 eta: 4:01:21 time: 0.1597 data_time: 0.0109 memory: 7124 grad_norm: 7.1949 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9830 loss: 0.9830 2022/09/07 16:06:37 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:06:37 - mmengine - INFO - Epoch(train) [77][3757/3757] lr: 1.0000e-04 eta: 4:01:15 time: 0.1375 data_time: 0.0074 memory: 7124 grad_norm: 7.1225 top1_acc: 0.5714 top5_acc: 0.8571 loss_cls: 1.1354 loss: 1.1354 2022/09/07 16:06:37 - mmengine - INFO - Saving checkpoint at 77 epochs 2022/09/07 16:08:55 - mmengine - INFO - Epoch(val) [77][100/310] eta: 0:03:43 time: 1.0620 data_time: 0.7595 memory: 7627 2022/09/07 16:11:15 - mmengine - INFO - Epoch(val) [77][200/310] eta: 0:02:32 time: 1.3860 data_time: 1.0829 memory: 7627 2022/09/07 16:13:17 - mmengine - INFO - Epoch(val) [77][300/310] eta: 0:00:10 time: 1.0972 data_time: 0.7951 memory: 7627 2022/09/07 16:13:34 - mmengine - INFO - Epoch(val) [77][310/310] acc/top1: 0.7492 acc/top5: 0.9168 acc/mean1: 0.7491 2022/09/07 16:13:52 - mmengine - INFO - Epoch(train) [78][100/3757] lr: 1.0000e-04 eta: 4:00:55 time: 0.1635 data_time: 0.0104 memory: 7627 grad_norm: 6.7944 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0129 loss: 1.0129 2022/09/07 16:14:08 - mmengine - INFO - Epoch(train) [78][200/3757] lr: 1.0000e-04 eta: 4:00:38 time: 0.1621 data_time: 0.0107 memory: 7124 grad_norm: 6.7675 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0224 loss: 1.0224 2022/09/07 16:14:24 - mmengine - INFO - Epoch(train) [78][300/3757] lr: 1.0000e-04 eta: 4:00:21 time: 0.1605 data_time: 0.0116 memory: 7124 grad_norm: 7.1269 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9810 loss: 0.9810 2022/09/07 16:14:41 - mmengine - INFO - Epoch(train) [78][400/3757] lr: 1.0000e-04 eta: 4:00:04 time: 0.1623 data_time: 0.0105 memory: 7124 grad_norm: 7.0727 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1000 loss: 1.1000 2022/09/07 16:14:57 - mmengine - INFO - Epoch(train) [78][500/3757] lr: 1.0000e-04 eta: 3:59:47 time: 0.1584 data_time: 0.0107 memory: 7124 grad_norm: 7.1220 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0671 loss: 1.0671 2022/09/07 16:15:15 - mmengine - INFO - Epoch(train) [78][600/3757] lr: 1.0000e-04 eta: 3:59:30 time: 0.1505 data_time: 0.0087 memory: 7124 grad_norm: 6.9289 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1540 loss: 1.1540 2022/09/07 16:15:33 - mmengine - INFO - Epoch(train) [78][700/3757] lr: 1.0000e-04 eta: 3:59:14 time: 0.2070 data_time: 0.0089 memory: 7124 grad_norm: 6.8189 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9692 loss: 0.9692 2022/09/07 16:15:35 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:15:49 - mmengine - INFO - Epoch(train) [78][800/3757] lr: 1.0000e-04 eta: 3:58:57 time: 0.1600 data_time: 0.0114 memory: 7124 grad_norm: 7.1464 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.1113 loss: 1.1113 2022/09/07 16:16:06 - mmengine - INFO - Epoch(train) [78][900/3757] lr: 1.0000e-04 eta: 3:58:40 time: 0.1612 data_time: 0.0112 memory: 7124 grad_norm: 6.8769 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1040 loss: 1.1040 2022/09/07 16:16:22 - mmengine - INFO - Epoch(train) [78][1000/3757] lr: 1.0000e-04 eta: 3:58:23 time: 0.1602 data_time: 0.0096 memory: 7124 grad_norm: 6.9677 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1616 loss: 1.1616 2022/09/07 16:16:38 - mmengine - INFO - Epoch(train) [78][1100/3757] lr: 1.0000e-04 eta: 3:58:06 time: 0.1578 data_time: 0.0105 memory: 7124 grad_norm: 7.0625 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0131 loss: 1.0131 2022/09/07 16:16:54 - mmengine - INFO - Epoch(train) [78][1200/3757] lr: 1.0000e-04 eta: 3:57:49 time: 0.1589 data_time: 0.0113 memory: 7124 grad_norm: 7.0668 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.1575 loss: 1.1575 2022/09/07 16:17:10 - mmengine - INFO - Epoch(train) [78][1300/3757] lr: 1.0000e-04 eta: 3:57:32 time: 0.1573 data_time: 0.0114 memory: 7124 grad_norm: 7.1844 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2593 loss: 1.2593 2022/09/07 16:17:26 - mmengine - INFO - Epoch(train) [78][1400/3757] lr: 1.0000e-04 eta: 3:57:15 time: 0.1605 data_time: 0.0122 memory: 7124 grad_norm: 6.7633 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9675 loss: 0.9675 2022/09/07 16:17:42 - mmengine - INFO - Epoch(train) [78][1500/3757] lr: 1.0000e-04 eta: 3:56:59 time: 0.1620 data_time: 0.0112 memory: 7124 grad_norm: 6.9645 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8601 loss: 0.8601 2022/09/07 16:17:59 - mmengine - INFO - Epoch(train) [78][1600/3757] lr: 1.0000e-04 eta: 3:56:42 time: 0.1596 data_time: 0.0110 memory: 7124 grad_norm: 6.8781 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2656 loss: 1.2656 2022/09/07 16:18:15 - mmengine - INFO - Epoch(train) [78][1700/3757] lr: 1.0000e-04 eta: 3:56:25 time: 0.1583 data_time: 0.0104 memory: 7124 grad_norm: 7.1586 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9665 loss: 0.9665 2022/09/07 16:18:17 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:18:31 - mmengine - INFO - Epoch(train) [78][1800/3757] lr: 1.0000e-04 eta: 3:56:08 time: 0.1639 data_time: 0.0113 memory: 7124 grad_norm: 6.7840 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1000 loss: 1.1000 2022/09/07 16:18:47 - mmengine - INFO - Epoch(train) [78][1900/3757] lr: 1.0000e-04 eta: 3:55:51 time: 0.1642 data_time: 0.0124 memory: 7124 grad_norm: 6.9216 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.7903 loss: 0.7903 2022/09/07 16:19:03 - mmengine - INFO - Epoch(train) [78][2000/3757] lr: 1.0000e-04 eta: 3:55:34 time: 0.1599 data_time: 0.0115 memory: 7124 grad_norm: 7.0304 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.0422 loss: 1.0422 2022/09/07 16:19:20 - mmengine - INFO - Epoch(train) [78][2100/3757] lr: 1.0000e-04 eta: 3:55:17 time: 0.1577 data_time: 0.0111 memory: 7124 grad_norm: 7.1320 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8140 loss: 0.8140 2022/09/07 16:19:36 - mmengine - INFO - Epoch(train) [78][2200/3757] lr: 1.0000e-04 eta: 3:55:00 time: 0.1525 data_time: 0.0103 memory: 7124 grad_norm: 7.2645 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2128 loss: 1.2128 2022/09/07 16:19:56 - mmengine - INFO - Epoch(train) [78][2300/3757] lr: 1.0000e-04 eta: 3:54:44 time: 0.1569 data_time: 0.0100 memory: 7124 grad_norm: 7.0036 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9162 loss: 0.9162 2022/09/07 16:20:12 - mmengine - INFO - Epoch(train) [78][2400/3757] lr: 1.0000e-04 eta: 3:54:27 time: 0.1657 data_time: 0.0113 memory: 7124 grad_norm: 7.1902 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3202 loss: 1.3202 2022/09/07 16:20:29 - mmengine - INFO - Epoch(train) [78][2500/3757] lr: 1.0000e-04 eta: 3:54:10 time: 0.1562 data_time: 0.0107 memory: 7124 grad_norm: 6.5481 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.0427 loss: 1.0427 2022/09/07 16:20:45 - mmengine - INFO - Epoch(train) [78][2600/3757] lr: 1.0000e-04 eta: 3:53:53 time: 0.1623 data_time: 0.0129 memory: 7124 grad_norm: 7.1095 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9416 loss: 0.9416 2022/09/07 16:21:01 - mmengine - INFO - Epoch(train) [78][2700/3757] lr: 1.0000e-04 eta: 3:53:37 time: 0.1587 data_time: 0.0108 memory: 7124 grad_norm: 6.9337 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8673 loss: 0.8673 2022/09/07 16:21:03 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:21:18 - mmengine - INFO - Epoch(train) [78][2800/3757] lr: 1.0000e-04 eta: 3:53:20 time: 0.1600 data_time: 0.0119 memory: 7124 grad_norm: 7.0623 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0940 loss: 1.0940 2022/09/07 16:21:34 - mmengine - INFO - Epoch(train) [78][2900/3757] lr: 1.0000e-04 eta: 3:53:03 time: 0.1593 data_time: 0.0102 memory: 7124 grad_norm: 7.3475 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.1060 loss: 1.1060 2022/09/07 16:21:50 - mmengine - INFO - Epoch(train) [78][3000/3757] lr: 1.0000e-04 eta: 3:52:46 time: 0.1627 data_time: 0.0114 memory: 7124 grad_norm: 6.9897 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2007 loss: 1.2007 2022/09/07 16:22:06 - mmengine - INFO - Epoch(train) [78][3100/3757] lr: 1.0000e-04 eta: 3:52:29 time: 0.1584 data_time: 0.0109 memory: 7124 grad_norm: 6.8669 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9774 loss: 0.9774 2022/09/07 16:22:22 - mmengine - INFO - Epoch(train) [78][3200/3757] lr: 1.0000e-04 eta: 3:52:12 time: 0.1621 data_time: 0.0123 memory: 7124 grad_norm: 6.8607 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9309 loss: 0.9309 2022/09/07 16:22:38 - mmengine - INFO - Epoch(train) [78][3300/3757] lr: 1.0000e-04 eta: 3:51:55 time: 0.1592 data_time: 0.0113 memory: 7124 grad_norm: 7.3888 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1796 loss: 1.1796 2022/09/07 16:22:55 - mmengine - INFO - Epoch(train) [78][3400/3757] lr: 1.0000e-04 eta: 3:51:38 time: 0.1588 data_time: 0.0112 memory: 7124 grad_norm: 6.7134 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8667 loss: 0.8667 2022/09/07 16:23:11 - mmengine - INFO - Epoch(train) [78][3500/3757] lr: 1.0000e-04 eta: 3:51:21 time: 0.1635 data_time: 0.0117 memory: 7124 grad_norm: 7.2458 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.3526 loss: 1.3526 2022/09/07 16:23:27 - mmengine - INFO - Epoch(train) [78][3600/3757] lr: 1.0000e-04 eta: 3:51:04 time: 0.1577 data_time: 0.0107 memory: 7124 grad_norm: 6.9387 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2024 loss: 1.2024 2022/09/07 16:23:43 - mmengine - INFO - Epoch(train) [78][3700/3757] lr: 1.0000e-04 eta: 3:50:47 time: 0.1570 data_time: 0.0120 memory: 7124 grad_norm: 7.2669 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.6990 loss: 0.6990 2022/09/07 16:23:45 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:23:52 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:23:52 - mmengine - INFO - Epoch(train) [78][3757/3757] lr: 1.0000e-04 eta: 3:50:40 time: 0.1452 data_time: 0.0080 memory: 7124 grad_norm: 6.8978 top1_acc: 0.5714 top5_acc: 0.8571 loss_cls: 0.9091 loss: 0.9091 2022/09/07 16:23:52 - mmengine - INFO - Saving checkpoint at 78 epochs 2022/09/07 16:26:11 - mmengine - INFO - Epoch(val) [78][100/310] eta: 0:04:12 time: 1.2035 data_time: 0.8993 memory: 7627 2022/09/07 16:28:25 - mmengine - INFO - Epoch(val) [78][200/310] eta: 0:02:06 time: 1.1475 data_time: 0.8467 memory: 7627 2022/09/07 16:30:33 - mmengine - INFO - Epoch(val) [78][300/310] eta: 0:00:13 time: 1.3218 data_time: 1.0210 memory: 7627 2022/09/07 16:30:53 - mmengine - INFO - Epoch(val) [78][310/310] acc/top1: 0.7533 acc/top5: 0.9176 acc/mean1: 0.7532 2022/09/07 16:30:53 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_76.pth is removed 2022/09/07 16:30:55 - mmengine - INFO - The best checkpoint with 0.7533 acc/top1 at 78 epoch is saved to best_acc/top1_epoch_78.pth. 2022/09/07 16:31:13 - mmengine - INFO - Epoch(train) [79][100/3757] lr: 1.0000e-04 eta: 3:50:20 time: 0.1630 data_time: 0.0104 memory: 7627 grad_norm: 7.0811 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0005 loss: 1.0005 2022/09/07 16:31:30 - mmengine - INFO - Epoch(train) [79][200/3757] lr: 1.0000e-04 eta: 3:50:03 time: 0.1686 data_time: 0.0117 memory: 7124 grad_norm: 7.2005 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0288 loss: 1.0288 2022/09/07 16:31:46 - mmengine - INFO - Epoch(train) [79][300/3757] lr: 1.0000e-04 eta: 3:49:47 time: 0.1628 data_time: 0.0113 memory: 7124 grad_norm: 6.7924 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0146 loss: 1.0146 2022/09/07 16:32:03 - mmengine - INFO - Epoch(train) [79][400/3757] lr: 1.0000e-04 eta: 3:49:30 time: 0.1643 data_time: 0.0106 memory: 7124 grad_norm: 6.9535 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8360 loss: 0.8360 2022/09/07 16:32:19 - mmengine - INFO - Epoch(train) [79][500/3757] lr: 1.0000e-04 eta: 3:49:13 time: 0.1617 data_time: 0.0122 memory: 7124 grad_norm: 6.7331 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8363 loss: 0.8363 2022/09/07 16:32:35 - mmengine - INFO - Epoch(train) [79][600/3757] lr: 1.0000e-04 eta: 3:48:56 time: 0.1607 data_time: 0.0116 memory: 7124 grad_norm: 7.0668 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 0.9109 loss: 0.9109 2022/09/07 16:32:52 - mmengine - INFO - Epoch(train) [79][700/3757] lr: 1.0000e-04 eta: 3:48:39 time: 0.1608 data_time: 0.0104 memory: 7124 grad_norm: 7.0212 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9847 loss: 0.9847 2022/09/07 16:33:08 - mmengine - INFO - Epoch(train) [79][800/3757] lr: 1.0000e-04 eta: 3:48:22 time: 0.1633 data_time: 0.0108 memory: 7124 grad_norm: 7.1316 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2419 loss: 1.2419 2022/09/07 16:33:25 - mmengine - INFO - Epoch(train) [79][900/3757] lr: 1.0000e-04 eta: 3:48:05 time: 0.1661 data_time: 0.0094 memory: 7124 grad_norm: 7.0940 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9779 loss: 0.9779 2022/09/07 16:33:34 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:33:41 - mmengine - INFO - Epoch(train) [79][1000/3757] lr: 1.0000e-04 eta: 3:47:49 time: 0.1617 data_time: 0.0109 memory: 7124 grad_norm: 6.8203 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0163 loss: 1.0163 2022/09/07 16:33:57 - mmengine - INFO - Epoch(train) [79][1100/3757] lr: 1.0000e-04 eta: 3:47:32 time: 0.1610 data_time: 0.0115 memory: 7124 grad_norm: 7.1511 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0541 loss: 1.0541 2022/09/07 16:34:14 - mmengine - INFO - Epoch(train) [79][1200/3757] lr: 1.0000e-04 eta: 3:47:15 time: 0.1639 data_time: 0.0113 memory: 7124 grad_norm: 6.9948 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1007 loss: 1.1007 2022/09/07 16:34:30 - mmengine - INFO - Epoch(train) [79][1300/3757] lr: 1.0000e-04 eta: 3:46:58 time: 0.1607 data_time: 0.0103 memory: 7124 grad_norm: 6.8953 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9448 loss: 0.9448 2022/09/07 16:34:47 - mmengine - INFO - Epoch(train) [79][1400/3757] lr: 1.0000e-04 eta: 3:46:41 time: 0.1586 data_time: 0.0113 memory: 7124 grad_norm: 7.2157 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1294 loss: 1.1294 2022/09/07 16:35:06 - mmengine - INFO - Epoch(train) [79][1500/3757] lr: 1.0000e-04 eta: 3:46:25 time: 0.1628 data_time: 0.0115 memory: 7124 grad_norm: 6.8537 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0798 loss: 1.0798 2022/09/07 16:35:22 - mmengine - INFO - Epoch(train) [79][1600/3757] lr: 1.0000e-04 eta: 3:46:08 time: 0.1625 data_time: 0.0124 memory: 7124 grad_norm: 7.0210 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9775 loss: 0.9775 2022/09/07 16:35:39 - mmengine - INFO - Epoch(train) [79][1700/3757] lr: 1.0000e-04 eta: 3:45:51 time: 0.1662 data_time: 0.0125 memory: 7124 grad_norm: 7.0383 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8533 loss: 0.8533 2022/09/07 16:35:55 - mmengine - INFO - Epoch(train) [79][1800/3757] lr: 1.0000e-04 eta: 3:45:34 time: 0.1624 data_time: 0.0113 memory: 7124 grad_norm: 7.1018 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2176 loss: 1.2176 2022/09/07 16:36:12 - mmengine - INFO - Epoch(train) [79][1900/3757] lr: 1.0000e-04 eta: 3:45:18 time: 0.1625 data_time: 0.0115 memory: 7124 grad_norm: 6.9493 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9828 loss: 0.9828 2022/09/07 16:36:21 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:36:28 - mmengine - INFO - Epoch(train) [79][2000/3757] lr: 1.0000e-04 eta: 3:45:01 time: 0.1603 data_time: 0.0119 memory: 7124 grad_norm: 6.9292 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0381 loss: 1.0381 2022/09/07 16:36:45 - mmengine - INFO - Epoch(train) [79][2100/3757] lr: 1.0000e-04 eta: 3:44:44 time: 0.1619 data_time: 0.0113 memory: 7124 grad_norm: 6.9929 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1543 loss: 1.1543 2022/09/07 16:37:02 - mmengine - INFO - Epoch(train) [79][2200/3757] lr: 1.0000e-04 eta: 3:44:27 time: 0.1616 data_time: 0.0111 memory: 7124 grad_norm: 6.8040 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1691 loss: 1.1691 2022/09/07 16:37:18 - mmengine - INFO - Epoch(train) [79][2300/3757] lr: 1.0000e-04 eta: 3:44:10 time: 0.1613 data_time: 0.0097 memory: 7124 grad_norm: 6.9429 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9089 loss: 0.9089 2022/09/07 16:37:35 - mmengine - INFO - Epoch(train) [79][2400/3757] lr: 1.0000e-04 eta: 3:43:54 time: 0.1672 data_time: 0.0116 memory: 7124 grad_norm: 7.2774 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0832 loss: 1.0832 2022/09/07 16:37:52 - mmengine - INFO - Epoch(train) [79][2500/3757] lr: 1.0000e-04 eta: 3:43:37 time: 0.1626 data_time: 0.0100 memory: 7124 grad_norm: 7.2534 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0330 loss: 1.0330 2022/09/07 16:38:08 - mmengine - INFO - Epoch(train) [79][2600/3757] lr: 1.0000e-04 eta: 3:43:20 time: 0.1642 data_time: 0.0114 memory: 7124 grad_norm: 6.9638 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0355 loss: 1.0355 2022/09/07 16:38:25 - mmengine - INFO - Epoch(train) [79][2700/3757] lr: 1.0000e-04 eta: 3:43:03 time: 0.1597 data_time: 0.0114 memory: 7124 grad_norm: 7.1100 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.0723 loss: 1.0723 2022/09/07 16:38:41 - mmengine - INFO - Epoch(train) [79][2800/3757] lr: 1.0000e-04 eta: 3:42:46 time: 0.1645 data_time: 0.0113 memory: 7124 grad_norm: 7.3791 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2489 loss: 1.2489 2022/09/07 16:38:58 - mmengine - INFO - Epoch(train) [79][2900/3757] lr: 1.0000e-04 eta: 3:42:30 time: 0.1720 data_time: 0.0102 memory: 7124 grad_norm: 7.0531 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 0.9882 loss: 0.9882 2022/09/07 16:39:07 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:39:15 - mmengine - INFO - Epoch(train) [79][3000/3757] lr: 1.0000e-04 eta: 3:42:13 time: 0.1671 data_time: 0.0118 memory: 7124 grad_norm: 7.1925 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.1472 loss: 1.1472 2022/09/07 16:39:31 - mmengine - INFO - Epoch(train) [79][3100/3757] lr: 1.0000e-04 eta: 3:41:56 time: 0.1657 data_time: 0.0113 memory: 7124 grad_norm: 7.1512 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1607 loss: 1.1607 2022/09/07 16:39:48 - mmengine - INFO - Epoch(train) [79][3200/3757] lr: 1.0000e-04 eta: 3:41:39 time: 0.1612 data_time: 0.0117 memory: 7124 grad_norm: 6.9932 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0614 loss: 1.0614 2022/09/07 16:40:04 - mmengine - INFO - Epoch(train) [79][3300/3757] lr: 1.0000e-04 eta: 3:41:22 time: 0.1602 data_time: 0.0118 memory: 7124 grad_norm: 6.8084 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1157 loss: 1.1157 2022/09/07 16:40:21 - mmengine - INFO - Epoch(train) [79][3400/3757] lr: 1.0000e-04 eta: 3:41:06 time: 0.1648 data_time: 0.0110 memory: 7124 grad_norm: 7.0403 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9260 loss: 0.9260 2022/09/07 16:40:38 - mmengine - INFO - Epoch(train) [79][3500/3757] lr: 1.0000e-04 eta: 3:40:49 time: 0.1743 data_time: 0.0105 memory: 7124 grad_norm: 7.1194 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1202 loss: 1.1202 2022/09/07 16:40:55 - mmengine - INFO - Epoch(train) [79][3600/3757] lr: 1.0000e-04 eta: 3:40:32 time: 0.1700 data_time: 0.0102 memory: 7124 grad_norm: 6.9466 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.7942 loss: 0.7942 2022/09/07 16:41:11 - mmengine - INFO - Epoch(train) [79][3700/3757] lr: 1.0000e-04 eta: 3:40:15 time: 0.1665 data_time: 0.0124 memory: 7124 grad_norm: 6.9507 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0016 loss: 1.0016 2022/09/07 16:41:20 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:41:20 - mmengine - INFO - Epoch(train) [79][3757/3757] lr: 1.0000e-04 eta: 3:40:09 time: 0.1408 data_time: 0.0080 memory: 7124 grad_norm: 6.9041 top1_acc: 0.8571 top5_acc: 1.0000 loss_cls: 0.9391 loss: 0.9391 2022/09/07 16:41:21 - mmengine - INFO - Saving checkpoint at 79 epochs 2022/09/07 16:43:45 - mmengine - INFO - Epoch(val) [79][100/310] eta: 0:04:19 time: 1.2378 data_time: 0.9295 memory: 7627 2022/09/07 16:46:04 - mmengine - INFO - Epoch(val) [79][200/310] eta: 0:02:13 time: 1.2173 data_time: 0.9073 memory: 7627 2022/09/07 16:48:12 - mmengine - INFO - Epoch(val) [79][300/310] eta: 0:00:12 time: 1.2464 data_time: 0.9433 memory: 7627 2022/09/07 16:48:28 - mmengine - INFO - Epoch(val) [79][310/310] acc/top1: 0.7521 acc/top5: 0.9176 acc/mean1: 0.7520 2022/09/07 16:48:46 - mmengine - INFO - Epoch(train) [80][100/3757] lr: 1.0000e-04 eta: 3:39:49 time: 0.1598 data_time: 0.0114 memory: 7627 grad_norm: 7.2556 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1280 loss: 1.1280 2022/09/07 16:49:02 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:49:03 - mmengine - INFO - Epoch(train) [80][200/3757] lr: 1.0000e-04 eta: 3:39:32 time: 0.1629 data_time: 0.0117 memory: 7124 grad_norm: 7.0077 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9916 loss: 0.9916 2022/09/07 16:49:19 - mmengine - INFO - Epoch(train) [80][300/3757] lr: 1.0000e-04 eta: 3:39:15 time: 0.1646 data_time: 0.0102 memory: 7124 grad_norm: 7.0586 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9791 loss: 0.9791 2022/09/07 16:49:36 - mmengine - INFO - Epoch(train) [80][400/3757] lr: 1.0000e-04 eta: 3:38:58 time: 0.1643 data_time: 0.0111 memory: 7124 grad_norm: 7.1475 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9870 loss: 0.9870 2022/09/07 16:49:52 - mmengine - INFO - Epoch(train) [80][500/3757] lr: 1.0000e-04 eta: 3:38:41 time: 0.1608 data_time: 0.0106 memory: 7124 grad_norm: 7.0917 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1075 loss: 1.1075 2022/09/07 16:50:09 - mmengine - INFO - Epoch(train) [80][600/3757] lr: 1.0000e-04 eta: 3:38:24 time: 0.1654 data_time: 0.0113 memory: 7124 grad_norm: 6.9074 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8188 loss: 0.8188 2022/09/07 16:50:25 - mmengine - INFO - Epoch(train) [80][700/3757] lr: 1.0000e-04 eta: 3:38:08 time: 0.1586 data_time: 0.0123 memory: 7124 grad_norm: 6.7772 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0586 loss: 1.0586 2022/09/07 16:50:41 - mmengine - INFO - Epoch(train) [80][800/3757] lr: 1.0000e-04 eta: 3:37:51 time: 0.1582 data_time: 0.0107 memory: 7124 grad_norm: 7.1769 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9766 loss: 0.9766 2022/09/07 16:50:58 - mmengine - INFO - Epoch(train) [80][900/3757] lr: 1.0000e-04 eta: 3:37:34 time: 0.1742 data_time: 0.0122 memory: 7124 grad_norm: 7.1152 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1477 loss: 1.1477 2022/09/07 16:51:14 - mmengine - INFO - Epoch(train) [80][1000/3757] lr: 1.0000e-04 eta: 3:37:17 time: 0.1656 data_time: 0.0115 memory: 7124 grad_norm: 6.9498 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1075 loss: 1.1075 2022/09/07 16:51:31 - mmengine - INFO - Epoch(train) [80][1100/3757] lr: 1.0000e-04 eta: 3:37:00 time: 0.1626 data_time: 0.0115 memory: 7124 grad_norm: 7.0002 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.8454 loss: 0.8454 2022/09/07 16:51:47 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:51:47 - mmengine - INFO - Epoch(train) [80][1200/3757] lr: 1.0000e-04 eta: 3:36:43 time: 0.1646 data_time: 0.0112 memory: 7124 grad_norm: 6.8903 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8103 loss: 0.8103 2022/09/07 16:52:04 - mmengine - INFO - Epoch(train) [80][1300/3757] lr: 1.0000e-04 eta: 3:36:26 time: 0.1630 data_time: 0.0110 memory: 7124 grad_norm: 6.8968 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9438 loss: 0.9438 2022/09/07 16:52:20 - mmengine - INFO - Epoch(train) [80][1400/3757] lr: 1.0000e-04 eta: 3:36:10 time: 0.1608 data_time: 0.0107 memory: 7124 grad_norm: 7.1386 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.0280 loss: 1.0280 2022/09/07 16:52:37 - mmengine - INFO - Epoch(train) [80][1500/3757] lr: 1.0000e-04 eta: 3:35:53 time: 0.1594 data_time: 0.0121 memory: 7124 grad_norm: 7.2180 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9880 loss: 0.9880 2022/09/07 16:52:53 - mmengine - INFO - Epoch(train) [80][1600/3757] lr: 1.0000e-04 eta: 3:35:36 time: 0.1651 data_time: 0.0113 memory: 7124 grad_norm: 6.7500 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8473 loss: 0.8473 2022/09/07 16:53:10 - mmengine - INFO - Epoch(train) [80][1700/3757] lr: 1.0000e-04 eta: 3:35:19 time: 0.1622 data_time: 0.0120 memory: 7124 grad_norm: 6.9627 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1805 loss: 1.1805 2022/09/07 16:53:26 - mmengine - INFO - Epoch(train) [80][1800/3757] lr: 1.0000e-04 eta: 3:35:02 time: 0.1618 data_time: 0.0114 memory: 7124 grad_norm: 7.1257 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.9674 loss: 0.9674 2022/09/07 16:53:43 - mmengine - INFO - Epoch(train) [80][1900/3757] lr: 1.0000e-04 eta: 3:34:45 time: 0.1621 data_time: 0.0113 memory: 7124 grad_norm: 6.9999 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0662 loss: 1.0662 2022/09/07 16:53:59 - mmengine - INFO - Epoch(train) [80][2000/3757] lr: 1.0000e-04 eta: 3:34:29 time: 0.1625 data_time: 0.0109 memory: 7124 grad_norm: 7.4018 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1490 loss: 1.1490 2022/09/07 16:54:16 - mmengine - INFO - Epoch(train) [80][2100/3757] lr: 1.0000e-04 eta: 3:34:12 time: 0.1640 data_time: 0.0121 memory: 7124 grad_norm: 7.1520 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1046 loss: 1.1046 2022/09/07 16:54:32 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:54:32 - mmengine - INFO - Epoch(train) [80][2200/3757] lr: 1.0000e-04 eta: 3:33:55 time: 0.1605 data_time: 0.0119 memory: 7124 grad_norm: 7.1657 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0781 loss: 1.0781 2022/09/07 16:54:48 - mmengine - INFO - Epoch(train) [80][2300/3757] lr: 1.0000e-04 eta: 3:33:38 time: 0.1607 data_time: 0.0096 memory: 7124 grad_norm: 6.7965 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9727 loss: 0.9727 2022/09/07 16:55:05 - mmengine - INFO - Epoch(train) [80][2400/3757] lr: 1.0000e-04 eta: 3:33:21 time: 0.1643 data_time: 0.0133 memory: 7124 grad_norm: 6.8643 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2537 loss: 1.2537 2022/09/07 16:55:21 - mmengine - INFO - Epoch(train) [80][2500/3757] lr: 1.0000e-04 eta: 3:33:04 time: 0.1628 data_time: 0.0106 memory: 7124 grad_norm: 6.8792 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9965 loss: 0.9965 2022/09/07 16:55:37 - mmengine - INFO - Epoch(train) [80][2600/3757] lr: 1.0000e-04 eta: 3:32:48 time: 0.1635 data_time: 0.0119 memory: 7124 grad_norm: 7.2498 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.9395 loss: 0.9395 2022/09/07 16:55:54 - mmengine - INFO - Epoch(train) [80][2700/3757] lr: 1.0000e-04 eta: 3:32:31 time: 0.1645 data_time: 0.0138 memory: 7124 grad_norm: 7.0426 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2658 loss: 1.2658 2022/09/07 16:56:10 - mmengine - INFO - Epoch(train) [80][2800/3757] lr: 1.0000e-04 eta: 3:32:14 time: 0.1605 data_time: 0.0109 memory: 7124 grad_norm: 7.0916 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2223 loss: 1.2223 2022/09/07 16:56:27 - mmengine - INFO - Epoch(train) [80][2900/3757] lr: 1.0000e-04 eta: 3:31:57 time: 0.1604 data_time: 0.0117 memory: 7124 grad_norm: 7.0910 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0731 loss: 1.0731 2022/09/07 16:56:43 - mmengine - INFO - Epoch(train) [80][3000/3757] lr: 1.0000e-04 eta: 3:31:40 time: 0.1624 data_time: 0.0105 memory: 7124 grad_norm: 7.0685 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1912 loss: 1.1912 2022/09/07 16:57:00 - mmengine - INFO - Epoch(train) [80][3100/3757] lr: 1.0000e-04 eta: 3:31:23 time: 0.1599 data_time: 0.0109 memory: 7124 grad_norm: 7.2779 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.7359 loss: 0.7359 2022/09/07 16:57:16 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:57:16 - mmengine - INFO - Epoch(train) [80][3200/3757] lr: 1.0000e-04 eta: 3:31:07 time: 0.1718 data_time: 0.0113 memory: 7124 grad_norm: 6.6672 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9942 loss: 0.9942 2022/09/07 16:57:33 - mmengine - INFO - Epoch(train) [80][3300/3757] lr: 1.0000e-04 eta: 3:30:50 time: 0.1608 data_time: 0.0107 memory: 7124 grad_norm: 7.3538 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1599 loss: 1.1599 2022/09/07 16:57:49 - mmengine - INFO - Epoch(train) [80][3400/3757] lr: 1.0000e-04 eta: 3:30:33 time: 0.1636 data_time: 0.0104 memory: 7124 grad_norm: 7.0515 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1411 loss: 1.1411 2022/09/07 16:58:06 - mmengine - INFO - Epoch(train) [80][3500/3757] lr: 1.0000e-04 eta: 3:30:16 time: 0.1647 data_time: 0.0110 memory: 7124 grad_norm: 7.1448 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 0.8939 loss: 0.8939 2022/09/07 16:58:22 - mmengine - INFO - Epoch(train) [80][3600/3757] lr: 1.0000e-04 eta: 3:29:59 time: 0.1636 data_time: 0.0111 memory: 7124 grad_norm: 6.6731 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0596 loss: 1.0596 2022/09/07 16:58:39 - mmengine - INFO - Epoch(train) [80][3700/3757] lr: 1.0000e-04 eta: 3:29:42 time: 0.1627 data_time: 0.0124 memory: 7124 grad_norm: 7.0287 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9154 loss: 0.9154 2022/09/07 16:58:48 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 16:58:48 - mmengine - INFO - Epoch(train) [80][3757/3757] lr: 1.0000e-04 eta: 3:29:36 time: 0.1387 data_time: 0.0071 memory: 7124 grad_norm: 6.8620 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9373 loss: 0.9373 2022/09/07 16:58:48 - mmengine - INFO - Saving checkpoint at 80 epochs 2022/09/07 17:01:06 - mmengine - INFO - Epoch(val) [80][100/310] eta: 0:03:44 time: 1.0687 data_time: 0.7634 memory: 7627 2022/09/07 17:03:28 - mmengine - INFO - Epoch(val) [80][200/310] eta: 0:02:32 time: 1.3889 data_time: 1.0848 memory: 7627 2022/09/07 17:05:35 - mmengine - INFO - Epoch(val) [80][300/310] eta: 0:00:11 time: 1.1029 data_time: 0.7997 memory: 7627 2022/09/07 17:05:52 - mmengine - INFO - Epoch(val) [80][310/310] acc/top1: 0.7518 acc/top5: 0.9167 acc/mean1: 0.7518 2022/09/07 17:06:10 - mmengine - INFO - Epoch(train) [81][100/3757] lr: 1.0000e-04 eta: 3:29:16 time: 0.1562 data_time: 0.0108 memory: 7627 grad_norm: 7.1000 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0642 loss: 1.0642 2022/09/07 17:06:26 - mmengine - INFO - Epoch(train) [81][200/3757] lr: 1.0000e-04 eta: 3:28:59 time: 0.1629 data_time: 0.0123 memory: 7124 grad_norm: 6.8907 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0974 loss: 1.0974 2022/09/07 17:06:43 - mmengine - INFO - Epoch(train) [81][300/3757] lr: 1.0000e-04 eta: 3:28:42 time: 0.1653 data_time: 0.0113 memory: 7124 grad_norm: 7.2840 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1450 loss: 1.1450 2022/09/07 17:06:59 - mmengine - INFO - Epoch(train) [81][400/3757] lr: 1.0000e-04 eta: 3:28:25 time: 0.1627 data_time: 0.0118 memory: 7124 grad_norm: 6.9707 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2785 loss: 1.2785 2022/09/07 17:07:06 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 17:07:16 - mmengine - INFO - Epoch(train) [81][500/3757] lr: 1.0000e-04 eta: 3:28:08 time: 0.1637 data_time: 0.0116 memory: 7124 grad_norm: 7.3322 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.2391 loss: 1.2391 2022/09/07 17:07:32 - mmengine - INFO - Epoch(train) [81][600/3757] lr: 1.0000e-04 eta: 3:27:52 time: 0.1604 data_time: 0.0122 memory: 7124 grad_norm: 7.1234 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.8312 loss: 0.8312 2022/09/07 17:07:49 - mmengine - INFO - Epoch(train) [81][700/3757] lr: 1.0000e-04 eta: 3:27:35 time: 0.1648 data_time: 0.0109 memory: 7124 grad_norm: 7.1020 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0878 loss: 1.0878 2022/09/07 17:08:06 - mmengine - INFO - Epoch(train) [81][800/3757] lr: 1.0000e-04 eta: 3:27:18 time: 0.1657 data_time: 0.0128 memory: 7124 grad_norm: 6.9220 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9997 loss: 0.9997 2022/09/07 17:08:22 - mmengine - INFO - Epoch(train) [81][900/3757] lr: 1.0000e-04 eta: 3:27:01 time: 0.1613 data_time: 0.0113 memory: 7124 grad_norm: 7.0801 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0091 loss: 1.0091 2022/09/07 17:08:38 - mmengine - INFO - Epoch(train) [81][1000/3757] lr: 1.0000e-04 eta: 3:26:44 time: 0.1610 data_time: 0.0109 memory: 7124 grad_norm: 7.1808 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9309 loss: 0.9309 2022/09/07 17:08:55 - mmengine - INFO - Epoch(train) [81][1100/3757] lr: 1.0000e-04 eta: 3:26:28 time: 0.1592 data_time: 0.0117 memory: 7124 grad_norm: 7.4118 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9510 loss: 0.9510 2022/09/07 17:09:11 - mmengine - INFO - Epoch(train) [81][1200/3757] lr: 1.0000e-04 eta: 3:26:11 time: 0.1588 data_time: 0.0113 memory: 7124 grad_norm: 7.2220 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0060 loss: 1.0060 2022/09/07 17:09:28 - mmengine - INFO - Epoch(train) [81][1300/3757] lr: 1.0000e-04 eta: 3:25:54 time: 0.1634 data_time: 0.0117 memory: 7124 grad_norm: 7.3620 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.8298 loss: 0.8298 2022/09/07 17:09:44 - mmengine - INFO - Epoch(train) [81][1400/3757] lr: 1.0000e-04 eta: 3:25:37 time: 0.1593 data_time: 0.0112 memory: 7124 grad_norm: 6.8416 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2255 loss: 1.2255 2022/09/07 17:09:51 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 17:10:00 - mmengine - INFO - Epoch(train) [81][1500/3757] lr: 1.0000e-04 eta: 3:25:20 time: 0.1579 data_time: 0.0111 memory: 7124 grad_norm: 7.1562 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1912 loss: 1.1912 2022/09/07 17:10:17 - mmengine - INFO - Epoch(train) [81][1600/3757] lr: 1.0000e-04 eta: 3:25:03 time: 0.1707 data_time: 0.0118 memory: 7124 grad_norm: 7.0778 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0545 loss: 1.0545 2022/09/07 17:10:33 - mmengine - INFO - Epoch(train) [81][1700/3757] lr: 1.0000e-04 eta: 3:24:46 time: 0.1583 data_time: 0.0108 memory: 7124 grad_norm: 7.0986 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9313 loss: 0.9313 2022/09/07 17:10:50 - mmengine - INFO - Epoch(train) [81][1800/3757] lr: 1.0000e-04 eta: 3:24:30 time: 0.1627 data_time: 0.0123 memory: 7124 grad_norm: 7.0481 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1042 loss: 1.1042 2022/09/07 17:11:06 - mmengine - INFO - Epoch(train) [81][1900/3757] lr: 1.0000e-04 eta: 3:24:13 time: 0.1623 data_time: 0.0109 memory: 7124 grad_norm: 6.9086 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9861 loss: 0.9861 2022/09/07 17:11:23 - mmengine - INFO - Epoch(train) [81][2000/3757] lr: 1.0000e-04 eta: 3:23:56 time: 0.1621 data_time: 0.0107 memory: 7124 grad_norm: 7.1189 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0256 loss: 1.0256 2022/09/07 17:11:39 - mmengine - INFO - Epoch(train) [81][2100/3757] lr: 1.0000e-04 eta: 3:23:39 time: 0.1645 data_time: 0.0123 memory: 7124 grad_norm: 7.0028 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9536 loss: 0.9536 2022/09/07 17:11:56 - mmengine - INFO - Epoch(train) [81][2200/3757] lr: 1.0000e-04 eta: 3:23:22 time: 0.1773 data_time: 0.0104 memory: 7124 grad_norm: 6.8811 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.0905 loss: 1.0905 2022/09/07 17:12:12 - mmengine - INFO - Epoch(train) [81][2300/3757] lr: 1.0000e-04 eta: 3:23:06 time: 0.1603 data_time: 0.0111 memory: 7124 grad_norm: 7.4871 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9897 loss: 0.9897 2022/09/07 17:12:28 - mmengine - INFO - Epoch(train) [81][2400/3757] lr: 1.0000e-04 eta: 3:22:49 time: 0.1671 data_time: 0.0105 memory: 7124 grad_norm: 6.8039 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0543 loss: 1.0543 2022/09/07 17:12:35 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 17:12:45 - mmengine - INFO - Epoch(train) [81][2500/3757] lr: 1.0000e-04 eta: 3:22:32 time: 0.1649 data_time: 0.0097 memory: 7124 grad_norm: 7.0365 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1492 loss: 1.1492 2022/09/07 17:13:01 - mmengine - INFO - Epoch(train) [81][2600/3757] lr: 1.0000e-04 eta: 3:22:15 time: 0.1607 data_time: 0.0100 memory: 7124 grad_norm: 7.1455 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1898 loss: 1.1898 2022/09/07 17:13:17 - mmengine - INFO - Epoch(train) [81][2700/3757] lr: 1.0000e-04 eta: 3:21:58 time: 0.1649 data_time: 0.0125 memory: 7124 grad_norm: 7.0577 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1011 loss: 1.1011 2022/09/07 17:13:34 - mmengine - INFO - Epoch(train) [81][2800/3757] lr: 1.0000e-04 eta: 3:21:41 time: 0.1607 data_time: 0.0113 memory: 7124 grad_norm: 7.2654 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1277 loss: 1.1277 2022/09/07 17:13:50 - mmengine - INFO - Epoch(train) [81][2900/3757] lr: 1.0000e-04 eta: 3:21:25 time: 0.1632 data_time: 0.0116 memory: 7124 grad_norm: 7.0989 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1669 loss: 1.1669 2022/09/07 17:14:07 - mmengine - INFO - Epoch(train) [81][3000/3757] lr: 1.0000e-04 eta: 3:21:08 time: 0.1652 data_time: 0.0128 memory: 7124 grad_norm: 6.9831 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.0747 loss: 1.0747 2022/09/07 17:14:23 - mmengine - INFO - Epoch(train) [81][3100/3757] lr: 1.0000e-04 eta: 3:20:51 time: 0.1617 data_time: 0.0112 memory: 7124 grad_norm: 7.3107 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0759 loss: 1.0759 2022/09/07 17:14:39 - mmengine - INFO - Epoch(train) [81][3200/3757] lr: 1.0000e-04 eta: 3:20:34 time: 0.1605 data_time: 0.0124 memory: 7124 grad_norm: 7.1220 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2913 loss: 1.2913 2022/09/07 17:14:56 - mmengine - INFO - Epoch(train) [81][3300/3757] lr: 1.0000e-04 eta: 3:20:17 time: 0.1613 data_time: 0.0100 memory: 7124 grad_norm: 7.0106 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1714 loss: 1.1714 2022/09/07 17:15:13 - mmengine - INFO - Epoch(train) [81][3400/3757] lr: 1.0000e-04 eta: 3:20:00 time: 0.1587 data_time: 0.0113 memory: 7124 grad_norm: 7.0870 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1277 loss: 1.1277 2022/09/07 17:15:19 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 17:15:29 - mmengine - INFO - Epoch(train) [81][3500/3757] lr: 1.0000e-04 eta: 3:19:44 time: 0.1604 data_time: 0.0117 memory: 7124 grad_norm: 7.0232 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8838 loss: 0.8838 2022/09/07 17:15:46 - mmengine - INFO - Epoch(train) [81][3600/3757] lr: 1.0000e-04 eta: 3:19:27 time: 0.1646 data_time: 0.0107 memory: 7124 grad_norm: 7.2604 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9927 loss: 0.9927 2022/09/07 17:16:02 - mmengine - INFO - Epoch(train) [81][3700/3757] lr: 1.0000e-04 eta: 3:19:10 time: 0.1653 data_time: 0.0116 memory: 7124 grad_norm: 7.5052 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0380 loss: 1.0380 2022/09/07 17:16:11 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 17:16:11 - mmengine - INFO - Epoch(train) [81][3757/3757] lr: 1.0000e-04 eta: 3:19:03 time: 0.1404 data_time: 0.0075 memory: 7124 grad_norm: 7.2049 top1_acc: 0.7143 top5_acc: 0.7143 loss_cls: 1.1295 loss: 1.1295 2022/09/07 17:16:11 - mmengine - INFO - Saving checkpoint at 81 epochs 2022/09/07 17:18:32 - mmengine - INFO - Epoch(val) [81][100/310] eta: 0:03:57 time: 1.1291 data_time: 0.8269 memory: 7627 2022/09/07 17:20:54 - mmengine - INFO - Epoch(val) [81][200/310] eta: 0:02:33 time: 1.3937 data_time: 1.0892 memory: 7627 2022/09/07 17:22:59 - mmengine - INFO - Epoch(val) [81][300/310] eta: 0:00:11 time: 1.1225 data_time: 0.8205 memory: 7627 2022/09/07 17:23:15 - mmengine - INFO - Epoch(val) [81][310/310] acc/top1: 0.7524 acc/top5: 0.9183 acc/mean1: 0.7524 2022/09/07 17:23:33 - mmengine - INFO - Epoch(train) [82][100/3757] lr: 1.0000e-04 eta: 3:18:43 time: 0.1564 data_time: 0.0099 memory: 7627 grad_norm: 7.2358 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1885 loss: 1.1885 2022/09/07 17:23:49 - mmengine - INFO - Epoch(train) [82][200/3757] lr: 1.0000e-04 eta: 3:18:26 time: 0.1619 data_time: 0.0114 memory: 7124 grad_norm: 7.0153 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1333 loss: 1.1333 2022/09/07 17:24:05 - mmengine - INFO - Epoch(train) [82][300/3757] lr: 1.0000e-04 eta: 3:18:10 time: 0.1630 data_time: 0.0109 memory: 7124 grad_norm: 6.7366 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8426 loss: 0.8426 2022/09/07 17:24:22 - mmengine - INFO - Epoch(train) [82][400/3757] lr: 1.0000e-04 eta: 3:17:53 time: 0.1617 data_time: 0.0121 memory: 7124 grad_norm: 7.0850 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.1695 loss: 1.1695 2022/09/07 17:24:38 - mmengine - INFO - Epoch(train) [82][500/3757] lr: 1.0000e-04 eta: 3:17:36 time: 0.1639 data_time: 0.0117 memory: 7124 grad_norm: 7.0651 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9765 loss: 0.9765 2022/09/07 17:24:54 - mmengine - INFO - Epoch(train) [82][600/3757] lr: 1.0000e-04 eta: 3:17:19 time: 0.1591 data_time: 0.0110 memory: 7124 grad_norm: 6.9279 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.1432 loss: 1.1432 2022/09/07 17:25:08 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 17:25:11 - mmengine - INFO - Epoch(train) [82][700/3757] lr: 1.0000e-04 eta: 3:17:02 time: 0.1625 data_time: 0.0109 memory: 7124 grad_norm: 7.1860 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2027 loss: 1.2027 2022/09/07 17:25:27 - mmengine - INFO - Epoch(train) [82][800/3757] lr: 1.0000e-04 eta: 3:16:45 time: 0.1634 data_time: 0.0116 memory: 7124 grad_norm: 7.1385 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.9459 loss: 0.9459 2022/09/07 17:25:43 - mmengine - INFO - Epoch(train) [82][900/3757] lr: 1.0000e-04 eta: 3:16:28 time: 0.1618 data_time: 0.0117 memory: 7124 grad_norm: 7.0477 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.1864 loss: 1.1864 2022/09/07 17:26:00 - mmengine - INFO - Epoch(train) [82][1000/3757] lr: 1.0000e-04 eta: 3:16:12 time: 0.1625 data_time: 0.0119 memory: 7124 grad_norm: 6.7976 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9490 loss: 0.9490 2022/09/07 17:26:16 - mmengine - INFO - Epoch(train) [82][1100/3757] lr: 1.0000e-04 eta: 3:15:55 time: 0.1626 data_time: 0.0129 memory: 7124 grad_norm: 7.3207 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2709 loss: 1.2709 2022/09/07 17:26:32 - mmengine - INFO - Epoch(train) [82][1200/3757] lr: 1.0000e-04 eta: 3:15:38 time: 0.1611 data_time: 0.0134 memory: 7124 grad_norm: 7.0759 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2125 loss: 1.2125 2022/09/07 17:26:49 - mmengine - INFO - Epoch(train) [82][1300/3757] lr: 1.0000e-04 eta: 3:15:21 time: 0.1615 data_time: 0.0111 memory: 7124 grad_norm: 6.7587 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8354 loss: 0.8354 2022/09/07 17:27:05 - mmengine - INFO - Epoch(train) [82][1400/3757] lr: 1.0000e-04 eta: 3:15:04 time: 0.1674 data_time: 0.0119 memory: 7124 grad_norm: 7.0943 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1465 loss: 1.1465 2022/09/07 17:27:21 - mmengine - INFO - Epoch(train) [82][1500/3757] lr: 1.0000e-04 eta: 3:14:47 time: 0.1663 data_time: 0.0113 memory: 7124 grad_norm: 7.1676 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0740 loss: 1.0740 2022/09/07 17:27:38 - mmengine - INFO - Epoch(train) [82][1600/3757] lr: 1.0000e-04 eta: 3:14:31 time: 0.1643 data_time: 0.0121 memory: 7124 grad_norm: 7.1657 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9761 loss: 0.9761 2022/09/07 17:27:51 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 17:27:54 - mmengine - INFO - Epoch(train) [82][1700/3757] lr: 1.0000e-04 eta: 3:14:14 time: 0.1634 data_time: 0.0108 memory: 7124 grad_norm: 7.1958 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.3433 loss: 1.3433 2022/09/07 17:28:11 - mmengine - INFO - Epoch(train) [82][1800/3757] lr: 1.0000e-04 eta: 3:13:57 time: 0.1631 data_time: 0.0112 memory: 7124 grad_norm: 6.9591 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2991 loss: 1.2991 2022/09/07 17:28:27 - mmengine - INFO - Epoch(train) [82][1900/3757] lr: 1.0000e-04 eta: 3:13:40 time: 0.1658 data_time: 0.0109 memory: 7124 grad_norm: 7.0908 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.3509 loss: 1.3509 2022/09/07 17:28:44 - mmengine - INFO - Epoch(train) [82][2000/3757] lr: 1.0000e-04 eta: 3:13:23 time: 0.1611 data_time: 0.0119 memory: 7124 grad_norm: 7.0484 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0135 loss: 1.0135 2022/09/07 17:29:00 - mmengine - INFO - Epoch(train) [82][2100/3757] lr: 1.0000e-04 eta: 3:13:07 time: 0.1700 data_time: 0.0117 memory: 7124 grad_norm: 7.4998 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0845 loss: 1.0845 2022/09/07 17:29:17 - mmengine - INFO - Epoch(train) [82][2200/3757] lr: 1.0000e-04 eta: 3:12:50 time: 0.1620 data_time: 0.0119 memory: 7124 grad_norm: 7.1523 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1356 loss: 1.1356 2022/09/07 17:29:33 - mmengine - INFO - Epoch(train) [82][2300/3757] lr: 1.0000e-04 eta: 3:12:33 time: 0.1530 data_time: 0.0104 memory: 7124 grad_norm: 7.3394 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1570 loss: 1.1570 2022/09/07 17:29:50 - mmengine - INFO - Epoch(train) [82][2400/3757] lr: 1.0000e-04 eta: 3:12:16 time: 0.1664 data_time: 0.0121 memory: 7124 grad_norm: 7.3597 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.3216 loss: 1.3216 2022/09/07 17:30:06 - mmengine - INFO - Epoch(train) [82][2500/3757] lr: 1.0000e-04 eta: 3:11:59 time: 0.1631 data_time: 0.0115 memory: 7124 grad_norm: 7.1819 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0586 loss: 1.0586 2022/09/07 17:30:23 - mmengine - INFO - Epoch(train) [82][2600/3757] lr: 1.0000e-04 eta: 3:11:43 time: 0.1637 data_time: 0.0116 memory: 7124 grad_norm: 7.3333 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1047 loss: 1.1047 2022/09/07 17:30:36 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 17:30:39 - mmengine - INFO - Epoch(train) [82][2700/3757] lr: 1.0000e-04 eta: 3:11:26 time: 0.1674 data_time: 0.0122 memory: 7124 grad_norm: 6.8862 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8385 loss: 0.8385 2022/09/07 17:30:56 - mmengine - INFO - Epoch(train) [82][2800/3757] lr: 1.0000e-04 eta: 3:11:09 time: 0.1607 data_time: 0.0104 memory: 7124 grad_norm: 6.9152 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1084 loss: 1.1084 2022/09/07 17:31:12 - mmengine - INFO - Epoch(train) [82][2900/3757] lr: 1.0000e-04 eta: 3:10:52 time: 0.1638 data_time: 0.0111 memory: 7124 grad_norm: 7.1445 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0935 loss: 1.0935 2022/09/07 17:31:28 - mmengine - INFO - Epoch(train) [82][3000/3757] lr: 1.0000e-04 eta: 3:10:35 time: 0.1592 data_time: 0.0108 memory: 7124 grad_norm: 7.0935 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.8775 loss: 0.8775 2022/09/07 17:31:45 - mmengine - INFO - Epoch(train) [82][3100/3757] lr: 1.0000e-04 eta: 3:10:19 time: 0.1660 data_time: 0.0120 memory: 7124 grad_norm: 6.9969 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0950 loss: 1.0950 2022/09/07 17:32:01 - mmengine - INFO - Epoch(train) [82][3200/3757] lr: 1.0000e-04 eta: 3:10:02 time: 0.1604 data_time: 0.0119 memory: 7124 grad_norm: 6.7054 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9846 loss: 0.9846 2022/09/07 17:32:18 - mmengine - INFO - Epoch(train) [82][3300/3757] lr: 1.0000e-04 eta: 3:09:45 time: 0.1611 data_time: 0.0107 memory: 7124 grad_norm: 7.0434 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0024 loss: 1.0024 2022/09/07 17:32:34 - mmengine - INFO - Epoch(train) [82][3400/3757] lr: 1.0000e-04 eta: 3:09:28 time: 0.1643 data_time: 0.0105 memory: 7124 grad_norm: 6.9327 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1215 loss: 1.1215 2022/09/07 17:32:50 - mmengine - INFO - Epoch(train) [82][3500/3757] lr: 1.0000e-04 eta: 3:09:11 time: 0.1606 data_time: 0.0125 memory: 7124 grad_norm: 6.9220 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9051 loss: 0.9051 2022/09/07 17:33:07 - mmengine - INFO - Epoch(train) [82][3600/3757] lr: 1.0000e-04 eta: 3:08:54 time: 0.1634 data_time: 0.0114 memory: 7124 grad_norm: 7.0566 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9121 loss: 0.9121 2022/09/07 17:33:21 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 17:33:24 - mmengine - INFO - Epoch(train) [82][3700/3757] lr: 1.0000e-04 eta: 3:08:38 time: 0.1651 data_time: 0.0120 memory: 7124 grad_norm: 7.2996 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 0.9725 loss: 0.9725 2022/09/07 17:33:32 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 17:33:32 - mmengine - INFO - Epoch(train) [82][3757/3757] lr: 1.0000e-04 eta: 3:08:31 time: 0.1380 data_time: 0.0073 memory: 7124 grad_norm: 6.9265 top1_acc: 0.5714 top5_acc: 0.8571 loss_cls: 0.9355 loss: 0.9355 2022/09/07 17:33:32 - mmengine - INFO - Saving checkpoint at 82 epochs 2022/09/07 17:35:54 - mmengine - INFO - Epoch(val) [82][100/310] eta: 0:04:38 time: 1.3275 data_time: 1.0246 memory: 7627 2022/09/07 17:38:09 - mmengine - INFO - Epoch(val) [82][200/310] eta: 0:02:02 time: 1.1176 data_time: 0.8154 memory: 7627 2022/09/07 17:40:19 - mmengine - INFO - Epoch(val) [82][300/310] eta: 0:00:12 time: 1.2752 data_time: 0.9710 memory: 7627 2022/09/07 17:40:40 - mmengine - INFO - Epoch(val) [82][310/310] acc/top1: 0.7514 acc/top5: 0.9185 acc/mean1: 0.7513 2022/09/07 17:40:58 - mmengine - INFO - Epoch(train) [83][100/3757] lr: 1.0000e-04 eta: 3:08:11 time: 0.1659 data_time: 0.0115 memory: 7627 grad_norm: 6.9952 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9599 loss: 0.9599 2022/09/07 17:41:14 - mmengine - INFO - Epoch(train) [83][200/3757] lr: 1.0000e-04 eta: 3:07:54 time: 0.1617 data_time: 0.0120 memory: 7124 grad_norm: 7.0532 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1974 loss: 1.1974 2022/09/07 17:41:30 - mmengine - INFO - Epoch(train) [83][300/3757] lr: 1.0000e-04 eta: 3:07:37 time: 0.1634 data_time: 0.0112 memory: 7124 grad_norm: 6.9716 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8717 loss: 0.8717 2022/09/07 17:41:47 - mmengine - INFO - Epoch(train) [83][400/3757] lr: 1.0000e-04 eta: 3:07:20 time: 0.1602 data_time: 0.0110 memory: 7124 grad_norm: 6.9715 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1735 loss: 1.1735 2022/09/07 17:42:03 - mmengine - INFO - Epoch(train) [83][500/3757] lr: 1.0000e-04 eta: 3:07:04 time: 0.1623 data_time: 0.0111 memory: 7124 grad_norm: 6.9999 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.1160 loss: 1.1160 2022/09/07 17:42:20 - mmengine - INFO - Epoch(train) [83][600/3757] lr: 1.0000e-04 eta: 3:06:47 time: 0.1670 data_time: 0.0117 memory: 7124 grad_norm: 7.0237 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.7716 loss: 0.7716 2022/09/07 17:42:36 - mmengine - INFO - Epoch(train) [83][700/3757] lr: 1.0000e-04 eta: 3:06:30 time: 0.1633 data_time: 0.0115 memory: 7124 grad_norm: 7.1773 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1284 loss: 1.1284 2022/09/07 17:42:52 - mmengine - INFO - Epoch(train) [83][800/3757] lr: 1.0000e-04 eta: 3:06:13 time: 0.1590 data_time: 0.0105 memory: 7124 grad_norm: 7.5335 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0189 loss: 1.0189 2022/09/07 17:43:09 - mmengine - INFO - Epoch(train) [83][900/3757] lr: 1.0000e-04 eta: 3:05:56 time: 0.1630 data_time: 0.0109 memory: 7124 grad_norm: 7.2545 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.9846 loss: 0.9846 2022/09/07 17:43:13 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 17:43:25 - mmengine - INFO - Epoch(train) [83][1000/3757] lr: 1.0000e-04 eta: 3:05:40 time: 0.1646 data_time: 0.0118 memory: 7124 grad_norm: 7.1483 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0834 loss: 1.0834 2022/09/07 17:43:41 - mmengine - INFO - Epoch(train) [83][1100/3757] lr: 1.0000e-04 eta: 3:05:23 time: 0.1651 data_time: 0.0114 memory: 7124 grad_norm: 7.1300 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0609 loss: 1.0609 2022/09/07 17:43:58 - mmengine - INFO - Epoch(train) [83][1200/3757] lr: 1.0000e-04 eta: 3:05:06 time: 0.1596 data_time: 0.0103 memory: 7124 grad_norm: 7.1812 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9482 loss: 0.9482 2022/09/07 17:44:15 - mmengine - INFO - Epoch(train) [83][1300/3757] lr: 1.0000e-04 eta: 3:04:49 time: 0.1661 data_time: 0.0127 memory: 7124 grad_norm: 6.8624 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1197 loss: 1.1197 2022/09/07 17:44:31 - mmengine - INFO - Epoch(train) [83][1400/3757] lr: 1.0000e-04 eta: 3:04:32 time: 0.1636 data_time: 0.0109 memory: 7124 grad_norm: 7.3061 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9368 loss: 0.9368 2022/09/07 17:44:48 - mmengine - INFO - Epoch(train) [83][1500/3757] lr: 1.0000e-04 eta: 3:04:16 time: 0.1646 data_time: 0.0110 memory: 7124 grad_norm: 7.0890 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.9921 loss: 0.9921 2022/09/07 17:45:05 - mmengine - INFO - Epoch(train) [83][1600/3757] lr: 1.0000e-04 eta: 3:03:59 time: 0.1620 data_time: 0.0118 memory: 7124 grad_norm: 7.2909 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9463 loss: 0.9463 2022/09/07 17:45:22 - mmengine - INFO - Epoch(train) [83][1700/3757] lr: 1.0000e-04 eta: 3:03:42 time: 0.1628 data_time: 0.0108 memory: 7124 grad_norm: 7.0761 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.0654 loss: 1.0654 2022/09/07 17:45:38 - mmengine - INFO - Epoch(train) [83][1800/3757] lr: 1.0000e-04 eta: 3:03:25 time: 0.1640 data_time: 0.0114 memory: 7124 grad_norm: 7.1800 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0190 loss: 1.0190 2022/09/07 17:45:55 - mmengine - INFO - Epoch(train) [83][1900/3757] lr: 1.0000e-04 eta: 3:03:09 time: 0.1622 data_time: 0.0114 memory: 7124 grad_norm: 7.3994 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2723 loss: 1.2723 2022/09/07 17:45:59 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 17:46:11 - mmengine - INFO - Epoch(train) [83][2000/3757] lr: 1.0000e-04 eta: 3:02:52 time: 0.1648 data_time: 0.0118 memory: 7124 grad_norm: 6.9064 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9233 loss: 0.9233 2022/09/07 17:46:28 - mmengine - INFO - Epoch(train) [83][2100/3757] lr: 1.0000e-04 eta: 3:02:35 time: 0.1663 data_time: 0.0138 memory: 7124 grad_norm: 7.0451 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9978 loss: 0.9978 2022/09/07 17:46:44 - mmengine - INFO - Epoch(train) [83][2200/3757] lr: 1.0000e-04 eta: 3:02:18 time: 0.1655 data_time: 0.0103 memory: 7124 grad_norm: 7.3004 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.3011 loss: 1.3011 2022/09/07 17:47:01 - mmengine - INFO - Epoch(train) [83][2300/3757] lr: 1.0000e-04 eta: 3:02:02 time: 0.1625 data_time: 0.0115 memory: 7124 grad_norm: 6.8072 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8862 loss: 0.8862 2022/09/07 17:47:17 - mmengine - INFO - Epoch(train) [83][2400/3757] lr: 1.0000e-04 eta: 3:01:45 time: 0.1629 data_time: 0.0119 memory: 7124 grad_norm: 7.1402 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9721 loss: 0.9721 2022/09/07 17:47:34 - mmengine - INFO - Epoch(train) [83][2500/3757] lr: 1.0000e-04 eta: 3:01:28 time: 0.1637 data_time: 0.0115 memory: 7124 grad_norm: 6.9788 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.9939 loss: 0.9939 2022/09/07 17:47:51 - mmengine - INFO - Epoch(train) [83][2600/3757] lr: 1.0000e-04 eta: 3:01:11 time: 0.1749 data_time: 0.0109 memory: 7124 grad_norm: 7.2294 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0929 loss: 1.0929 2022/09/07 17:48:07 - mmengine - INFO - Epoch(train) [83][2700/3757] lr: 1.0000e-04 eta: 3:00:55 time: 0.1669 data_time: 0.0122 memory: 7124 grad_norm: 7.2362 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0936 loss: 1.0936 2022/09/07 17:48:24 - mmengine - INFO - Epoch(train) [83][2800/3757] lr: 1.0000e-04 eta: 3:00:38 time: 0.1614 data_time: 0.0111 memory: 7124 grad_norm: 7.3043 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2088 loss: 1.2088 2022/09/07 17:48:41 - mmengine - INFO - Epoch(train) [83][2900/3757] lr: 1.0000e-04 eta: 3:00:21 time: 0.1625 data_time: 0.0125 memory: 7124 grad_norm: 7.2049 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8911 loss: 0.8911 2022/09/07 17:48:45 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 17:48:57 - mmengine - INFO - Epoch(train) [83][3000/3757] lr: 1.0000e-04 eta: 3:00:04 time: 0.1632 data_time: 0.0115 memory: 7124 grad_norm: 7.1136 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1923 loss: 1.1923 2022/09/07 17:49:14 - mmengine - INFO - Epoch(train) [83][3100/3757] lr: 1.0000e-04 eta: 2:59:47 time: 0.1634 data_time: 0.0116 memory: 7124 grad_norm: 7.3275 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0361 loss: 1.0361 2022/09/07 17:49:30 - mmengine - INFO - Epoch(train) [83][3200/3757] lr: 1.0000e-04 eta: 2:59:31 time: 0.1684 data_time: 0.0108 memory: 7124 grad_norm: 7.3117 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8355 loss: 0.8355 2022/09/07 17:49:47 - mmengine - INFO - Epoch(train) [83][3300/3757] lr: 1.0000e-04 eta: 2:59:14 time: 0.1593 data_time: 0.0114 memory: 7124 grad_norm: 7.0771 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9732 loss: 0.9732 2022/09/07 17:50:03 - mmengine - INFO - Epoch(train) [83][3400/3757] lr: 1.0000e-04 eta: 2:58:57 time: 0.1682 data_time: 0.0104 memory: 7124 grad_norm: 7.6045 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8945 loss: 0.8945 2022/09/07 17:50:20 - mmengine - INFO - Epoch(train) [83][3500/3757] lr: 1.0000e-04 eta: 2:58:40 time: 0.1710 data_time: 0.0113 memory: 7124 grad_norm: 6.8232 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0050 loss: 1.0050 2022/09/07 17:50:37 - mmengine - INFO - Epoch(train) [83][3600/3757] lr: 1.0000e-04 eta: 2:58:24 time: 0.1676 data_time: 0.0127 memory: 7124 grad_norm: 6.9592 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1997 loss: 1.1997 2022/09/07 17:50:53 - mmengine - INFO - Epoch(train) [83][3700/3757] lr: 1.0000e-04 eta: 2:58:07 time: 0.1610 data_time: 0.0105 memory: 7124 grad_norm: 7.1752 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.9402 loss: 0.9402 2022/09/07 17:51:03 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 17:51:03 - mmengine - INFO - Epoch(train) [83][3757/3757] lr: 1.0000e-04 eta: 2:58:00 time: 0.1404 data_time: 0.0084 memory: 7124 grad_norm: 7.3924 top1_acc: 0.4286 top5_acc: 0.5714 loss_cls: 1.3161 loss: 1.3161 2022/09/07 17:51:03 - mmengine - INFO - Saving checkpoint at 83 epochs 2022/09/07 17:53:26 - mmengine - INFO - Epoch(val) [83][100/310] eta: 0:04:03 time: 1.1591 data_time: 0.8545 memory: 7627 2022/09/07 17:55:50 - mmengine - INFO - Epoch(val) [83][200/310] eta: 0:02:36 time: 1.4214 data_time: 1.1167 memory: 7627 2022/09/07 17:57:57 - mmengine - INFO - Epoch(val) [83][300/310] eta: 0:00:11 time: 1.1462 data_time: 0.8465 memory: 7627 2022/09/07 17:58:14 - mmengine - INFO - Epoch(val) [83][310/310] acc/top1: 0.7526 acc/top5: 0.9179 acc/mean1: 0.7525 2022/09/07 17:58:33 - mmengine - INFO - Epoch(train) [84][100/3757] lr: 1.0000e-04 eta: 2:57:40 time: 0.1675 data_time: 0.0108 memory: 7627 grad_norm: 7.0224 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1031 loss: 1.1031 2022/09/07 17:58:44 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 17:58:49 - mmengine - INFO - Epoch(train) [84][200/3757] lr: 1.0000e-04 eta: 2:57:23 time: 0.1634 data_time: 0.0112 memory: 7124 grad_norm: 7.0123 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9963 loss: 0.9963 2022/09/07 17:59:05 - mmengine - INFO - Epoch(train) [84][300/3757] lr: 1.0000e-04 eta: 2:57:07 time: 0.1622 data_time: 0.0120 memory: 7124 grad_norm: 7.2680 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1488 loss: 1.1488 2022/09/07 17:59:22 - mmengine - INFO - Epoch(train) [84][400/3757] lr: 1.0000e-04 eta: 2:56:50 time: 0.1700 data_time: 0.0139 memory: 7124 grad_norm: 7.0971 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9276 loss: 0.9276 2022/09/07 17:59:38 - mmengine - INFO - Epoch(train) [84][500/3757] lr: 1.0000e-04 eta: 2:56:33 time: 0.1601 data_time: 0.0105 memory: 7124 grad_norm: 7.2346 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8780 loss: 0.8780 2022/09/07 17:59:55 - mmengine - INFO - Epoch(train) [84][600/3757] lr: 1.0000e-04 eta: 2:56:16 time: 0.1585 data_time: 0.0105 memory: 7124 grad_norm: 7.1921 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.7834 loss: 0.7834 2022/09/07 18:00:15 - mmengine - INFO - Epoch(train) [84][700/3757] lr: 1.0000e-04 eta: 2:56:00 time: 0.1621 data_time: 0.0121 memory: 7124 grad_norm: 7.1544 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.9350 loss: 0.9350 2022/09/07 18:00:31 - mmengine - INFO - Epoch(train) [84][800/3757] lr: 1.0000e-04 eta: 2:55:43 time: 0.1640 data_time: 0.0114 memory: 7124 grad_norm: 7.3811 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0284 loss: 1.0284 2022/09/07 18:00:47 - mmengine - INFO - Epoch(train) [84][900/3757] lr: 1.0000e-04 eta: 2:55:27 time: 0.1588 data_time: 0.0109 memory: 7124 grad_norm: 7.1078 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0977 loss: 1.0977 2022/09/07 18:01:03 - mmengine - INFO - Epoch(train) [84][1000/3757] lr: 1.0000e-04 eta: 2:55:10 time: 0.1610 data_time: 0.0120 memory: 7124 grad_norm: 7.4189 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.2057 loss: 1.2057 2022/09/07 18:01:20 - mmengine - INFO - Epoch(train) [84][1100/3757] lr: 1.0000e-04 eta: 2:54:53 time: 0.1606 data_time: 0.0116 memory: 7124 grad_norm: 7.2574 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3172 loss: 1.3172 2022/09/07 18:01:31 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:01:36 - mmengine - INFO - Epoch(train) [84][1200/3757] lr: 1.0000e-04 eta: 2:54:36 time: 0.1579 data_time: 0.0122 memory: 7124 grad_norm: 6.9777 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9775 loss: 0.9775 2022/09/07 18:01:52 - mmengine - INFO - Epoch(train) [84][1300/3757] lr: 1.0000e-04 eta: 2:54:19 time: 0.1669 data_time: 0.0114 memory: 7124 grad_norm: 7.2800 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2043 loss: 1.2043 2022/09/07 18:02:08 - mmengine - INFO - Epoch(train) [84][1400/3757] lr: 1.0000e-04 eta: 2:54:02 time: 0.1573 data_time: 0.0106 memory: 7124 grad_norm: 7.0696 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.3086 loss: 1.3086 2022/09/07 18:02:24 - mmengine - INFO - Epoch(train) [84][1500/3757] lr: 1.0000e-04 eta: 2:53:45 time: 0.1583 data_time: 0.0113 memory: 7124 grad_norm: 6.9851 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.8672 loss: 0.8672 2022/09/07 18:02:40 - mmengine - INFO - Epoch(train) [84][1600/3757] lr: 1.0000e-04 eta: 2:53:29 time: 0.1599 data_time: 0.0107 memory: 7124 grad_norm: 7.1200 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.2273 loss: 1.2273 2022/09/07 18:02:57 - mmengine - INFO - Epoch(train) [84][1700/3757] lr: 1.0000e-04 eta: 2:53:12 time: 0.1545 data_time: 0.0092 memory: 7124 grad_norm: 7.2119 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0994 loss: 1.0994 2022/09/07 18:03:13 - mmengine - INFO - Epoch(train) [84][1800/3757] lr: 1.0000e-04 eta: 2:52:55 time: 0.1646 data_time: 0.0111 memory: 7124 grad_norm: 7.2649 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.8081 loss: 0.8081 2022/09/07 18:03:29 - mmengine - INFO - Epoch(train) [84][1900/3757] lr: 1.0000e-04 eta: 2:52:38 time: 0.1568 data_time: 0.0110 memory: 7124 grad_norm: 7.2720 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0572 loss: 1.0572 2022/09/07 18:03:45 - mmengine - INFO - Epoch(train) [84][2000/3757] lr: 1.0000e-04 eta: 2:52:21 time: 0.1572 data_time: 0.0119 memory: 7124 grad_norm: 7.2803 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.7484 loss: 0.7484 2022/09/07 18:04:02 - mmengine - INFO - Epoch(train) [84][2100/3757] lr: 1.0000e-04 eta: 2:52:04 time: 0.1595 data_time: 0.0103 memory: 7124 grad_norm: 7.1556 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0751 loss: 1.0751 2022/09/07 18:04:13 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:04:18 - mmengine - INFO - Epoch(train) [84][2200/3757] lr: 1.0000e-04 eta: 2:51:48 time: 0.1581 data_time: 0.0117 memory: 7124 grad_norm: 7.2375 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0909 loss: 1.0909 2022/09/07 18:04:34 - mmengine - INFO - Epoch(train) [84][2300/3757] lr: 1.0000e-04 eta: 2:51:31 time: 0.1556 data_time: 0.0106 memory: 7124 grad_norm: 7.1339 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9996 loss: 0.9996 2022/09/07 18:04:50 - mmengine - INFO - Epoch(train) [84][2400/3757] lr: 1.0000e-04 eta: 2:51:14 time: 0.1587 data_time: 0.0107 memory: 7124 grad_norm: 6.9505 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9343 loss: 0.9343 2022/09/07 18:05:06 - mmengine - INFO - Epoch(train) [84][2500/3757] lr: 1.0000e-04 eta: 2:50:57 time: 0.1583 data_time: 0.0111 memory: 7124 grad_norm: 7.2727 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1638 loss: 1.1638 2022/09/07 18:05:23 - mmengine - INFO - Epoch(train) [84][2600/3757] lr: 1.0000e-04 eta: 2:50:40 time: 0.1636 data_time: 0.0104 memory: 7124 grad_norm: 7.2132 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8969 loss: 0.8969 2022/09/07 18:05:39 - mmengine - INFO - Epoch(train) [84][2700/3757] lr: 1.0000e-04 eta: 2:50:23 time: 0.1597 data_time: 0.0105 memory: 7124 grad_norm: 7.1836 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1992 loss: 1.1992 2022/09/07 18:05:55 - mmengine - INFO - Epoch(train) [84][2800/3757] lr: 1.0000e-04 eta: 2:50:07 time: 0.1605 data_time: 0.0108 memory: 7124 grad_norm: 6.9358 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.1143 loss: 1.1143 2022/09/07 18:06:11 - mmengine - INFO - Epoch(train) [84][2900/3757] lr: 1.0000e-04 eta: 2:49:50 time: 0.1602 data_time: 0.0109 memory: 7124 grad_norm: 6.9403 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9370 loss: 0.9370 2022/09/07 18:06:28 - mmengine - INFO - Epoch(train) [84][3000/3757] lr: 1.0000e-04 eta: 2:49:33 time: 0.1607 data_time: 0.0106 memory: 7124 grad_norm: 7.1812 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8323 loss: 0.8323 2022/09/07 18:06:44 - mmengine - INFO - Epoch(train) [84][3100/3757] lr: 1.0000e-04 eta: 2:49:16 time: 0.1595 data_time: 0.0119 memory: 7124 grad_norm: 7.1272 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.1518 loss: 1.1518 2022/09/07 18:06:55 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:07:00 - mmengine - INFO - Epoch(train) [84][3200/3757] lr: 1.0000e-04 eta: 2:48:59 time: 0.1577 data_time: 0.0108 memory: 7124 grad_norm: 7.0816 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9752 loss: 0.9752 2022/09/07 18:07:16 - mmengine - INFO - Epoch(train) [84][3300/3757] lr: 1.0000e-04 eta: 2:48:42 time: 0.1582 data_time: 0.0108 memory: 7124 grad_norm: 6.6599 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9090 loss: 0.9090 2022/09/07 18:07:32 - mmengine - INFO - Epoch(train) [84][3400/3757] lr: 1.0000e-04 eta: 2:48:26 time: 0.1661 data_time: 0.0155 memory: 7124 grad_norm: 7.2659 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9473 loss: 0.9473 2022/09/07 18:07:49 - mmengine - INFO - Epoch(train) [84][3500/3757] lr: 1.0000e-04 eta: 2:48:09 time: 0.1554 data_time: 0.0098 memory: 7124 grad_norm: 7.0139 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0330 loss: 1.0330 2022/09/07 18:08:05 - mmengine - INFO - Epoch(train) [84][3600/3757] lr: 1.0000e-04 eta: 2:47:52 time: 0.1584 data_time: 0.0105 memory: 7124 grad_norm: 7.0945 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1032 loss: 1.1032 2022/09/07 18:08:21 - mmengine - INFO - Epoch(train) [84][3700/3757] lr: 1.0000e-04 eta: 2:47:35 time: 0.1622 data_time: 0.0102 memory: 7124 grad_norm: 7.3619 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0670 loss: 1.0670 2022/09/07 18:08:30 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:08:30 - mmengine - INFO - Epoch(train) [84][3757/3757] lr: 1.0000e-04 eta: 2:47:28 time: 0.1461 data_time: 0.0099 memory: 7124 grad_norm: 7.0582 top1_acc: 0.8571 top5_acc: 0.8571 loss_cls: 1.1441 loss: 1.1441 2022/09/07 18:08:30 - mmengine - INFO - Saving checkpoint at 84 epochs 2022/09/07 18:10:48 - mmengine - INFO - Epoch(val) [84][100/310] eta: 0:03:38 time: 1.0384 data_time: 0.7357 memory: 7627 2022/09/07 18:13:08 - mmengine - INFO - Epoch(val) [84][200/310] eta: 0:02:31 time: 1.3766 data_time: 1.0758 memory: 7627 2022/09/07 18:15:11 - mmengine - INFO - Epoch(val) [84][300/310] eta: 0:00:11 time: 1.1066 data_time: 0.8072 memory: 7627 2022/09/07 18:15:27 - mmengine - INFO - Epoch(val) [84][310/310] acc/top1: 0.7518 acc/top5: 0.9180 acc/mean1: 0.7517 2022/09/07 18:15:45 - mmengine - INFO - Epoch(train) [85][100/3757] lr: 1.0000e-04 eta: 2:47:08 time: 0.1590 data_time: 0.0106 memory: 7627 grad_norm: 7.1889 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.0104 loss: 1.0104 2022/09/07 18:16:01 - mmengine - INFO - Epoch(train) [85][200/3757] lr: 1.0000e-04 eta: 2:46:52 time: 0.1577 data_time: 0.0118 memory: 7124 grad_norm: 6.9565 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1835 loss: 1.1835 2022/09/07 18:16:17 - mmengine - INFO - Epoch(train) [85][300/3757] lr: 1.0000e-04 eta: 2:46:35 time: 0.1585 data_time: 0.0106 memory: 7124 grad_norm: 7.1141 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.1932 loss: 1.1932 2022/09/07 18:16:33 - mmengine - INFO - Epoch(train) [85][400/3757] lr: 1.0000e-04 eta: 2:46:18 time: 0.1608 data_time: 0.0113 memory: 7124 grad_norm: 7.2435 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0208 loss: 1.0208 2022/09/07 18:16:35 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:16:49 - mmengine - INFO - Epoch(train) [85][500/3757] lr: 1.0000e-04 eta: 2:46:01 time: 0.1591 data_time: 0.0108 memory: 7124 grad_norm: 7.4902 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8608 loss: 0.8608 2022/09/07 18:17:05 - mmengine - INFO - Epoch(train) [85][600/3757] lr: 1.0000e-04 eta: 2:45:44 time: 0.1583 data_time: 0.0108 memory: 7124 grad_norm: 7.3197 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1175 loss: 1.1175 2022/09/07 18:17:22 - mmengine - INFO - Epoch(train) [85][700/3757] lr: 1.0000e-04 eta: 2:45:27 time: 0.1627 data_time: 0.0112 memory: 7124 grad_norm: 7.0484 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8709 loss: 0.8709 2022/09/07 18:17:38 - mmengine - INFO - Epoch(train) [85][800/3757] lr: 1.0000e-04 eta: 2:45:11 time: 0.1621 data_time: 0.0114 memory: 7124 grad_norm: 7.3187 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2251 loss: 1.2251 2022/09/07 18:17:54 - mmengine - INFO - Epoch(train) [85][900/3757] lr: 1.0000e-04 eta: 2:44:54 time: 0.1623 data_time: 0.0106 memory: 7124 grad_norm: 7.2362 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0867 loss: 1.0867 2022/09/07 18:18:10 - mmengine - INFO - Epoch(train) [85][1000/3757] lr: 1.0000e-04 eta: 2:44:37 time: 0.1575 data_time: 0.0113 memory: 7124 grad_norm: 7.4246 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1444 loss: 1.1444 2022/09/07 18:18:26 - mmengine - INFO - Epoch(train) [85][1100/3757] lr: 1.0000e-04 eta: 2:44:20 time: 0.1607 data_time: 0.0114 memory: 7124 grad_norm: 7.1055 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1647 loss: 1.1647 2022/09/07 18:18:43 - mmengine - INFO - Epoch(train) [85][1200/3757] lr: 1.0000e-04 eta: 2:44:03 time: 0.1625 data_time: 0.0109 memory: 7124 grad_norm: 7.0429 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2742 loss: 1.2742 2022/09/07 18:18:59 - mmengine - INFO - Epoch(train) [85][1300/3757] lr: 1.0000e-04 eta: 2:43:46 time: 0.1592 data_time: 0.0110 memory: 7124 grad_norm: 7.0946 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9067 loss: 0.9067 2022/09/07 18:19:15 - mmengine - INFO - Epoch(train) [85][1400/3757] lr: 1.0000e-04 eta: 2:43:30 time: 0.1598 data_time: 0.0111 memory: 7124 grad_norm: 7.0913 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0056 loss: 1.0056 2022/09/07 18:19:17 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:19:31 - mmengine - INFO - Epoch(train) [85][1500/3757] lr: 1.0000e-04 eta: 2:43:13 time: 0.1589 data_time: 0.0090 memory: 7124 grad_norm: 7.1661 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9438 loss: 0.9438 2022/09/07 18:19:47 - mmengine - INFO - Epoch(train) [85][1600/3757] lr: 1.0000e-04 eta: 2:42:56 time: 0.1580 data_time: 0.0108 memory: 7124 grad_norm: 7.1665 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0385 loss: 1.0385 2022/09/07 18:20:03 - mmengine - INFO - Epoch(train) [85][1700/3757] lr: 1.0000e-04 eta: 2:42:39 time: 0.1640 data_time: 0.0124 memory: 7124 grad_norm: 6.8601 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8396 loss: 0.8396 2022/09/07 18:20:19 - mmengine - INFO - Epoch(train) [85][1800/3757] lr: 1.0000e-04 eta: 2:42:22 time: 0.1606 data_time: 0.0113 memory: 7124 grad_norm: 6.9043 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.7740 loss: 0.7740 2022/09/07 18:20:36 - mmengine - INFO - Epoch(train) [85][1900/3757] lr: 1.0000e-04 eta: 2:42:05 time: 0.1599 data_time: 0.0105 memory: 7124 grad_norm: 7.0439 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9060 loss: 0.9060 2022/09/07 18:20:52 - mmengine - INFO - Epoch(train) [85][2000/3757] lr: 1.0000e-04 eta: 2:41:49 time: 0.1617 data_time: 0.0106 memory: 7124 grad_norm: 7.2485 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0187 loss: 1.0187 2022/09/07 18:21:08 - mmengine - INFO - Epoch(train) [85][2100/3757] lr: 1.0000e-04 eta: 2:41:32 time: 0.1589 data_time: 0.0107 memory: 7124 grad_norm: 7.0139 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1605 loss: 1.1605 2022/09/07 18:21:25 - mmengine - INFO - Epoch(train) [85][2200/3757] lr: 1.0000e-04 eta: 2:41:15 time: 0.1616 data_time: 0.0111 memory: 7124 grad_norm: 7.2651 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9535 loss: 0.9535 2022/09/07 18:21:41 - mmengine - INFO - Epoch(train) [85][2300/3757] lr: 1.0000e-04 eta: 2:40:58 time: 0.1602 data_time: 0.0112 memory: 7124 grad_norm: 7.1262 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0380 loss: 1.0380 2022/09/07 18:21:57 - mmengine - INFO - Epoch(train) [85][2400/3757] lr: 1.0000e-04 eta: 2:40:41 time: 0.1577 data_time: 0.0121 memory: 7124 grad_norm: 6.8389 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9480 loss: 0.9480 2022/09/07 18:21:59 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:22:13 - mmengine - INFO - Epoch(train) [85][2500/3757] lr: 1.0000e-04 eta: 2:40:25 time: 0.1614 data_time: 0.0123 memory: 7124 grad_norm: 7.3713 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.0982 loss: 1.0982 2022/09/07 18:22:29 - mmengine - INFO - Epoch(train) [85][2600/3757] lr: 1.0000e-04 eta: 2:40:08 time: 0.1593 data_time: 0.0111 memory: 7124 grad_norm: 7.0817 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1692 loss: 1.1692 2022/09/07 18:22:46 - mmengine - INFO - Epoch(train) [85][2700/3757] lr: 1.0000e-04 eta: 2:39:51 time: 0.1670 data_time: 0.0118 memory: 7124 grad_norm: 7.1516 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.0152 loss: 1.0152 2022/09/07 18:23:02 - mmengine - INFO - Epoch(train) [85][2800/3757] lr: 1.0000e-04 eta: 2:39:34 time: 0.1610 data_time: 0.0115 memory: 7124 grad_norm: 7.0162 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8543 loss: 0.8543 2022/09/07 18:23:18 - mmengine - INFO - Epoch(train) [85][2900/3757] lr: 1.0000e-04 eta: 2:39:17 time: 0.1631 data_time: 0.0122 memory: 7124 grad_norm: 7.0174 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.7099 loss: 0.7099 2022/09/07 18:23:34 - mmengine - INFO - Epoch(train) [85][3000/3757] lr: 1.0000e-04 eta: 2:39:00 time: 0.1602 data_time: 0.0110 memory: 7124 grad_norm: 7.2860 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8942 loss: 0.8942 2022/09/07 18:23:50 - mmengine - INFO - Epoch(train) [85][3100/3757] lr: 1.0000e-04 eta: 2:38:44 time: 0.1612 data_time: 0.0119 memory: 7124 grad_norm: 7.3595 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8330 loss: 0.8330 2022/09/07 18:24:06 - mmengine - INFO - Epoch(train) [85][3200/3757] lr: 1.0000e-04 eta: 2:38:27 time: 0.1608 data_time: 0.0115 memory: 7124 grad_norm: 7.1811 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9223 loss: 0.9223 2022/09/07 18:24:23 - mmengine - INFO - Epoch(train) [85][3300/3757] lr: 1.0000e-04 eta: 2:38:10 time: 0.1600 data_time: 0.0106 memory: 7124 grad_norm: 7.0131 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9228 loss: 0.9228 2022/09/07 18:24:39 - mmengine - INFO - Epoch(train) [85][3400/3757] lr: 1.0000e-04 eta: 2:37:53 time: 0.1628 data_time: 0.0113 memory: 7124 grad_norm: 7.0807 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2185 loss: 1.2185 2022/09/07 18:24:41 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:24:56 - mmengine - INFO - Epoch(train) [85][3500/3757] lr: 1.0000e-04 eta: 2:37:37 time: 0.1641 data_time: 0.0109 memory: 7124 grad_norm: 7.3233 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2628 loss: 1.2628 2022/09/07 18:25:13 - mmengine - INFO - Epoch(train) [85][3600/3757] lr: 1.0000e-04 eta: 2:37:20 time: 0.1659 data_time: 0.0113 memory: 7124 grad_norm: 7.2116 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8924 loss: 0.8924 2022/09/07 18:25:29 - mmengine - INFO - Epoch(train) [85][3700/3757] lr: 1.0000e-04 eta: 2:37:03 time: 0.1640 data_time: 0.0122 memory: 7124 grad_norm: 6.9508 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8424 loss: 0.8424 2022/09/07 18:25:38 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:25:38 - mmengine - INFO - Epoch(train) [85][3757/3757] lr: 1.0000e-04 eta: 2:36:56 time: 0.1411 data_time: 0.0078 memory: 7124 grad_norm: 7.2409 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1215 loss: 1.1215 2022/09/07 18:25:38 - mmengine - INFO - Saving checkpoint at 85 epochs 2022/09/07 18:28:02 - mmengine - INFO - Epoch(val) [85][100/310] eta: 0:04:28 time: 1.2770 data_time: 0.9730 memory: 7627 2022/09/07 18:30:21 - mmengine - INFO - Epoch(val) [85][200/310] eta: 0:02:26 time: 1.3299 data_time: 1.0268 memory: 7627 2022/09/07 18:32:31 - mmengine - INFO - Epoch(val) [85][300/310] eta: 0:00:11 time: 1.1797 data_time: 0.8778 memory: 7627 2022/09/07 18:32:48 - mmengine - INFO - Epoch(val) [85][310/310] acc/top1: 0.7508 acc/top5: 0.9179 acc/mean1: 0.7507 2022/09/07 18:33:07 - mmengine - INFO - Epoch(train) [86][100/3757] lr: 1.0000e-04 eta: 2:36:37 time: 0.1629 data_time: 0.0102 memory: 7627 grad_norm: 6.9560 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 0.9149 loss: 0.9149 2022/09/07 18:33:24 - mmengine - INFO - Epoch(train) [86][200/3757] lr: 1.0000e-04 eta: 2:36:20 time: 0.1725 data_time: 0.0120 memory: 7124 grad_norm: 7.0540 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0986 loss: 1.0986 2022/09/07 18:33:40 - mmengine - INFO - Epoch(train) [86][300/3757] lr: 1.0000e-04 eta: 2:36:03 time: 0.1646 data_time: 0.0112 memory: 7124 grad_norm: 7.1320 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1651 loss: 1.1651 2022/09/07 18:33:57 - mmengine - INFO - Epoch(train) [86][400/3757] lr: 1.0000e-04 eta: 2:35:46 time: 0.1646 data_time: 0.0125 memory: 7124 grad_norm: 7.0533 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1567 loss: 1.1567 2022/09/07 18:34:14 - mmengine - INFO - Epoch(train) [86][500/3757] lr: 1.0000e-04 eta: 2:35:30 time: 0.1670 data_time: 0.0116 memory: 7124 grad_norm: 7.3200 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1506 loss: 1.1506 2022/09/07 18:34:31 - mmengine - INFO - Epoch(train) [86][600/3757] lr: 1.0000e-04 eta: 2:35:13 time: 0.1649 data_time: 0.0112 memory: 7124 grad_norm: 7.1554 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0705 loss: 1.0705 2022/09/07 18:34:40 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:34:48 - mmengine - INFO - Epoch(train) [86][700/3757] lr: 1.0000e-04 eta: 2:34:56 time: 0.1628 data_time: 0.0120 memory: 7124 grad_norm: 7.1400 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9715 loss: 0.9715 2022/09/07 18:35:04 - mmengine - INFO - Epoch(train) [86][800/3757] lr: 1.0000e-04 eta: 2:34:40 time: 0.1727 data_time: 0.0122 memory: 7124 grad_norm: 7.2696 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9292 loss: 0.9292 2022/09/07 18:35:21 - mmengine - INFO - Epoch(train) [86][900/3757] lr: 1.0000e-04 eta: 2:34:23 time: 0.1645 data_time: 0.0129 memory: 7124 grad_norm: 6.9207 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0009 loss: 1.0009 2022/09/07 18:35:38 - mmengine - INFO - Epoch(train) [86][1000/3757] lr: 1.0000e-04 eta: 2:34:06 time: 0.1615 data_time: 0.0106 memory: 7124 grad_norm: 7.0599 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0202 loss: 1.0202 2022/09/07 18:35:54 - mmengine - INFO - Epoch(train) [86][1100/3757] lr: 1.0000e-04 eta: 2:33:49 time: 0.1638 data_time: 0.0101 memory: 7124 grad_norm: 7.4360 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9860 loss: 0.9860 2022/09/07 18:36:11 - mmengine - INFO - Epoch(train) [86][1200/3757] lr: 1.0000e-04 eta: 2:33:33 time: 0.1619 data_time: 0.0117 memory: 7124 grad_norm: 7.2196 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0817 loss: 1.0817 2022/09/07 18:36:27 - mmengine - INFO - Epoch(train) [86][1300/3757] lr: 1.0000e-04 eta: 2:33:16 time: 0.1609 data_time: 0.0095 memory: 7124 grad_norm: 6.9770 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9780 loss: 0.9780 2022/09/07 18:36:43 - mmengine - INFO - Epoch(train) [86][1400/3757] lr: 1.0000e-04 eta: 2:32:59 time: 0.1626 data_time: 0.0116 memory: 7124 grad_norm: 7.2545 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1117 loss: 1.1117 2022/09/07 18:37:00 - mmengine - INFO - Epoch(train) [86][1500/3757] lr: 1.0000e-04 eta: 2:32:42 time: 0.1650 data_time: 0.0114 memory: 7124 grad_norm: 6.8144 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9349 loss: 0.9349 2022/09/07 18:37:16 - mmengine - INFO - Epoch(train) [86][1600/3757] lr: 1.0000e-04 eta: 2:32:26 time: 0.1633 data_time: 0.0115 memory: 7124 grad_norm: 7.3098 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0048 loss: 1.0048 2022/09/07 18:37:26 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:37:33 - mmengine - INFO - Epoch(train) [86][1700/3757] lr: 1.0000e-04 eta: 2:32:09 time: 0.1644 data_time: 0.0113 memory: 7124 grad_norm: 7.1336 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1202 loss: 1.1202 2022/09/07 18:37:50 - mmengine - INFO - Epoch(train) [86][1800/3757] lr: 1.0000e-04 eta: 2:31:52 time: 0.1661 data_time: 0.0121 memory: 7124 grad_norm: 7.2352 top1_acc: 0.3750 top5_acc: 1.0000 loss_cls: 0.8625 loss: 0.8625 2022/09/07 18:38:06 - mmengine - INFO - Epoch(train) [86][1900/3757] lr: 1.0000e-04 eta: 2:31:35 time: 0.1626 data_time: 0.0107 memory: 7124 grad_norm: 7.1366 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.8635 loss: 0.8635 2022/09/07 18:38:22 - mmengine - INFO - Epoch(train) [86][2000/3757] lr: 1.0000e-04 eta: 2:31:19 time: 0.1621 data_time: 0.0101 memory: 7124 grad_norm: 7.2648 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8906 loss: 0.8906 2022/09/07 18:38:39 - mmengine - INFO - Epoch(train) [86][2100/3757] lr: 1.0000e-04 eta: 2:31:02 time: 0.1634 data_time: 0.0112 memory: 7124 grad_norm: 6.8600 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0782 loss: 1.0782 2022/09/07 18:38:55 - mmengine - INFO - Epoch(train) [86][2200/3757] lr: 1.0000e-04 eta: 2:30:45 time: 0.1621 data_time: 0.0112 memory: 7124 grad_norm: 7.0816 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8699 loss: 0.8699 2022/09/07 18:39:12 - mmengine - INFO - Epoch(train) [86][2300/3757] lr: 1.0000e-04 eta: 2:30:28 time: 0.1640 data_time: 0.0112 memory: 7124 grad_norm: 7.2785 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0406 loss: 1.0406 2022/09/07 18:39:28 - mmengine - INFO - Epoch(train) [86][2400/3757] lr: 1.0000e-04 eta: 2:30:11 time: 0.1616 data_time: 0.0112 memory: 7124 grad_norm: 7.3628 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9423 loss: 0.9423 2022/09/07 18:39:45 - mmengine - INFO - Epoch(train) [86][2500/3757] lr: 1.0000e-04 eta: 2:29:55 time: 0.1610 data_time: 0.0105 memory: 7124 grad_norm: 7.1140 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9931 loss: 0.9931 2022/09/07 18:40:01 - mmengine - INFO - Epoch(train) [86][2600/3757] lr: 1.0000e-04 eta: 2:29:38 time: 0.1614 data_time: 0.0108 memory: 7124 grad_norm: 7.0727 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9879 loss: 0.9879 2022/09/07 18:40:10 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:40:18 - mmengine - INFO - Epoch(train) [86][2700/3757] lr: 1.0000e-04 eta: 2:29:21 time: 0.1660 data_time: 0.0109 memory: 7124 grad_norm: 7.0335 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.7726 loss: 0.7726 2022/09/07 18:40:35 - mmengine - INFO - Epoch(train) [86][2800/3757] lr: 1.0000e-04 eta: 2:29:05 time: 0.1684 data_time: 0.0105 memory: 7124 grad_norm: 7.4522 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.3721 loss: 1.3721 2022/09/07 18:40:51 - mmengine - INFO - Epoch(train) [86][2900/3757] lr: 1.0000e-04 eta: 2:28:48 time: 0.1636 data_time: 0.0116 memory: 7124 grad_norm: 7.2157 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0815 loss: 1.0815 2022/09/07 18:41:08 - mmengine - INFO - Epoch(train) [86][3000/3757] lr: 1.0000e-04 eta: 2:28:31 time: 0.1615 data_time: 0.0114 memory: 7124 grad_norm: 7.3821 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9937 loss: 0.9937 2022/09/07 18:41:24 - mmengine - INFO - Epoch(train) [86][3100/3757] lr: 1.0000e-04 eta: 2:28:14 time: 0.1637 data_time: 0.0107 memory: 7124 grad_norm: 7.0246 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0308 loss: 1.0308 2022/09/07 18:41:40 - mmengine - INFO - Epoch(train) [86][3200/3757] lr: 1.0000e-04 eta: 2:27:57 time: 0.1608 data_time: 0.0103 memory: 7124 grad_norm: 7.1692 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2931 loss: 1.2931 2022/09/07 18:41:57 - mmengine - INFO - Epoch(train) [86][3300/3757] lr: 1.0000e-04 eta: 2:27:41 time: 0.1663 data_time: 0.0103 memory: 7124 grad_norm: 7.0981 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9681 loss: 0.9681 2022/09/07 18:42:13 - mmengine - INFO - Epoch(train) [86][3400/3757] lr: 1.0000e-04 eta: 2:27:24 time: 0.1596 data_time: 0.0114 memory: 7124 grad_norm: 7.2384 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9520 loss: 0.9520 2022/09/07 18:42:30 - mmengine - INFO - Epoch(train) [86][3500/3757] lr: 1.0000e-04 eta: 2:27:07 time: 0.1597 data_time: 0.0111 memory: 7124 grad_norm: 7.3506 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0780 loss: 1.0780 2022/09/07 18:42:46 - mmengine - INFO - Epoch(train) [86][3600/3757] lr: 1.0000e-04 eta: 2:26:50 time: 0.1617 data_time: 0.0110 memory: 7124 grad_norm: 7.2235 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.7917 loss: 0.7917 2022/09/07 18:42:55 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:43:02 - mmengine - INFO - Epoch(train) [86][3700/3757] lr: 1.0000e-04 eta: 2:26:34 time: 0.1633 data_time: 0.0109 memory: 7124 grad_norm: 6.9465 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1372 loss: 1.1372 2022/09/07 18:43:11 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:43:11 - mmengine - INFO - Epoch(train) [86][3757/3757] lr: 1.0000e-04 eta: 2:26:27 time: 0.1405 data_time: 0.0076 memory: 7124 grad_norm: 7.1102 top1_acc: 0.7143 top5_acc: 1.0000 loss_cls: 1.0384 loss: 1.0384 2022/09/07 18:43:11 - mmengine - INFO - Saving checkpoint at 86 epochs 2022/09/07 18:45:33 - mmengine - INFO - Epoch(val) [86][100/310] eta: 0:04:16 time: 1.2205 data_time: 0.9136 memory: 7627 2022/09/07 18:47:47 - mmengine - INFO - Epoch(val) [86][200/310] eta: 0:02:06 time: 1.1510 data_time: 0.8461 memory: 7627 2022/09/07 18:49:56 - mmengine - INFO - Epoch(val) [86][300/310] eta: 0:00:13 time: 1.3225 data_time: 1.0187 memory: 7627 2022/09/07 18:50:16 - mmengine - INFO - Epoch(val) [86][310/310] acc/top1: 0.7538 acc/top5: 0.9174 acc/mean1: 0.7538 2022/09/07 18:50:16 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_78.pth is removed 2022/09/07 18:50:18 - mmengine - INFO - The best checkpoint with 0.7538 acc/top1 at 86 epoch is saved to best_acc/top1_epoch_86.pth. 2022/09/07 18:50:36 - mmengine - INFO - Epoch(train) [87][100/3757] lr: 1.0000e-04 eta: 2:26:07 time: 0.1632 data_time: 0.0131 memory: 7627 grad_norm: 7.1234 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1678 loss: 1.1678 2022/09/07 18:50:52 - mmengine - INFO - Epoch(train) [87][200/3757] lr: 1.0000e-04 eta: 2:25:50 time: 0.1604 data_time: 0.0116 memory: 7124 grad_norm: 7.2966 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9851 loss: 0.9851 2022/09/07 18:51:08 - mmengine - INFO - Epoch(train) [87][300/3757] lr: 1.0000e-04 eta: 2:25:33 time: 0.1598 data_time: 0.0109 memory: 7124 grad_norm: 7.3582 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0907 loss: 1.0907 2022/09/07 18:51:25 - mmengine - INFO - Epoch(train) [87][400/3757] lr: 1.0000e-04 eta: 2:25:17 time: 0.1609 data_time: 0.0110 memory: 7124 grad_norm: 7.2450 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.4118 loss: 1.4118 2022/09/07 18:51:41 - mmengine - INFO - Epoch(train) [87][500/3757] lr: 1.0000e-04 eta: 2:25:00 time: 0.1610 data_time: 0.0106 memory: 7124 grad_norm: 7.4613 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0287 loss: 1.0287 2022/09/07 18:51:57 - mmengine - INFO - Epoch(train) [87][600/3757] lr: 1.0000e-04 eta: 2:24:43 time: 0.1570 data_time: 0.0117 memory: 7124 grad_norm: 7.1330 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1015 loss: 1.1015 2022/09/07 18:52:14 - mmengine - INFO - Epoch(train) [87][700/3757] lr: 1.0000e-04 eta: 2:24:26 time: 0.1604 data_time: 0.0109 memory: 7124 grad_norm: 7.1455 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.7615 loss: 0.7615 2022/09/07 18:52:30 - mmengine - INFO - Epoch(train) [87][800/3757] lr: 1.0000e-04 eta: 2:24:10 time: 0.1617 data_time: 0.0108 memory: 7124 grad_norm: 7.0022 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8871 loss: 0.8871 2022/09/07 18:52:46 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:52:46 - mmengine - INFO - Epoch(train) [87][900/3757] lr: 1.0000e-04 eta: 2:23:53 time: 0.1594 data_time: 0.0102 memory: 7124 grad_norm: 7.1966 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9377 loss: 0.9377 2022/09/07 18:53:02 - mmengine - INFO - Epoch(train) [87][1000/3757] lr: 1.0000e-04 eta: 2:23:36 time: 0.1609 data_time: 0.0105 memory: 7124 grad_norm: 7.2468 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0535 loss: 1.0535 2022/09/07 18:53:18 - mmengine - INFO - Epoch(train) [87][1100/3757] lr: 1.0000e-04 eta: 2:23:19 time: 0.1561 data_time: 0.0101 memory: 7124 grad_norm: 6.9705 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0080 loss: 1.0080 2022/09/07 18:53:35 - mmengine - INFO - Epoch(train) [87][1200/3757] lr: 1.0000e-04 eta: 2:23:02 time: 0.1634 data_time: 0.0113 memory: 7124 grad_norm: 7.1612 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0380 loss: 1.0380 2022/09/07 18:53:51 - mmengine - INFO - Epoch(train) [87][1300/3757] lr: 1.0000e-04 eta: 2:22:46 time: 0.1630 data_time: 0.0108 memory: 7124 grad_norm: 7.4909 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0247 loss: 1.0247 2022/09/07 18:54:08 - mmengine - INFO - Epoch(train) [87][1400/3757] lr: 1.0000e-04 eta: 2:22:29 time: 0.1650 data_time: 0.0117 memory: 7124 grad_norm: 7.4035 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1590 loss: 1.1590 2022/09/07 18:54:25 - mmengine - INFO - Epoch(train) [87][1500/3757] lr: 1.0000e-04 eta: 2:22:12 time: 0.1644 data_time: 0.0113 memory: 7124 grad_norm: 7.0837 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8723 loss: 0.8723 2022/09/07 18:54:42 - mmengine - INFO - Epoch(train) [87][1600/3757] lr: 1.0000e-04 eta: 2:21:55 time: 0.1668 data_time: 0.0110 memory: 7124 grad_norm: 7.4189 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 0.8182 loss: 0.8182 2022/09/07 18:54:59 - mmengine - INFO - Epoch(train) [87][1700/3757] lr: 1.0000e-04 eta: 2:21:39 time: 0.1760 data_time: 0.0256 memory: 7124 grad_norm: 7.0986 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1852 loss: 1.1852 2022/09/07 18:55:15 - mmengine - INFO - Epoch(train) [87][1800/3757] lr: 1.0000e-04 eta: 2:21:22 time: 0.1740 data_time: 0.0114 memory: 7124 grad_norm: 7.1645 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0448 loss: 1.0448 2022/09/07 18:55:32 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:55:32 - mmengine - INFO - Epoch(train) [87][1900/3757] lr: 1.0000e-04 eta: 2:21:05 time: 0.1621 data_time: 0.0121 memory: 7124 grad_norm: 7.1640 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8862 loss: 0.8862 2022/09/07 18:55:49 - mmengine - INFO - Epoch(train) [87][2000/3757] lr: 1.0000e-04 eta: 2:20:49 time: 0.1606 data_time: 0.0106 memory: 7124 grad_norm: 7.3047 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.7888 loss: 0.7888 2022/09/07 18:56:06 - mmengine - INFO - Epoch(train) [87][2100/3757] lr: 1.0000e-04 eta: 2:20:32 time: 0.1763 data_time: 0.0233 memory: 7124 grad_norm: 7.1221 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.9081 loss: 0.9081 2022/09/07 18:56:23 - mmengine - INFO - Epoch(train) [87][2200/3757] lr: 1.0000e-04 eta: 2:20:15 time: 0.1618 data_time: 0.0117 memory: 7124 grad_norm: 7.4110 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2624 loss: 1.2624 2022/09/07 18:56:40 - mmengine - INFO - Epoch(train) [87][2300/3757] lr: 1.0000e-04 eta: 2:19:59 time: 0.1636 data_time: 0.0111 memory: 7124 grad_norm: 7.2185 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0490 loss: 1.0490 2022/09/07 18:56:56 - mmengine - INFO - Epoch(train) [87][2400/3757] lr: 1.0000e-04 eta: 2:19:42 time: 0.1670 data_time: 0.0112 memory: 7124 grad_norm: 7.3275 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0484 loss: 1.0484 2022/09/07 18:57:13 - mmengine - INFO - Epoch(train) [87][2500/3757] lr: 1.0000e-04 eta: 2:19:25 time: 0.1679 data_time: 0.0116 memory: 7124 grad_norm: 6.9275 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9186 loss: 0.9186 2022/09/07 18:57:30 - mmengine - INFO - Epoch(train) [87][2600/3757] lr: 1.0000e-04 eta: 2:19:09 time: 0.1661 data_time: 0.0106 memory: 7124 grad_norm: 6.9275 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2271 loss: 1.2271 2022/09/07 18:57:47 - mmengine - INFO - Epoch(train) [87][2700/3757] lr: 1.0000e-04 eta: 2:18:52 time: 0.1675 data_time: 0.0134 memory: 7124 grad_norm: 7.6335 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1939 loss: 1.1939 2022/09/07 18:58:03 - mmengine - INFO - Epoch(train) [87][2800/3757] lr: 1.0000e-04 eta: 2:18:35 time: 0.1646 data_time: 0.0106 memory: 7124 grad_norm: 7.3980 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0714 loss: 1.0714 2022/09/07 18:58:20 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 18:58:20 - mmengine - INFO - Epoch(train) [87][2900/3757] lr: 1.0000e-04 eta: 2:18:18 time: 0.1631 data_time: 0.0107 memory: 7124 grad_norm: 7.3003 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9884 loss: 0.9884 2022/09/07 18:58:37 - mmengine - INFO - Epoch(train) [87][3000/3757] lr: 1.0000e-04 eta: 2:18:02 time: 0.1632 data_time: 0.0107 memory: 7124 grad_norm: 7.0951 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1420 loss: 1.1420 2022/09/07 18:58:53 - mmengine - INFO - Epoch(train) [87][3100/3757] lr: 1.0000e-04 eta: 2:17:45 time: 0.1662 data_time: 0.0117 memory: 7124 grad_norm: 7.2570 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8917 loss: 0.8917 2022/09/07 18:59:10 - mmengine - INFO - Epoch(train) [87][3200/3757] lr: 1.0000e-04 eta: 2:17:28 time: 0.1641 data_time: 0.0116 memory: 7124 grad_norm: 7.1282 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1584 loss: 1.1584 2022/09/07 18:59:27 - mmengine - INFO - Epoch(train) [87][3300/3757] lr: 1.0000e-04 eta: 2:17:11 time: 0.1733 data_time: 0.0119 memory: 7124 grad_norm: 7.2721 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2210 loss: 1.2210 2022/09/07 18:59:44 - mmengine - INFO - Epoch(train) [87][3400/3757] lr: 1.0000e-04 eta: 2:16:55 time: 0.1667 data_time: 0.0113 memory: 7124 grad_norm: 6.9323 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0629 loss: 1.0629 2022/09/07 19:00:00 - mmengine - INFO - Epoch(train) [87][3500/3757] lr: 1.0000e-04 eta: 2:16:38 time: 0.1639 data_time: 0.0104 memory: 7124 grad_norm: 7.4085 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.0338 loss: 1.0338 2022/09/07 19:00:18 - mmengine - INFO - Epoch(train) [87][3600/3757] lr: 1.0000e-04 eta: 2:16:21 time: 0.1838 data_time: 0.0116 memory: 7124 grad_norm: 7.4816 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1757 loss: 1.1757 2022/09/07 19:00:34 - mmengine - INFO - Epoch(train) [87][3700/3757] lr: 1.0000e-04 eta: 2:16:05 time: 0.1656 data_time: 0.0105 memory: 7124 grad_norm: 7.3918 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1325 loss: 1.1325 2022/09/07 19:00:44 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:00:44 - mmengine - INFO - Epoch(train) [87][3757/3757] lr: 1.0000e-04 eta: 2:15:58 time: 0.1397 data_time: 0.0081 memory: 7124 grad_norm: 7.3448 top1_acc: 0.8571 top5_acc: 1.0000 loss_cls: 1.1672 loss: 1.1672 2022/09/07 19:00:44 - mmengine - INFO - Saving checkpoint at 87 epochs 2022/09/07 19:03:09 - mmengine - INFO - Epoch(val) [87][100/310] eta: 0:04:24 time: 1.2589 data_time: 0.9556 memory: 7627 2022/09/07 19:05:29 - mmengine - INFO - Epoch(val) [87][200/310] eta: 0:02:16 time: 1.2439 data_time: 0.9410 memory: 7627 2022/09/07 19:07:39 - mmengine - INFO - Epoch(val) [87][300/310] eta: 0:00:12 time: 1.2785 data_time: 0.9689 memory: 7627 2022/09/07 19:07:56 - mmengine - INFO - Epoch(val) [87][310/310] acc/top1: 0.7521 acc/top5: 0.9189 acc/mean1: 0.7520 2022/09/07 19:08:15 - mmengine - INFO - Epoch(train) [88][100/3757] lr: 1.0000e-04 eta: 2:15:38 time: 0.1670 data_time: 0.0108 memory: 7627 grad_norm: 7.2237 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0837 loss: 1.0837 2022/09/07 19:08:22 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:08:31 - mmengine - INFO - Epoch(train) [88][200/3757] lr: 1.0000e-04 eta: 2:15:22 time: 0.1663 data_time: 0.0113 memory: 7124 grad_norm: 7.0662 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1985 loss: 1.1985 2022/09/07 19:08:48 - mmengine - INFO - Epoch(train) [88][300/3757] lr: 1.0000e-04 eta: 2:15:05 time: 0.1683 data_time: 0.0125 memory: 7124 grad_norm: 7.2441 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0027 loss: 1.0027 2022/09/07 19:09:05 - mmengine - INFO - Epoch(train) [88][400/3757] lr: 1.0000e-04 eta: 2:14:48 time: 0.1697 data_time: 0.0114 memory: 7124 grad_norm: 7.0657 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2427 loss: 1.2427 2022/09/07 19:09:21 - mmengine - INFO - Epoch(train) [88][500/3757] lr: 1.0000e-04 eta: 2:14:31 time: 0.1639 data_time: 0.0114 memory: 7124 grad_norm: 7.1272 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.1617 loss: 1.1617 2022/09/07 19:09:38 - mmengine - INFO - Epoch(train) [88][600/3757] lr: 1.0000e-04 eta: 2:14:15 time: 0.1629 data_time: 0.0112 memory: 7124 grad_norm: 6.9028 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.0740 loss: 1.0740 2022/09/07 19:09:55 - mmengine - INFO - Epoch(train) [88][700/3757] lr: 1.0000e-04 eta: 2:13:58 time: 0.1649 data_time: 0.0114 memory: 7124 grad_norm: 7.3065 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0221 loss: 1.0221 2022/09/07 19:10:11 - mmengine - INFO - Epoch(train) [88][800/3757] lr: 1.0000e-04 eta: 2:13:41 time: 0.1661 data_time: 0.0115 memory: 7124 grad_norm: 7.4441 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1829 loss: 1.1829 2022/09/07 19:10:28 - mmengine - INFO - Epoch(train) [88][900/3757] lr: 1.0000e-04 eta: 2:13:25 time: 0.1634 data_time: 0.0125 memory: 7124 grad_norm: 7.3782 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0129 loss: 1.0129 2022/09/07 19:10:45 - mmengine - INFO - Epoch(train) [88][1000/3757] lr: 1.0000e-04 eta: 2:13:08 time: 0.1618 data_time: 0.0111 memory: 7124 grad_norm: 6.9752 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0676 loss: 1.0676 2022/09/07 19:11:02 - mmengine - INFO - Epoch(train) [88][1100/3757] lr: 1.0000e-04 eta: 2:12:51 time: 0.1677 data_time: 0.0101 memory: 7124 grad_norm: 7.1909 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.8405 loss: 0.8405 2022/09/07 19:11:08 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:11:18 - mmengine - INFO - Epoch(train) [88][1200/3757] lr: 1.0000e-04 eta: 2:12:34 time: 0.1654 data_time: 0.0116 memory: 7124 grad_norm: 7.3906 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9830 loss: 0.9830 2022/09/07 19:11:35 - mmengine - INFO - Epoch(train) [88][1300/3757] lr: 1.0000e-04 eta: 2:12:18 time: 0.1688 data_time: 0.0127 memory: 7124 grad_norm: 7.2588 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0577 loss: 1.0577 2022/09/07 19:11:52 - mmengine - INFO - Epoch(train) [88][1400/3757] lr: 1.0000e-04 eta: 2:12:01 time: 0.1613 data_time: 0.0108 memory: 7124 grad_norm: 6.9563 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9959 loss: 0.9959 2022/09/07 19:12:09 - mmengine - INFO - Epoch(train) [88][1500/3757] lr: 1.0000e-04 eta: 2:11:44 time: 0.1661 data_time: 0.0125 memory: 7124 grad_norm: 7.1196 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0459 loss: 1.0459 2022/09/07 19:12:25 - mmengine - INFO - Epoch(train) [88][1600/3757] lr: 1.0000e-04 eta: 2:11:28 time: 0.1649 data_time: 0.0106 memory: 7124 grad_norm: 7.1498 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1559 loss: 1.1559 2022/09/07 19:12:42 - mmengine - INFO - Epoch(train) [88][1700/3757] lr: 1.0000e-04 eta: 2:11:11 time: 0.1611 data_time: 0.0120 memory: 7124 grad_norm: 7.0025 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0205 loss: 1.0205 2022/09/07 19:12:58 - mmengine - INFO - Epoch(train) [88][1800/3757] lr: 1.0000e-04 eta: 2:10:54 time: 0.1672 data_time: 0.0126 memory: 7124 grad_norm: 6.9893 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1938 loss: 1.1938 2022/09/07 19:13:15 - mmengine - INFO - Epoch(train) [88][1900/3757] lr: 1.0000e-04 eta: 2:10:37 time: 0.1672 data_time: 0.0126 memory: 7124 grad_norm: 7.2150 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1195 loss: 1.1195 2022/09/07 19:13:32 - mmengine - INFO - Epoch(train) [88][2000/3757] lr: 1.0000e-04 eta: 2:10:21 time: 0.1634 data_time: 0.0119 memory: 7124 grad_norm: 6.9711 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.7956 loss: 0.7956 2022/09/07 19:13:49 - mmengine - INFO - Epoch(train) [88][2100/3757] lr: 1.0000e-04 eta: 2:10:04 time: 0.1641 data_time: 0.0125 memory: 7124 grad_norm: 7.4207 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 0.9519 loss: 0.9519 2022/09/07 19:13:55 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:14:05 - mmengine - INFO - Epoch(train) [88][2200/3757] lr: 1.0000e-04 eta: 2:09:47 time: 0.1675 data_time: 0.0133 memory: 7124 grad_norm: 7.1888 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9627 loss: 0.9627 2022/09/07 19:14:22 - mmengine - INFO - Epoch(train) [88][2300/3757] lr: 1.0000e-04 eta: 2:09:31 time: 0.1651 data_time: 0.0117 memory: 7124 grad_norm: 6.9270 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9540 loss: 0.9540 2022/09/07 19:14:39 - mmengine - INFO - Epoch(train) [88][2400/3757] lr: 1.0000e-04 eta: 2:09:14 time: 0.1694 data_time: 0.0153 memory: 7124 grad_norm: 7.2260 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.3229 loss: 1.3229 2022/09/07 19:14:56 - mmengine - INFO - Epoch(train) [88][2500/3757] lr: 1.0000e-04 eta: 2:08:57 time: 0.1638 data_time: 0.0095 memory: 7124 grad_norm: 7.2853 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0948 loss: 1.0948 2022/09/07 19:15:12 - mmengine - INFO - Epoch(train) [88][2600/3757] lr: 1.0000e-04 eta: 2:08:40 time: 0.1622 data_time: 0.0106 memory: 7124 grad_norm: 7.5313 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1732 loss: 1.1732 2022/09/07 19:15:29 - mmengine - INFO - Epoch(train) [88][2700/3757] lr: 1.0000e-04 eta: 2:08:24 time: 0.1629 data_time: 0.0121 memory: 7124 grad_norm: 7.1718 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9144 loss: 0.9144 2022/09/07 19:15:46 - mmengine - INFO - Epoch(train) [88][2800/3757] lr: 1.0000e-04 eta: 2:08:07 time: 0.1625 data_time: 0.0123 memory: 7124 grad_norm: 6.9304 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9728 loss: 0.9728 2022/09/07 19:16:02 - mmengine - INFO - Epoch(train) [88][2900/3757] lr: 1.0000e-04 eta: 2:07:50 time: 0.1673 data_time: 0.0114 memory: 7124 grad_norm: 6.9813 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 0.9189 loss: 0.9189 2022/09/07 19:16:19 - mmengine - INFO - Epoch(train) [88][3000/3757] lr: 1.0000e-04 eta: 2:07:34 time: 0.1752 data_time: 0.0121 memory: 7124 grad_norm: 6.8721 top1_acc: 0.3750 top5_acc: 0.5000 loss_cls: 1.1378 loss: 1.1378 2022/09/07 19:16:36 - mmengine - INFO - Epoch(train) [88][3100/3757] lr: 1.0000e-04 eta: 2:07:17 time: 0.1677 data_time: 0.0109 memory: 7124 grad_norm: 6.9906 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.8718 loss: 0.8718 2022/09/07 19:16:43 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:16:53 - mmengine - INFO - Epoch(train) [88][3200/3757] lr: 1.0000e-04 eta: 2:07:00 time: 0.1656 data_time: 0.0120 memory: 7124 grad_norm: 7.2263 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0474 loss: 1.0474 2022/09/07 19:17:09 - mmengine - INFO - Epoch(train) [88][3300/3757] lr: 1.0000e-04 eta: 2:06:43 time: 0.1722 data_time: 0.0108 memory: 7124 grad_norm: 7.2962 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9632 loss: 0.9632 2022/09/07 19:17:26 - mmengine - INFO - Epoch(train) [88][3400/3757] lr: 1.0000e-04 eta: 2:06:27 time: 0.1641 data_time: 0.0108 memory: 7124 grad_norm: 7.2069 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9954 loss: 0.9954 2022/09/07 19:17:43 - mmengine - INFO - Epoch(train) [88][3500/3757] lr: 1.0000e-04 eta: 2:06:10 time: 0.1645 data_time: 0.0115 memory: 7124 grad_norm: 7.5915 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0113 loss: 1.0113 2022/09/07 19:18:00 - mmengine - INFO - Epoch(train) [88][3600/3757] lr: 1.0000e-04 eta: 2:05:53 time: 0.1634 data_time: 0.0117 memory: 7124 grad_norm: 7.2719 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9722 loss: 0.9722 2022/09/07 19:18:17 - mmengine - INFO - Epoch(train) [88][3700/3757] lr: 1.0000e-04 eta: 2:05:37 time: 0.1636 data_time: 0.0113 memory: 7124 grad_norm: 7.1981 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9561 loss: 0.9561 2022/09/07 19:18:26 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:18:26 - mmengine - INFO - Epoch(train) [88][3757/3757] lr: 1.0000e-04 eta: 2:05:30 time: 0.1408 data_time: 0.0079 memory: 7124 grad_norm: 7.0816 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.7428 loss: 0.7428 2022/09/07 19:18:26 - mmengine - INFO - Saving checkpoint at 88 epochs 2022/09/07 19:20:49 - mmengine - INFO - Epoch(val) [88][100/310] eta: 0:04:10 time: 1.1910 data_time: 0.8842 memory: 7627 2022/09/07 19:23:12 - mmengine - INFO - Epoch(val) [88][200/310] eta: 0:02:32 time: 1.3859 data_time: 1.0772 memory: 7627 2022/09/07 19:25:18 - mmengine - INFO - Epoch(val) [88][300/310] eta: 0:00:11 time: 1.1864 data_time: 0.8775 memory: 7627 2022/09/07 19:25:36 - mmengine - INFO - Epoch(val) [88][310/310] acc/top1: 0.7528 acc/top5: 0.9191 acc/mean1: 0.7527 2022/09/07 19:25:55 - mmengine - INFO - Epoch(train) [89][100/3757] lr: 1.0000e-04 eta: 2:05:10 time: 0.1631 data_time: 0.0101 memory: 7627 grad_norm: 7.3944 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1076 loss: 1.1076 2022/09/07 19:26:12 - mmengine - INFO - Epoch(train) [89][200/3757] lr: 1.0000e-04 eta: 2:04:54 time: 0.1807 data_time: 0.0111 memory: 7124 grad_norm: 6.9780 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.0091 loss: 1.0091 2022/09/07 19:26:29 - mmengine - INFO - Epoch(train) [89][300/3757] lr: 1.0000e-04 eta: 2:04:37 time: 0.1712 data_time: 0.0109 memory: 7124 grad_norm: 6.9113 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2650 loss: 1.2650 2022/09/07 19:26:43 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:26:46 - mmengine - INFO - Epoch(train) [89][400/3757] lr: 1.0000e-04 eta: 2:04:20 time: 0.1649 data_time: 0.0118 memory: 7124 grad_norm: 7.4096 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0917 loss: 1.0917 2022/09/07 19:27:02 - mmengine - INFO - Epoch(train) [89][500/3757] lr: 1.0000e-04 eta: 2:04:04 time: 0.1653 data_time: 0.0105 memory: 7124 grad_norm: 7.5084 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9776 loss: 0.9776 2022/09/07 19:27:19 - mmengine - INFO - Epoch(train) [89][600/3757] lr: 1.0000e-04 eta: 2:03:47 time: 0.1686 data_time: 0.0108 memory: 7124 grad_norm: 7.1343 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9096 loss: 0.9096 2022/09/07 19:27:36 - mmengine - INFO - Epoch(train) [89][700/3757] lr: 1.0000e-04 eta: 2:03:30 time: 0.1631 data_time: 0.0116 memory: 7124 grad_norm: 7.0650 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8984 loss: 0.8984 2022/09/07 19:27:52 - mmengine - INFO - Epoch(train) [89][800/3757] lr: 1.0000e-04 eta: 2:03:13 time: 0.1618 data_time: 0.0110 memory: 7124 grad_norm: 7.5852 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.2061 loss: 1.2061 2022/09/07 19:28:09 - mmengine - INFO - Epoch(train) [89][900/3757] lr: 1.0000e-04 eta: 2:02:57 time: 0.1660 data_time: 0.0114 memory: 7124 grad_norm: 7.5548 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1494 loss: 1.1494 2022/09/07 19:28:26 - mmengine - INFO - Epoch(train) [89][1000/3757] lr: 1.0000e-04 eta: 2:02:40 time: 0.1695 data_time: 0.0122 memory: 7124 grad_norm: 7.2444 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0099 loss: 1.0099 2022/09/07 19:28:43 - mmengine - INFO - Epoch(train) [89][1100/3757] lr: 1.0000e-04 eta: 2:02:23 time: 0.1626 data_time: 0.0114 memory: 7124 grad_norm: 7.0331 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0086 loss: 1.0086 2022/09/07 19:28:59 - mmengine - INFO - Epoch(train) [89][1200/3757] lr: 1.0000e-04 eta: 2:02:07 time: 0.1665 data_time: 0.0114 memory: 7124 grad_norm: 7.1717 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8977 loss: 0.8977 2022/09/07 19:29:16 - mmengine - INFO - Epoch(train) [89][1300/3757] lr: 1.0000e-04 eta: 2:01:50 time: 0.1633 data_time: 0.0106 memory: 7124 grad_norm: 7.2352 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9638 loss: 0.9638 2022/09/07 19:29:30 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:29:33 - mmengine - INFO - Epoch(train) [89][1400/3757] lr: 1.0000e-04 eta: 2:01:33 time: 0.1667 data_time: 0.0114 memory: 7124 grad_norm: 7.1767 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.9642 loss: 0.9642 2022/09/07 19:29:50 - mmengine - INFO - Epoch(train) [89][1500/3757] lr: 1.0000e-04 eta: 2:01:17 time: 0.1846 data_time: 0.0121 memory: 7124 grad_norm: 7.2115 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1577 loss: 1.1577 2022/09/07 19:30:06 - mmengine - INFO - Epoch(train) [89][1600/3757] lr: 1.0000e-04 eta: 2:01:00 time: 0.1690 data_time: 0.0110 memory: 7124 grad_norm: 7.0491 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8954 loss: 0.8954 2022/09/07 19:30:23 - mmengine - INFO - Epoch(train) [89][1700/3757] lr: 1.0000e-04 eta: 2:00:43 time: 0.1631 data_time: 0.0111 memory: 7124 grad_norm: 7.3326 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8112 loss: 0.8112 2022/09/07 19:30:40 - mmengine - INFO - Epoch(train) [89][1800/3757] lr: 1.0000e-04 eta: 2:00:26 time: 0.1714 data_time: 0.0164 memory: 7124 grad_norm: 6.9447 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.2405 loss: 1.2405 2022/09/07 19:30:57 - mmengine - INFO - Epoch(train) [89][1900/3757] lr: 1.0000e-04 eta: 2:00:10 time: 0.1661 data_time: 0.0116 memory: 7124 grad_norm: 7.3229 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1828 loss: 1.1828 2022/09/07 19:31:13 - mmengine - INFO - Epoch(train) [89][2000/3757] lr: 1.0000e-04 eta: 1:59:53 time: 0.1626 data_time: 0.0117 memory: 7124 grad_norm: 7.0568 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1273 loss: 1.1273 2022/09/07 19:31:30 - mmengine - INFO - Epoch(train) [89][2100/3757] lr: 1.0000e-04 eta: 1:59:36 time: 0.1607 data_time: 0.0110 memory: 7124 grad_norm: 7.2888 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9524 loss: 0.9524 2022/09/07 19:31:47 - mmengine - INFO - Epoch(train) [89][2200/3757] lr: 1.0000e-04 eta: 1:59:20 time: 0.1595 data_time: 0.0119 memory: 7124 grad_norm: 7.2806 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.9428 loss: 0.9428 2022/09/07 19:32:03 - mmengine - INFO - Epoch(train) [89][2300/3757] lr: 1.0000e-04 eta: 1:59:03 time: 0.1661 data_time: 0.0125 memory: 7124 grad_norm: 7.3815 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0605 loss: 1.0605 2022/09/07 19:32:17 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:32:20 - mmengine - INFO - Epoch(train) [89][2400/3757] lr: 1.0000e-04 eta: 1:58:46 time: 0.1680 data_time: 0.0120 memory: 7124 grad_norm: 7.2630 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0013 loss: 1.0013 2022/09/07 19:32:36 - mmengine - INFO - Epoch(train) [89][2500/3757] lr: 1.0000e-04 eta: 1:58:29 time: 0.1620 data_time: 0.0109 memory: 7124 grad_norm: 7.2322 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0281 loss: 1.0281 2022/09/07 19:32:53 - mmengine - INFO - Epoch(train) [89][2600/3757] lr: 1.0000e-04 eta: 1:58:13 time: 0.1609 data_time: 0.0119 memory: 7124 grad_norm: 7.2925 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9818 loss: 0.9818 2022/09/07 19:33:10 - mmengine - INFO - Epoch(train) [89][2700/3757] lr: 1.0000e-04 eta: 1:57:56 time: 0.1737 data_time: 0.0105 memory: 7124 grad_norm: 7.1680 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9263 loss: 0.9263 2022/09/07 19:33:26 - mmengine - INFO - Epoch(train) [89][2800/3757] lr: 1.0000e-04 eta: 1:57:39 time: 0.1636 data_time: 0.0121 memory: 7124 grad_norm: 7.2397 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 0.9547 loss: 0.9547 2022/09/07 19:33:43 - mmengine - INFO - Epoch(train) [89][2900/3757] lr: 1.0000e-04 eta: 1:57:23 time: 0.1620 data_time: 0.0111 memory: 7124 grad_norm: 7.4338 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0870 loss: 1.0870 2022/09/07 19:34:00 - mmengine - INFO - Epoch(train) [89][3000/3757] lr: 1.0000e-04 eta: 1:57:06 time: 0.1629 data_time: 0.0109 memory: 7124 grad_norm: 7.2472 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1209 loss: 1.1209 2022/09/07 19:34:17 - mmengine - INFO - Epoch(train) [89][3100/3757] lr: 1.0000e-04 eta: 1:56:49 time: 0.1669 data_time: 0.0123 memory: 7124 grad_norm: 7.0655 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0691 loss: 1.0691 2022/09/07 19:34:34 - mmengine - INFO - Epoch(train) [89][3200/3757] lr: 1.0000e-04 eta: 1:56:32 time: 0.1677 data_time: 0.0120 memory: 7124 grad_norm: 7.2172 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2261 loss: 1.2261 2022/09/07 19:34:50 - mmengine - INFO - Epoch(train) [89][3300/3757] lr: 1.0000e-04 eta: 1:56:16 time: 0.1720 data_time: 0.0207 memory: 7124 grad_norm: 6.8949 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.3209 loss: 1.3209 2022/09/07 19:35:04 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:35:07 - mmengine - INFO - Epoch(train) [89][3400/3757] lr: 1.0000e-04 eta: 1:55:59 time: 0.1674 data_time: 0.0119 memory: 7124 grad_norm: 7.2288 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0386 loss: 1.0386 2022/09/07 19:35:24 - mmengine - INFO - Epoch(train) [89][3500/3757] lr: 1.0000e-04 eta: 1:55:42 time: 0.1657 data_time: 0.0123 memory: 7124 grad_norm: 6.9713 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9657 loss: 0.9657 2022/09/07 19:35:41 - mmengine - INFO - Epoch(train) [89][3600/3757] lr: 1.0000e-04 eta: 1:55:26 time: 0.1695 data_time: 0.0181 memory: 7124 grad_norm: 6.8954 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0171 loss: 1.0171 2022/09/07 19:35:57 - mmengine - INFO - Epoch(train) [89][3700/3757] lr: 1.0000e-04 eta: 1:55:09 time: 0.1630 data_time: 0.0110 memory: 7124 grad_norm: 7.4006 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2283 loss: 1.2283 2022/09/07 19:36:07 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:36:07 - mmengine - INFO - Epoch(train) [89][3757/3757] lr: 1.0000e-04 eta: 1:55:02 time: 0.1401 data_time: 0.0077 memory: 7124 grad_norm: 7.3198 top1_acc: 0.5714 top5_acc: 0.8571 loss_cls: 1.0310 loss: 1.0310 2022/09/07 19:36:07 - mmengine - INFO - Saving checkpoint at 89 epochs 2022/09/07 19:38:32 - mmengine - INFO - Epoch(val) [89][100/310] eta: 0:04:25 time: 1.2639 data_time: 0.9591 memory: 7627 2022/09/07 19:40:52 - mmengine - INFO - Epoch(val) [89][200/310] eta: 0:02:13 time: 1.2152 data_time: 0.9118 memory: 7627 2022/09/07 19:43:02 - mmengine - INFO - Epoch(val) [89][300/310] eta: 0:00:12 time: 1.2352 data_time: 0.9321 memory: 7627 2022/09/07 19:43:19 - mmengine - INFO - Epoch(val) [89][310/310] acc/top1: 0.7538 acc/top5: 0.9185 acc/mean1: 0.7537 2022/09/07 19:43:38 - mmengine - INFO - Epoch(train) [90][100/3757] lr: 1.0000e-04 eta: 1:54:43 time: 0.1603 data_time: 0.0108 memory: 7627 grad_norm: 7.2838 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2217 loss: 1.2217 2022/09/07 19:43:54 - mmengine - INFO - Epoch(train) [90][200/3757] lr: 1.0000e-04 eta: 1:54:26 time: 0.1623 data_time: 0.0116 memory: 7124 grad_norm: 7.5320 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1525 loss: 1.1525 2022/09/07 19:44:11 - mmengine - INFO - Epoch(train) [90][300/3757] lr: 1.0000e-04 eta: 1:54:09 time: 0.1622 data_time: 0.0122 memory: 7124 grad_norm: 7.4416 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0415 loss: 1.0415 2022/09/07 19:44:27 - mmengine - INFO - Epoch(train) [90][400/3757] lr: 1.0000e-04 eta: 1:53:52 time: 0.1664 data_time: 0.0126 memory: 7124 grad_norm: 7.5117 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0172 loss: 1.0172 2022/09/07 19:44:44 - mmengine - INFO - Epoch(train) [90][500/3757] lr: 1.0000e-04 eta: 1:53:36 time: 0.1633 data_time: 0.0123 memory: 7124 grad_norm: 7.3412 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 0.9726 loss: 0.9726 2022/09/07 19:45:00 - mmengine - INFO - Epoch(train) [90][600/3757] lr: 1.0000e-04 eta: 1:53:19 time: 0.1646 data_time: 0.0115 memory: 7124 grad_norm: 7.5288 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8609 loss: 0.8609 2022/09/07 19:45:05 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:45:17 - mmengine - INFO - Epoch(train) [90][700/3757] lr: 1.0000e-04 eta: 1:53:02 time: 0.1687 data_time: 0.0110 memory: 7124 grad_norm: 7.2628 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9852 loss: 0.9852 2022/09/07 19:45:33 - mmengine - INFO - Epoch(train) [90][800/3757] lr: 1.0000e-04 eta: 1:52:45 time: 0.1606 data_time: 0.0118 memory: 7124 grad_norm: 7.3875 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1010 loss: 1.1010 2022/09/07 19:45:50 - mmengine - INFO - Epoch(train) [90][900/3757] lr: 1.0000e-04 eta: 1:52:29 time: 0.1629 data_time: 0.0111 memory: 7124 grad_norm: 6.8762 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.7548 loss: 0.7548 2022/09/07 19:46:06 - mmengine - INFO - Epoch(train) [90][1000/3757] lr: 1.0000e-04 eta: 1:52:12 time: 0.1679 data_time: 0.0115 memory: 7124 grad_norm: 7.4202 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0967 loss: 1.0967 2022/09/07 19:46:23 - mmengine - INFO - Epoch(train) [90][1100/3757] lr: 1.0000e-04 eta: 1:51:55 time: 0.1658 data_time: 0.0135 memory: 7124 grad_norm: 7.2952 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8502 loss: 0.8502 2022/09/07 19:46:39 - mmengine - INFO - Epoch(train) [90][1200/3757] lr: 1.0000e-04 eta: 1:51:39 time: 0.1603 data_time: 0.0113 memory: 7124 grad_norm: 7.2311 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0171 loss: 1.0171 2022/09/07 19:46:56 - mmengine - INFO - Epoch(train) [90][1300/3757] lr: 1.0000e-04 eta: 1:51:22 time: 0.1621 data_time: 0.0101 memory: 7124 grad_norm: 7.0641 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0863 loss: 1.0863 2022/09/07 19:47:12 - mmengine - INFO - Epoch(train) [90][1400/3757] lr: 1.0000e-04 eta: 1:51:05 time: 0.1631 data_time: 0.0116 memory: 7124 grad_norm: 7.0895 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0103 loss: 1.0103 2022/09/07 19:47:29 - mmengine - INFO - Epoch(train) [90][1500/3757] lr: 1.0000e-04 eta: 1:50:48 time: 0.1657 data_time: 0.0109 memory: 7124 grad_norm: 7.3232 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0812 loss: 1.0812 2022/09/07 19:47:46 - mmengine - INFO - Epoch(train) [90][1600/3757] lr: 1.0000e-04 eta: 1:50:32 time: 0.1637 data_time: 0.0115 memory: 7124 grad_norm: 7.1885 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1551 loss: 1.1551 2022/09/07 19:47:50 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:48:02 - mmengine - INFO - Epoch(train) [90][1700/3757] lr: 1.0000e-04 eta: 1:50:15 time: 0.1657 data_time: 0.0123 memory: 7124 grad_norm: 7.1992 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0896 loss: 1.0896 2022/09/07 19:48:19 - mmengine - INFO - Epoch(train) [90][1800/3757] lr: 1.0000e-04 eta: 1:49:58 time: 0.1618 data_time: 0.0113 memory: 7124 grad_norm: 7.1974 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1814 loss: 1.1814 2022/09/07 19:48:35 - mmengine - INFO - Epoch(train) [90][1900/3757] lr: 1.0000e-04 eta: 1:49:41 time: 0.1608 data_time: 0.0112 memory: 7124 grad_norm: 7.1022 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9812 loss: 0.9812 2022/09/07 19:48:51 - mmengine - INFO - Epoch(train) [90][2000/3757] lr: 1.0000e-04 eta: 1:49:25 time: 0.1639 data_time: 0.0113 memory: 7124 grad_norm: 7.2258 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.8842 loss: 0.8842 2022/09/07 19:49:08 - mmengine - INFO - Epoch(train) [90][2100/3757] lr: 1.0000e-04 eta: 1:49:08 time: 0.1624 data_time: 0.0110 memory: 7124 grad_norm: 7.1465 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8610 loss: 0.8610 2022/09/07 19:49:24 - mmengine - INFO - Epoch(train) [90][2200/3757] lr: 1.0000e-04 eta: 1:48:51 time: 0.1618 data_time: 0.0110 memory: 7124 grad_norm: 7.0842 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0008 loss: 1.0008 2022/09/07 19:49:41 - mmengine - INFO - Epoch(train) [90][2300/3757] lr: 1.0000e-04 eta: 1:48:34 time: 0.1710 data_time: 0.0114 memory: 7124 grad_norm: 7.4465 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1080 loss: 1.1080 2022/09/07 19:49:57 - mmengine - INFO - Epoch(train) [90][2400/3757] lr: 1.0000e-04 eta: 1:48:18 time: 0.1619 data_time: 0.0110 memory: 7124 grad_norm: 7.2101 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9203 loss: 0.9203 2022/09/07 19:50:14 - mmengine - INFO - Epoch(train) [90][2500/3757] lr: 1.0000e-04 eta: 1:48:01 time: 0.1629 data_time: 0.0140 memory: 7124 grad_norm: 7.5248 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0618 loss: 1.0618 2022/09/07 19:50:30 - mmengine - INFO - Epoch(train) [90][2600/3757] lr: 1.0000e-04 eta: 1:47:44 time: 0.1614 data_time: 0.0110 memory: 7124 grad_norm: 7.4998 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1090 loss: 1.1090 2022/09/07 19:50:34 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:50:46 - mmengine - INFO - Epoch(train) [90][2700/3757] lr: 1.0000e-04 eta: 1:47:28 time: 0.1692 data_time: 0.0106 memory: 7124 grad_norm: 7.4287 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.0422 loss: 1.0422 2022/09/07 19:51:03 - mmengine - INFO - Epoch(train) [90][2800/3757] lr: 1.0000e-04 eta: 1:47:11 time: 0.1680 data_time: 0.0125 memory: 7124 grad_norm: 7.2923 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9599 loss: 0.9599 2022/09/07 19:51:19 - mmengine - INFO - Epoch(train) [90][2900/3757] lr: 1.0000e-04 eta: 1:46:54 time: 0.1624 data_time: 0.0118 memory: 7124 grad_norm: 7.0379 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1826 loss: 1.1826 2022/09/07 19:51:36 - mmengine - INFO - Epoch(train) [90][3000/3757] lr: 1.0000e-04 eta: 1:46:37 time: 0.1610 data_time: 0.0114 memory: 7124 grad_norm: 7.1521 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8615 loss: 0.8615 2022/09/07 19:51:53 - mmengine - INFO - Epoch(train) [90][3100/3757] lr: 1.0000e-04 eta: 1:46:21 time: 0.1687 data_time: 0.0124 memory: 7124 grad_norm: 7.5901 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1879 loss: 1.1879 2022/09/07 19:52:09 - mmengine - INFO - Epoch(train) [90][3200/3757] lr: 1.0000e-04 eta: 1:46:04 time: 0.1657 data_time: 0.0112 memory: 7124 grad_norm: 7.3342 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0467 loss: 1.0467 2022/09/07 19:52:26 - mmengine - INFO - Epoch(train) [90][3300/3757] lr: 1.0000e-04 eta: 1:45:47 time: 0.1655 data_time: 0.0107 memory: 7124 grad_norm: 7.2513 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.0838 loss: 1.0838 2022/09/07 19:52:43 - mmengine - INFO - Epoch(train) [90][3400/3757] lr: 1.0000e-04 eta: 1:45:31 time: 0.1653 data_time: 0.0125 memory: 7124 grad_norm: 7.1267 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0432 loss: 1.0432 2022/09/07 19:53:00 - mmengine - INFO - Epoch(train) [90][3500/3757] lr: 1.0000e-04 eta: 1:45:14 time: 0.1625 data_time: 0.0111 memory: 7124 grad_norm: 7.1208 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0923 loss: 1.0923 2022/09/07 19:53:17 - mmengine - INFO - Epoch(train) [90][3600/3757] lr: 1.0000e-04 eta: 1:44:57 time: 0.1622 data_time: 0.0110 memory: 7124 grad_norm: 7.2121 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8893 loss: 0.8893 2022/09/07 19:53:21 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:53:34 - mmengine - INFO - Epoch(train) [90][3700/3757] lr: 1.0000e-04 eta: 1:44:40 time: 0.1810 data_time: 0.0270 memory: 7124 grad_norm: 7.0177 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8953 loss: 0.8953 2022/09/07 19:53:43 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 19:53:43 - mmengine - INFO - Epoch(train) [90][3757/3757] lr: 1.0000e-04 eta: 1:44:34 time: 0.1422 data_time: 0.0077 memory: 7124 grad_norm: 7.1204 top1_acc: 0.7143 top5_acc: 1.0000 loss_cls: 0.9364 loss: 0.9364 2022/09/07 19:53:43 - mmengine - INFO - Saving checkpoint at 90 epochs 2022/09/07 19:56:06 - mmengine - INFO - Epoch(val) [90][100/310] eta: 0:04:32 time: 1.2961 data_time: 0.9909 memory: 7627 2022/09/07 19:58:23 - mmengine - INFO - Epoch(val) [90][200/310] eta: 0:02:22 time: 1.2949 data_time: 0.9894 memory: 7627 2022/09/07 20:00:35 - mmengine - INFO - Epoch(val) [90][300/310] eta: 0:00:13 time: 1.3747 data_time: 1.0711 memory: 7627 2022/09/07 20:00:56 - mmengine - INFO - Epoch(val) [90][310/310] acc/top1: 0.7544 acc/top5: 0.9199 acc/mean1: 0.7543 2022/09/07 20:00:57 - mmengine - INFO - The previous best checkpoint /mnt/lustre/daiwenxun/X/mmaction2/work_dirs/tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb/best_acc/top1_epoch_86.pth is removed 2022/09/07 20:00:59 - mmengine - INFO - The best checkpoint with 0.7544 acc/top1 at 90 epoch is saved to best_acc/top1_epoch_90.pth. 2022/09/07 20:01:17 - mmengine - INFO - Epoch(train) [91][100/3757] lr: 1.0000e-05 eta: 1:44:14 time: 0.1657 data_time: 0.0109 memory: 7627 grad_norm: 7.0230 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1161 loss: 1.1161 2022/09/07 20:01:34 - mmengine - INFO - Epoch(train) [91][200/3757] lr: 1.0000e-05 eta: 1:43:57 time: 0.1722 data_time: 0.0116 memory: 7124 grad_norm: 7.2899 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1234 loss: 1.1234 2022/09/07 20:01:51 - mmengine - INFO - Epoch(train) [91][300/3757] lr: 1.0000e-05 eta: 1:43:41 time: 0.1786 data_time: 0.0117 memory: 7124 grad_norm: 7.2369 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 0.8090 loss: 0.8090 2022/09/07 20:02:07 - mmengine - INFO - Epoch(train) [91][400/3757] lr: 1.0000e-05 eta: 1:43:24 time: 0.1632 data_time: 0.0112 memory: 7124 grad_norm: 7.1499 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0231 loss: 1.0231 2022/09/07 20:02:24 - mmengine - INFO - Epoch(train) [91][500/3757] lr: 1.0000e-05 eta: 1:43:07 time: 0.1605 data_time: 0.0106 memory: 7124 grad_norm: 7.0712 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8501 loss: 0.8501 2022/09/07 20:02:40 - mmengine - INFO - Epoch(train) [91][600/3757] lr: 1.0000e-05 eta: 1:42:51 time: 0.1579 data_time: 0.0106 memory: 7124 grad_norm: 7.3773 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.1654 loss: 1.1654 2022/09/07 20:02:57 - mmengine - INFO - Epoch(train) [91][700/3757] lr: 1.0000e-05 eta: 1:42:34 time: 0.1632 data_time: 0.0124 memory: 7124 grad_norm: 7.3795 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0525 loss: 1.0525 2022/09/07 20:03:14 - mmengine - INFO - Epoch(train) [91][800/3757] lr: 1.0000e-05 eta: 1:42:17 time: 0.1761 data_time: 0.0112 memory: 7124 grad_norm: 7.1204 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1754 loss: 1.1754 2022/09/07 20:03:25 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 20:03:30 - mmengine - INFO - Epoch(train) [91][900/3757] lr: 1.0000e-05 eta: 1:42:00 time: 0.1617 data_time: 0.0115 memory: 7124 grad_norm: 7.1204 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0167 loss: 1.0167 2022/09/07 20:03:47 - mmengine - INFO - Epoch(train) [91][1000/3757] lr: 1.0000e-05 eta: 1:41:44 time: 0.1629 data_time: 0.0123 memory: 7124 grad_norm: 7.2264 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2904 loss: 1.2904 2022/09/07 20:04:04 - mmengine - INFO - Epoch(train) [91][1100/3757] lr: 1.0000e-05 eta: 1:41:27 time: 0.1692 data_time: 0.0125 memory: 7124 grad_norm: 7.3027 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0797 loss: 1.0797 2022/09/07 20:04:20 - mmengine - INFO - Epoch(train) [91][1200/3757] lr: 1.0000e-05 eta: 1:41:10 time: 0.1694 data_time: 0.0118 memory: 7124 grad_norm: 7.1963 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0004 loss: 1.0004 2022/09/07 20:04:37 - mmengine - INFO - Epoch(train) [91][1300/3757] lr: 1.0000e-05 eta: 1:40:54 time: 0.1636 data_time: 0.0108 memory: 7124 grad_norm: 7.4570 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1580 loss: 1.1580 2022/09/07 20:04:53 - mmengine - INFO - Epoch(train) [91][1400/3757] lr: 1.0000e-05 eta: 1:40:37 time: 0.1648 data_time: 0.0121 memory: 7124 grad_norm: 7.1559 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0579 loss: 1.0579 2022/09/07 20:05:10 - mmengine - INFO - Epoch(train) [91][1500/3757] lr: 1.0000e-05 eta: 1:40:20 time: 0.1664 data_time: 0.0106 memory: 7124 grad_norm: 7.2592 top1_acc: 0.2500 top5_acc: 1.0000 loss_cls: 1.3927 loss: 1.3927 2022/09/07 20:05:27 - mmengine - INFO - Epoch(train) [91][1600/3757] lr: 1.0000e-05 eta: 1:40:03 time: 0.1663 data_time: 0.0115 memory: 7124 grad_norm: 7.1699 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8626 loss: 0.8626 2022/09/07 20:05:44 - mmengine - INFO - Epoch(train) [91][1700/3757] lr: 1.0000e-05 eta: 1:39:47 time: 0.1643 data_time: 0.0100 memory: 7124 grad_norm: 7.2177 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9287 loss: 0.9287 2022/09/07 20:06:00 - mmengine - INFO - Epoch(train) [91][1800/3757] lr: 1.0000e-05 eta: 1:39:30 time: 0.1631 data_time: 0.0125 memory: 7124 grad_norm: 7.1266 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.1106 loss: 1.1106 2022/09/07 20:06:12 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 20:06:17 - mmengine - INFO - Epoch(train) [91][1900/3757] lr: 1.0000e-05 eta: 1:39:13 time: 0.1609 data_time: 0.0121 memory: 7124 grad_norm: 7.2624 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1122 loss: 1.1122 2022/09/07 20:06:34 - mmengine - INFO - Epoch(train) [91][2000/3757] lr: 1.0000e-05 eta: 1:38:57 time: 0.1708 data_time: 0.0117 memory: 7124 grad_norm: 7.3963 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2855 loss: 1.2855 2022/09/07 20:06:50 - mmengine - INFO - Epoch(train) [91][2100/3757] lr: 1.0000e-05 eta: 1:38:40 time: 0.1670 data_time: 0.0108 memory: 7124 grad_norm: 7.1284 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1778 loss: 1.1778 2022/09/07 20:07:07 - mmengine - INFO - Epoch(train) [91][2200/3757] lr: 1.0000e-05 eta: 1:38:23 time: 0.1654 data_time: 0.0130 memory: 7124 grad_norm: 7.1159 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8400 loss: 0.8400 2022/09/07 20:07:24 - mmengine - INFO - Epoch(train) [91][2300/3757] lr: 1.0000e-05 eta: 1:38:07 time: 0.1695 data_time: 0.0116 memory: 7124 grad_norm: 7.1324 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.7731 loss: 0.7731 2022/09/07 20:07:41 - mmengine - INFO - Epoch(train) [91][2400/3757] lr: 1.0000e-05 eta: 1:37:50 time: 0.1673 data_time: 0.0132 memory: 7124 grad_norm: 7.0127 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8996 loss: 0.8996 2022/09/07 20:07:58 - mmengine - INFO - Epoch(train) [91][2500/3757] lr: 1.0000e-05 eta: 1:37:33 time: 0.1626 data_time: 0.0119 memory: 7124 grad_norm: 7.1570 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9823 loss: 0.9823 2022/09/07 20:08:14 - mmengine - INFO - Epoch(train) [91][2600/3757] lr: 1.0000e-05 eta: 1:37:16 time: 0.1674 data_time: 0.0116 memory: 7124 grad_norm: 6.9289 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8373 loss: 0.8373 2022/09/07 20:08:31 - mmengine - INFO - Epoch(train) [91][2700/3757] lr: 1.0000e-05 eta: 1:37:00 time: 0.1618 data_time: 0.0111 memory: 7124 grad_norm: 7.4024 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0840 loss: 1.0840 2022/09/07 20:08:47 - mmengine - INFO - Epoch(train) [91][2800/3757] lr: 1.0000e-05 eta: 1:36:43 time: 0.1631 data_time: 0.0111 memory: 7124 grad_norm: 6.8615 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9495 loss: 0.9495 2022/09/07 20:08:59 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 20:09:04 - mmengine - INFO - Epoch(train) [91][2900/3757] lr: 1.0000e-05 eta: 1:36:26 time: 0.1671 data_time: 0.0115 memory: 7124 grad_norm: 7.1293 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.9007 loss: 0.9007 2022/09/07 20:09:21 - mmengine - INFO - Epoch(train) [91][3000/3757] lr: 1.0000e-05 eta: 1:36:10 time: 0.1651 data_time: 0.0120 memory: 7124 grad_norm: 7.3304 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9642 loss: 0.9642 2022/09/07 20:09:38 - mmengine - INFO - Epoch(train) [91][3100/3757] lr: 1.0000e-05 eta: 1:35:53 time: 0.1662 data_time: 0.0114 memory: 7124 grad_norm: 7.2111 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8930 loss: 0.8930 2022/09/07 20:09:55 - mmengine - INFO - Epoch(train) [91][3200/3757] lr: 1.0000e-05 eta: 1:35:36 time: 0.1641 data_time: 0.0124 memory: 7124 grad_norm: 7.2370 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0720 loss: 1.0720 2022/09/07 20:10:12 - mmengine - INFO - Epoch(train) [91][3300/3757] lr: 1.0000e-05 eta: 1:35:20 time: 0.1625 data_time: 0.0112 memory: 7124 grad_norm: 7.0454 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8663 loss: 0.8663 2022/09/07 20:10:28 - mmengine - INFO - Epoch(train) [91][3400/3757] lr: 1.0000e-05 eta: 1:35:03 time: 0.1623 data_time: 0.0111 memory: 7124 grad_norm: 7.2146 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1051 loss: 1.1051 2022/09/07 20:10:45 - mmengine - INFO - Epoch(train) [91][3500/3757] lr: 1.0000e-05 eta: 1:34:46 time: 0.1730 data_time: 0.0158 memory: 7124 grad_norm: 7.0009 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9572 loss: 0.9572 2022/09/07 20:11:02 - mmengine - INFO - Epoch(train) [91][3600/3757] lr: 1.0000e-05 eta: 1:34:29 time: 0.1611 data_time: 0.0103 memory: 7124 grad_norm: 7.2728 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.9772 loss: 0.9772 2022/09/07 20:11:19 - mmengine - INFO - Epoch(train) [91][3700/3757] lr: 1.0000e-05 eta: 1:34:13 time: 0.1638 data_time: 0.0113 memory: 7124 grad_norm: 7.2826 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2265 loss: 1.2265 2022/09/07 20:11:27 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 20:11:27 - mmengine - INFO - Epoch(train) [91][3757/3757] lr: 1.0000e-05 eta: 1:34:06 time: 0.1385 data_time: 0.0067 memory: 7124 grad_norm: 6.9873 top1_acc: 0.7143 top5_acc: 1.0000 loss_cls: 0.8185 loss: 0.8185 2022/09/07 20:11:27 - mmengine - INFO - Saving checkpoint at 91 epochs 2022/09/07 20:13:54 - mmengine - INFO - Epoch(val) [91][100/310] eta: 0:04:38 time: 1.3251 data_time: 1.0191 memory: 7627 2022/09/07 20:16:10 - mmengine - INFO - Epoch(val) [91][200/310] eta: 0:02:06 time: 1.1543 data_time: 0.8473 memory: 7627 2022/09/07 20:18:19 - mmengine - INFO - Epoch(val) [91][300/310] eta: 0:00:13 time: 1.3066 data_time: 0.9807 memory: 7627 2022/09/07 20:18:40 - mmengine - INFO - Epoch(val) [91][310/310] acc/top1: 0.7536 acc/top5: 0.9194 acc/mean1: 0.7535 2022/09/07 20:19:05 - mmengine - INFO - Epoch(train) [92][100/3757] lr: 1.0000e-05 eta: 1:33:47 time: 0.2279 data_time: 0.0152 memory: 7627 grad_norm: 7.2830 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.1515 loss: 1.1515 2022/09/07 20:19:08 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 20:19:27 - mmengine - INFO - Epoch(train) [92][200/3757] lr: 1.0000e-05 eta: 1:33:31 time: 0.2218 data_time: 0.0158 memory: 7124 grad_norm: 7.1058 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0625 loss: 1.0625 2022/09/07 20:19:49 - mmengine - INFO - Epoch(train) [92][300/3757] lr: 1.0000e-05 eta: 1:33:15 time: 0.2147 data_time: 0.0150 memory: 7124 grad_norm: 7.1877 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0512 loss: 1.0512 2022/09/07 20:20:10 - mmengine - INFO - Epoch(train) [92][400/3757] lr: 1.0000e-05 eta: 1:32:58 time: 0.2043 data_time: 0.0144 memory: 7124 grad_norm: 7.4353 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.2083 loss: 1.2083 2022/09/07 20:20:31 - mmengine - INFO - Epoch(train) [92][500/3757] lr: 1.0000e-05 eta: 1:32:42 time: 0.2168 data_time: 0.0131 memory: 7124 grad_norm: 7.3049 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1113 loss: 1.1113 2022/09/07 20:20:53 - mmengine - INFO - Epoch(train) [92][600/3757] lr: 1.0000e-05 eta: 1:32:26 time: 0.2072 data_time: 0.0150 memory: 7124 grad_norm: 6.8600 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8201 loss: 0.8201 2022/09/07 20:21:14 - mmengine - INFO - Epoch(train) [92][700/3757] lr: 1.0000e-05 eta: 1:32:10 time: 0.2108 data_time: 0.0159 memory: 7124 grad_norm: 6.9725 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9699 loss: 0.9699 2022/09/07 20:21:35 - mmengine - INFO - Epoch(train) [92][800/3757] lr: 1.0000e-05 eta: 1:31:53 time: 0.2098 data_time: 0.0146 memory: 7124 grad_norm: 7.1012 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0183 loss: 1.0183 2022/09/07 20:21:56 - mmengine - INFO - Epoch(train) [92][900/3757] lr: 1.0000e-05 eta: 1:31:37 time: 0.2198 data_time: 0.0141 memory: 7124 grad_norm: 7.2213 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8790 loss: 0.8790 2022/09/07 20:22:18 - mmengine - INFO - Epoch(train) [92][1000/3757] lr: 1.0000e-05 eta: 1:31:21 time: 0.2163 data_time: 0.0387 memory: 7124 grad_norm: 7.3350 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8994 loss: 0.8994 2022/09/07 20:22:39 - mmengine - INFO - Epoch(train) [92][1100/3757] lr: 1.0000e-05 eta: 1:31:04 time: 0.2066 data_time: 0.0158 memory: 7124 grad_norm: 7.4730 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9988 loss: 0.9988 2022/09/07 20:22:42 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 20:23:00 - mmengine - INFO - Epoch(train) [92][1200/3757] lr: 1.0000e-05 eta: 1:30:48 time: 0.2015 data_time: 0.0150 memory: 7124 grad_norm: 7.0874 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.1126 loss: 1.1126 2022/09/07 20:23:21 - mmengine - INFO - Epoch(train) [92][1300/3757] lr: 1.0000e-05 eta: 1:30:32 time: 0.2107 data_time: 0.0148 memory: 7124 grad_norm: 7.0714 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.7858 loss: 0.7858 2022/09/07 20:23:42 - mmengine - INFO - Epoch(train) [92][1400/3757] lr: 1.0000e-05 eta: 1:30:16 time: 0.2068 data_time: 0.0143 memory: 7124 grad_norm: 7.2962 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0728 loss: 1.0728 2022/09/07 20:24:03 - mmengine - INFO - Epoch(train) [92][1500/3757] lr: 1.0000e-05 eta: 1:29:59 time: 0.2025 data_time: 0.0138 memory: 7124 grad_norm: 7.2884 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9301 loss: 0.9301 2022/09/07 20:24:25 - mmengine - INFO - Epoch(train) [92][1600/3757] lr: 1.0000e-05 eta: 1:29:43 time: 0.2210 data_time: 0.0154 memory: 7124 grad_norm: 7.1430 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9268 loss: 0.9268 2022/09/07 20:24:45 - mmengine - INFO - Epoch(train) [92][1700/3757] lr: 1.0000e-05 eta: 1:29:27 time: 0.1937 data_time: 0.0128 memory: 7124 grad_norm: 6.9931 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0361 loss: 1.0361 2022/09/07 20:25:07 - mmengine - INFO - Epoch(train) [92][1800/3757] lr: 1.0000e-05 eta: 1:29:10 time: 0.1998 data_time: 0.0154 memory: 7124 grad_norm: 7.1660 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0510 loss: 1.0510 2022/09/07 20:25:28 - mmengine - INFO - Epoch(train) [92][1900/3757] lr: 1.0000e-05 eta: 1:28:54 time: 0.2067 data_time: 0.0155 memory: 7124 grad_norm: 7.4031 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9946 loss: 0.9946 2022/09/07 20:25:49 - mmengine - INFO - Epoch(train) [92][2000/3757] lr: 1.0000e-05 eta: 1:28:38 time: 0.2194 data_time: 0.0155 memory: 7124 grad_norm: 7.0161 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8613 loss: 0.8613 2022/09/07 20:26:10 - mmengine - INFO - Epoch(train) [92][2100/3757] lr: 1.0000e-05 eta: 1:28:21 time: 0.2060 data_time: 0.0151 memory: 7124 grad_norm: 7.2810 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9988 loss: 0.9988 2022/09/07 20:26:12 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 20:26:31 - mmengine - INFO - Epoch(train) [92][2200/3757] lr: 1.0000e-05 eta: 1:28:05 time: 0.2026 data_time: 0.0173 memory: 7124 grad_norm: 7.2318 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0624 loss: 1.0624 2022/09/07 20:26:52 - mmengine - INFO - Epoch(train) [92][2300/3757] lr: 1.0000e-05 eta: 1:27:49 time: 0.1995 data_time: 0.0162 memory: 7124 grad_norm: 7.0922 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9809 loss: 0.9809 2022/09/07 20:27:13 - mmengine - INFO - Epoch(train) [92][2400/3757] lr: 1.0000e-05 eta: 1:27:32 time: 0.2049 data_time: 0.0151 memory: 7124 grad_norm: 7.0974 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1216 loss: 1.1216 2022/09/07 20:27:34 - mmengine - INFO - Epoch(train) [92][2500/3757] lr: 1.0000e-05 eta: 1:27:16 time: 0.2221 data_time: 0.0138 memory: 7124 grad_norm: 7.2903 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1662 loss: 1.1662 2022/09/07 20:27:55 - mmengine - INFO - Epoch(train) [92][2600/3757] lr: 1.0000e-05 eta: 1:27:00 time: 0.2084 data_time: 0.0141 memory: 7124 grad_norm: 7.1597 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1005 loss: 1.1005 2022/09/07 20:28:17 - mmengine - INFO - Epoch(train) [92][2700/3757] lr: 1.0000e-05 eta: 1:26:43 time: 0.2045 data_time: 0.0150 memory: 7124 grad_norm: 7.4057 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3653 loss: 1.3653 2022/09/07 20:28:39 - mmengine - INFO - Epoch(train) [92][2800/3757] lr: 1.0000e-05 eta: 1:26:27 time: 0.2720 data_time: 0.0147 memory: 7124 grad_norm: 7.3339 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0496 loss: 1.0496 2022/09/07 20:29:00 - mmengine - INFO - Epoch(train) [92][2900/3757] lr: 1.0000e-05 eta: 1:26:11 time: 0.1920 data_time: 0.0168 memory: 7124 grad_norm: 7.0751 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 0.9906 loss: 0.9906 2022/09/07 20:29:20 - mmengine - INFO - Epoch(train) [92][3000/3757] lr: 1.0000e-05 eta: 1:25:54 time: 0.2015 data_time: 0.0157 memory: 7124 grad_norm: 7.1915 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2224 loss: 1.2224 2022/09/07 20:29:42 - mmengine - INFO - Epoch(train) [92][3100/3757] lr: 1.0000e-05 eta: 1:25:38 time: 0.2166 data_time: 0.0145 memory: 7124 grad_norm: 7.3894 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0863 loss: 1.0863 2022/09/07 20:29:44 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 20:30:03 - mmengine - INFO - Epoch(train) [92][3200/3757] lr: 1.0000e-05 eta: 1:25:22 time: 0.2104 data_time: 0.0132 memory: 7124 grad_norm: 7.0525 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8581 loss: 0.8581 2022/09/07 20:30:24 - mmengine - INFO - Epoch(train) [92][3300/3757] lr: 1.0000e-05 eta: 1:25:05 time: 0.1964 data_time: 0.0153 memory: 7124 grad_norm: 7.2301 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0384 loss: 1.0384 2022/09/07 20:30:46 - mmengine - INFO - Epoch(train) [92][3400/3757] lr: 1.0000e-05 eta: 1:24:49 time: 0.2733 data_time: 0.0515 memory: 7124 grad_norm: 7.1095 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8860 loss: 0.8860 2022/09/07 20:31:10 - mmengine - INFO - Epoch(train) [92][3500/3757] lr: 1.0000e-05 eta: 1:24:33 time: 0.2617 data_time: 0.0149 memory: 7124 grad_norm: 7.1370 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9148 loss: 0.9148 2022/09/07 20:31:31 - mmengine - INFO - Epoch(train) [92][3600/3757] lr: 1.0000e-05 eta: 1:24:16 time: 0.2070 data_time: 0.0166 memory: 7124 grad_norm: 7.1397 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1020 loss: 1.1020 2022/09/07 20:31:53 - mmengine - INFO - Epoch(train) [92][3700/3757] lr: 1.0000e-05 eta: 1:24:00 time: 0.1982 data_time: 0.0136 memory: 7124 grad_norm: 7.1262 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1388 loss: 1.1388 2022/09/07 20:32:04 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 20:32:04 - mmengine - INFO - Epoch(train) [92][3757/3757] lr: 1.0000e-05 eta: 1:23:54 time: 0.2048 data_time: 0.0100 memory: 7124 grad_norm: 6.9894 top1_acc: 0.7143 top5_acc: 0.7143 loss_cls: 1.0790 loss: 1.0790 2022/09/07 20:32:04 - mmengine - INFO - Saving checkpoint at 92 epochs 2022/09/07 20:35:06 - mmengine - INFO - Epoch(val) [92][100/310] eta: 0:04:27 time: 1.2727 data_time: 0.9609 memory: 7627 2022/09/07 20:38:04 - mmengine - INFO - Epoch(val) [92][200/310] eta: 0:03:13 time: 1.7563 data_time: 1.4473 memory: 7627 2022/09/07 20:40:38 - mmengine - INFO - Epoch(val) [92][300/310] eta: 0:00:13 time: 1.3045 data_time: 0.9917 memory: 7627 2022/09/07 20:40:52 - mmengine - INFO - Epoch(val) [92][310/310] acc/top1: 0.7538 acc/top5: 0.9191 acc/mean1: 0.7538 2022/09/07 20:41:15 - mmengine - INFO - Epoch(train) [93][100/3757] lr: 1.0000e-05 eta: 1:23:34 time: 0.1974 data_time: 0.0134 memory: 7627 grad_norm: 6.9962 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.7663 loss: 0.7663 2022/09/07 20:41:35 - mmengine - INFO - Epoch(train) [93][200/3757] lr: 1.0000e-05 eta: 1:23:18 time: 0.1989 data_time: 0.0130 memory: 7124 grad_norm: 7.1342 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.8949 loss: 0.8949 2022/09/07 20:41:54 - mmengine - INFO - Epoch(train) [93][300/3757] lr: 1.0000e-05 eta: 1:23:01 time: 0.1880 data_time: 0.0143 memory: 7124 grad_norm: 6.8430 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.7701 loss: 0.7701 2022/09/07 20:42:05 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 20:42:14 - mmengine - INFO - Epoch(train) [93][400/3757] lr: 1.0000e-05 eta: 1:22:45 time: 0.1928 data_time: 0.0140 memory: 7124 grad_norm: 6.9185 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9434 loss: 0.9434 2022/09/07 20:42:33 - mmengine - INFO - Epoch(train) [93][500/3757] lr: 1.0000e-05 eta: 1:22:28 time: 0.1870 data_time: 0.0144 memory: 7124 grad_norm: 6.9155 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8425 loss: 0.8425 2022/09/07 20:42:53 - mmengine - INFO - Epoch(train) [93][600/3757] lr: 1.0000e-05 eta: 1:22:12 time: 0.1909 data_time: 0.0133 memory: 7124 grad_norm: 7.2176 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.7734 loss: 0.7734 2022/09/07 20:43:12 - mmengine - INFO - Epoch(train) [93][700/3757] lr: 1.0000e-05 eta: 1:21:55 time: 0.1934 data_time: 0.0132 memory: 7124 grad_norm: 7.2102 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8454 loss: 0.8454 2022/09/07 20:43:31 - mmengine - INFO - Epoch(train) [93][800/3757] lr: 1.0000e-05 eta: 1:21:39 time: 0.1947 data_time: 0.0125 memory: 7124 grad_norm: 7.1458 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0916 loss: 1.0916 2022/09/07 20:43:50 - mmengine - INFO - Epoch(train) [93][900/3757] lr: 1.0000e-05 eta: 1:21:22 time: 0.1892 data_time: 0.0129 memory: 7124 grad_norm: 7.0706 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8167 loss: 0.8167 2022/09/07 20:44:09 - mmengine - INFO - Epoch(train) [93][1000/3757] lr: 1.0000e-05 eta: 1:21:05 time: 0.1800 data_time: 0.0133 memory: 7124 grad_norm: 7.1990 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9654 loss: 0.9654 2022/09/07 20:44:28 - mmengine - INFO - Epoch(train) [93][1100/3757] lr: 1.0000e-05 eta: 1:20:49 time: 0.1915 data_time: 0.0145 memory: 7124 grad_norm: 7.3161 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9696 loss: 0.9696 2022/09/07 20:44:47 - mmengine - INFO - Epoch(train) [93][1200/3757] lr: 1.0000e-05 eta: 1:20:32 time: 0.1851 data_time: 0.0160 memory: 7124 grad_norm: 7.3915 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8676 loss: 0.8676 2022/09/07 20:45:06 - mmengine - INFO - Epoch(train) [93][1300/3757] lr: 1.0000e-05 eta: 1:20:16 time: 0.1872 data_time: 0.0151 memory: 7124 grad_norm: 7.1654 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9853 loss: 0.9853 2022/09/07 20:45:17 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 20:45:25 - mmengine - INFO - Epoch(train) [93][1400/3757] lr: 1.0000e-05 eta: 1:19:59 time: 0.1868 data_time: 0.0136 memory: 7124 grad_norm: 6.9978 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 0.8597 loss: 0.8597 2022/09/07 20:45:46 - mmengine - INFO - Epoch(train) [93][1500/3757] lr: 1.0000e-05 eta: 1:19:43 time: 0.2277 data_time: 0.0327 memory: 7124 grad_norm: 6.8803 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8405 loss: 0.8405 2022/09/07 20:46:08 - mmengine - INFO - Epoch(train) [93][1600/3757] lr: 1.0000e-05 eta: 1:19:26 time: 0.2128 data_time: 0.0165 memory: 7124 grad_norm: 7.0543 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0426 loss: 1.0426 2022/09/07 20:46:30 - mmengine - INFO - Epoch(train) [93][1700/3757] lr: 1.0000e-05 eta: 1:19:10 time: 0.2227 data_time: 0.0331 memory: 7124 grad_norm: 7.3103 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9906 loss: 0.9906 2022/09/07 20:46:51 - mmengine - INFO - Epoch(train) [93][1800/3757] lr: 1.0000e-05 eta: 1:18:54 time: 0.2061 data_time: 0.0149 memory: 7124 grad_norm: 6.9691 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0817 loss: 1.0817 2022/09/07 20:47:12 - mmengine - INFO - Epoch(train) [93][1900/3757] lr: 1.0000e-05 eta: 1:18:37 time: 0.1936 data_time: 0.0149 memory: 7124 grad_norm: 7.1553 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0005 loss: 1.0005 2022/09/07 20:47:33 - mmengine - INFO - Epoch(train) [93][2000/3757] lr: 1.0000e-05 eta: 1:18:21 time: 0.2096 data_time: 0.0153 memory: 7124 grad_norm: 7.0243 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0217 loss: 1.0217 2022/09/07 20:47:54 - mmengine - INFO - Epoch(train) [93][2100/3757] lr: 1.0000e-05 eta: 1:18:04 time: 0.1984 data_time: 0.0142 memory: 7124 grad_norm: 7.1569 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9188 loss: 0.9188 2022/09/07 20:48:15 - mmengine - INFO - Epoch(train) [93][2200/3757] lr: 1.0000e-05 eta: 1:17:48 time: 0.2064 data_time: 0.0157 memory: 7124 grad_norm: 7.0110 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8624 loss: 0.8624 2022/09/07 20:48:37 - mmengine - INFO - Epoch(train) [93][2300/3757] lr: 1.0000e-05 eta: 1:17:31 time: 0.2208 data_time: 0.0139 memory: 7124 grad_norm: 7.1931 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8972 loss: 0.8972 2022/09/07 20:48:48 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 20:48:58 - mmengine - INFO - Epoch(train) [93][2400/3757] lr: 1.0000e-05 eta: 1:17:15 time: 0.2157 data_time: 0.0230 memory: 7124 grad_norm: 6.9972 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2270 loss: 1.2270 2022/09/07 20:49:18 - mmengine - INFO - Epoch(train) [93][2500/3757] lr: 1.0000e-05 eta: 1:16:59 time: 0.2096 data_time: 0.0144 memory: 7124 grad_norm: 7.2117 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0595 loss: 1.0595 2022/09/07 20:49:39 - mmengine - INFO - Epoch(train) [93][2600/3757] lr: 1.0000e-05 eta: 1:16:42 time: 0.1966 data_time: 0.0144 memory: 7124 grad_norm: 7.1608 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0606 loss: 1.0606 2022/09/07 20:50:00 - mmengine - INFO - Epoch(train) [93][2700/3757] lr: 1.0000e-05 eta: 1:16:26 time: 0.2087 data_time: 0.0155 memory: 7124 grad_norm: 7.4049 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9332 loss: 0.9332 2022/09/07 20:50:21 - mmengine - INFO - Epoch(train) [93][2800/3757] lr: 1.0000e-05 eta: 1:16:09 time: 0.1904 data_time: 0.0148 memory: 7124 grad_norm: 7.3426 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2667 loss: 1.2667 2022/09/07 20:50:42 - mmengine - INFO - Epoch(train) [93][2900/3757] lr: 1.0000e-05 eta: 1:15:53 time: 0.2053 data_time: 0.0136 memory: 7124 grad_norm: 7.1152 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8513 loss: 0.8513 2022/09/07 20:51:03 - mmengine - INFO - Epoch(train) [93][3000/3757] lr: 1.0000e-05 eta: 1:15:36 time: 0.2107 data_time: 0.0171 memory: 7124 grad_norm: 7.0697 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1615 loss: 1.1615 2022/09/07 20:51:24 - mmengine - INFO - Epoch(train) [93][3100/3757] lr: 1.0000e-05 eta: 1:15:20 time: 0.2067 data_time: 0.0124 memory: 7124 grad_norm: 7.2316 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2376 loss: 1.2376 2022/09/07 20:51:45 - mmengine - INFO - Epoch(train) [93][3200/3757] lr: 1.0000e-05 eta: 1:15:03 time: 0.2080 data_time: 0.0153 memory: 7124 grad_norm: 7.3237 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8229 loss: 0.8229 2022/09/07 20:52:06 - mmengine - INFO - Epoch(train) [93][3300/3757] lr: 1.0000e-05 eta: 1:14:47 time: 0.2059 data_time: 0.0176 memory: 7124 grad_norm: 7.1987 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0753 loss: 1.0753 2022/09/07 20:52:18 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 20:52:27 - mmengine - INFO - Epoch(train) [93][3400/3757] lr: 1.0000e-05 eta: 1:14:31 time: 0.2312 data_time: 0.0169 memory: 7124 grad_norm: 7.2897 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0181 loss: 1.0181 2022/09/07 20:52:48 - mmengine - INFO - Epoch(train) [93][3500/3757] lr: 1.0000e-05 eta: 1:14:14 time: 0.2174 data_time: 0.0301 memory: 7124 grad_norm: 6.7350 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.9016 loss: 0.9016 2022/09/07 20:53:09 - mmengine - INFO - Epoch(train) [93][3600/3757] lr: 1.0000e-05 eta: 1:13:58 time: 0.2123 data_time: 0.0157 memory: 7124 grad_norm: 7.1368 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0609 loss: 1.0609 2022/09/07 20:53:30 - mmengine - INFO - Epoch(train) [93][3700/3757] lr: 1.0000e-05 eta: 1:13:41 time: 0.2111 data_time: 0.0265 memory: 7124 grad_norm: 7.2854 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1891 loss: 1.1891 2022/09/07 20:53:41 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 20:53:41 - mmengine - INFO - Epoch(train) [93][3757/3757] lr: 1.0000e-05 eta: 1:13:34 time: 0.1707 data_time: 0.0098 memory: 7124 grad_norm: 7.0771 top1_acc: 0.7143 top5_acc: 0.8571 loss_cls: 1.1356 loss: 1.1356 2022/09/07 20:53:41 - mmengine - INFO - Saving checkpoint at 93 epochs 2022/09/07 20:56:43 - mmengine - INFO - Epoch(val) [93][100/310] eta: 0:05:08 time: 1.4694 data_time: 1.1573 memory: 7627 2022/09/07 20:59:36 - mmengine - INFO - Epoch(val) [93][200/310] eta: 0:03:13 time: 1.7568 data_time: 1.4481 memory: 7627 2022/09/07 21:02:12 - mmengine - INFO - Epoch(val) [93][300/310] eta: 0:00:13 time: 1.3449 data_time: 1.0351 memory: 7627 2022/09/07 21:02:32 - mmengine - INFO - Epoch(val) [93][310/310] acc/top1: 0.7529 acc/top5: 0.9187 acc/mean1: 0.7528 2022/09/07 21:02:54 - mmengine - INFO - Epoch(train) [94][100/3757] lr: 1.0000e-05 eta: 1:13:15 time: 0.1947 data_time: 0.0147 memory: 7627 grad_norm: 7.2925 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1036 loss: 1.1036 2022/09/07 21:03:15 - mmengine - INFO - Epoch(train) [94][200/3757] lr: 1.0000e-05 eta: 1:12:58 time: 0.1904 data_time: 0.0146 memory: 7124 grad_norm: 7.3155 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.9184 loss: 0.9184 2022/09/07 21:03:35 - mmengine - INFO - Epoch(train) [94][300/3757] lr: 1.0000e-05 eta: 1:12:42 time: 0.1964 data_time: 0.0155 memory: 7124 grad_norm: 7.0939 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8168 loss: 0.8168 2022/09/07 21:03:54 - mmengine - INFO - Epoch(train) [94][400/3757] lr: 1.0000e-05 eta: 1:12:25 time: 0.1876 data_time: 0.0129 memory: 7124 grad_norm: 7.0064 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8495 loss: 0.8495 2022/09/07 21:04:14 - mmengine - INFO - Epoch(train) [94][500/3757] lr: 1.0000e-05 eta: 1:12:09 time: 0.1934 data_time: 0.0119 memory: 7124 grad_norm: 7.1721 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1949 loss: 1.1949 2022/09/07 21:04:34 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 21:04:34 - mmengine - INFO - Epoch(train) [94][600/3757] lr: 1.0000e-05 eta: 1:11:52 time: 0.1965 data_time: 0.0148 memory: 7124 grad_norm: 7.0278 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8985 loss: 0.8985 2022/09/07 21:04:54 - mmengine - INFO - Epoch(train) [94][700/3757] lr: 1.0000e-05 eta: 1:11:36 time: 0.1898 data_time: 0.0146 memory: 7124 grad_norm: 7.1997 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0134 loss: 1.0134 2022/09/07 21:05:14 - mmengine - INFO - Epoch(train) [94][800/3757] lr: 1.0000e-05 eta: 1:11:19 time: 0.1951 data_time: 0.0155 memory: 7124 grad_norm: 7.2214 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1954 loss: 1.1954 2022/09/07 21:05:34 - mmengine - INFO - Epoch(train) [94][900/3757] lr: 1.0000e-05 eta: 1:11:03 time: 0.1990 data_time: 0.0161 memory: 7124 grad_norm: 7.2224 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9645 loss: 0.9645 2022/09/07 21:05:55 - mmengine - INFO - Epoch(train) [94][1000/3757] lr: 1.0000e-05 eta: 1:10:46 time: 0.2079 data_time: 0.0285 memory: 7124 grad_norm: 7.2254 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0412 loss: 1.0412 2022/09/07 21:06:15 - mmengine - INFO - Epoch(train) [94][1100/3757] lr: 1.0000e-05 eta: 1:10:29 time: 0.1946 data_time: 0.0147 memory: 7124 grad_norm: 7.1950 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0024 loss: 1.0024 2022/09/07 21:06:35 - mmengine - INFO - Epoch(train) [94][1200/3757] lr: 1.0000e-05 eta: 1:10:13 time: 0.1934 data_time: 0.0151 memory: 7124 grad_norm: 7.2628 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1802 loss: 1.1802 2022/09/07 21:06:55 - mmengine - INFO - Epoch(train) [94][1300/3757] lr: 1.0000e-05 eta: 1:09:56 time: 0.1924 data_time: 0.0134 memory: 7124 grad_norm: 7.2875 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0906 loss: 1.0906 2022/09/07 21:07:16 - mmengine - INFO - Epoch(train) [94][1400/3757] lr: 1.0000e-05 eta: 1:09:40 time: 0.2148 data_time: 0.0128 memory: 7124 grad_norm: 7.3693 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8210 loss: 0.8210 2022/09/07 21:07:35 - mmengine - INFO - Epoch(train) [94][1500/3757] lr: 1.0000e-05 eta: 1:09:23 time: 0.1938 data_time: 0.0136 memory: 7124 grad_norm: 7.0036 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9303 loss: 0.9303 2022/09/07 21:07:55 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 21:07:55 - mmengine - INFO - Epoch(train) [94][1600/3757] lr: 1.0000e-05 eta: 1:09:07 time: 0.1980 data_time: 0.0153 memory: 7124 grad_norm: 7.3903 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2878 loss: 1.2878 2022/09/07 21:08:14 - mmengine - INFO - Epoch(train) [94][1700/3757] lr: 1.0000e-05 eta: 1:08:50 time: 0.1918 data_time: 0.0146 memory: 7124 grad_norm: 7.1577 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.0528 loss: 1.0528 2022/09/07 21:08:34 - mmengine - INFO - Epoch(train) [94][1800/3757] lr: 1.0000e-05 eta: 1:08:33 time: 0.1880 data_time: 0.0144 memory: 7124 grad_norm: 7.2256 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2509 loss: 1.2509 2022/09/07 21:08:54 - mmengine - INFO - Epoch(train) [94][1900/3757] lr: 1.0000e-05 eta: 1:08:17 time: 0.1867 data_time: 0.0135 memory: 7124 grad_norm: 6.9554 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.0918 loss: 1.0918 2022/09/07 21:09:12 - mmengine - INFO - Epoch(train) [94][2000/3757] lr: 1.0000e-05 eta: 1:08:00 time: 0.1737 data_time: 0.0116 memory: 7124 grad_norm: 7.5305 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8912 loss: 0.8912 2022/09/07 21:09:32 - mmengine - INFO - Epoch(train) [94][2100/3757] lr: 1.0000e-05 eta: 1:07:43 time: 0.2177 data_time: 0.0418 memory: 7124 grad_norm: 7.1464 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.0119 loss: 1.0119 2022/09/07 21:09:52 - mmengine - INFO - Epoch(train) [94][2200/3757] lr: 1.0000e-05 eta: 1:07:27 time: 0.2084 data_time: 0.0363 memory: 7124 grad_norm: 7.4911 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1589 loss: 1.1589 2022/09/07 21:10:11 - mmengine - INFO - Epoch(train) [94][2300/3757] lr: 1.0000e-05 eta: 1:07:10 time: 0.1860 data_time: 0.0145 memory: 7124 grad_norm: 7.5751 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1504 loss: 1.1504 2022/09/07 21:10:31 - mmengine - INFO - Epoch(train) [94][2400/3757] lr: 1.0000e-05 eta: 1:06:54 time: 0.2283 data_time: 0.0130 memory: 7124 grad_norm: 7.1048 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8977 loss: 0.8977 2022/09/07 21:10:50 - mmengine - INFO - Epoch(train) [94][2500/3757] lr: 1.0000e-05 eta: 1:06:37 time: 0.2143 data_time: 0.0123 memory: 7124 grad_norm: 7.3162 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0368 loss: 1.0368 2022/09/07 21:11:10 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 21:11:10 - mmengine - INFO - Epoch(train) [94][2600/3757] lr: 1.0000e-05 eta: 1:06:20 time: 0.1935 data_time: 0.0207 memory: 7124 grad_norm: 7.1906 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1750 loss: 1.1750 2022/09/07 21:11:29 - mmengine - INFO - Epoch(train) [94][2700/3757] lr: 1.0000e-05 eta: 1:06:04 time: 0.1931 data_time: 0.0162 memory: 7124 grad_norm: 7.2690 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0238 loss: 1.0238 2022/09/07 21:11:48 - mmengine - INFO - Epoch(train) [94][2800/3757] lr: 1.0000e-05 eta: 1:05:47 time: 0.1823 data_time: 0.0133 memory: 7124 grad_norm: 7.5240 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9541 loss: 0.9541 2022/09/07 21:12:07 - mmengine - INFO - Epoch(train) [94][2900/3757] lr: 1.0000e-05 eta: 1:05:30 time: 0.1886 data_time: 0.0133 memory: 7124 grad_norm: 7.4208 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0483 loss: 1.0483 2022/09/07 21:12:26 - mmengine - INFO - Epoch(train) [94][3000/3757] lr: 1.0000e-05 eta: 1:05:14 time: 0.1890 data_time: 0.0154 memory: 7124 grad_norm: 7.3393 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.6469 loss: 0.6469 2022/09/07 21:12:45 - mmengine - INFO - Epoch(train) [94][3100/3757] lr: 1.0000e-05 eta: 1:04:57 time: 0.1867 data_time: 0.0137 memory: 7124 grad_norm: 7.2135 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1678 loss: 1.1678 2022/09/07 21:13:05 - mmengine - INFO - Epoch(train) [94][3200/3757] lr: 1.0000e-05 eta: 1:04:40 time: 0.1861 data_time: 0.0132 memory: 7124 grad_norm: 7.0362 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9600 loss: 0.9600 2022/09/07 21:13:26 - mmengine - INFO - Epoch(train) [94][3300/3757] lr: 1.0000e-05 eta: 1:04:24 time: 0.2049 data_time: 0.0148 memory: 7124 grad_norm: 7.0421 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8038 loss: 0.8038 2022/09/07 21:13:48 - mmengine - INFO - Epoch(train) [94][3400/3757] lr: 1.0000e-05 eta: 1:04:07 time: 0.2143 data_time: 0.0159 memory: 7124 grad_norm: 6.6871 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8357 loss: 0.8357 2022/09/07 21:14:10 - mmengine - INFO - Epoch(train) [94][3500/3757] lr: 1.0000e-05 eta: 1:03:51 time: 0.2083 data_time: 0.0150 memory: 7124 grad_norm: 6.9753 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8872 loss: 0.8872 2022/09/07 21:14:31 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 21:14:31 - mmengine - INFO - Epoch(train) [94][3600/3757] lr: 1.0000e-05 eta: 1:03:34 time: 0.2411 data_time: 0.0167 memory: 7124 grad_norm: 7.2747 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1957 loss: 1.1957 2022/09/07 21:14:53 - mmengine - INFO - Epoch(train) [94][3700/3757] lr: 1.0000e-05 eta: 1:03:18 time: 0.2193 data_time: 0.0158 memory: 7124 grad_norm: 7.5139 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.2077 loss: 1.2077 2022/09/07 21:15:05 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 21:15:05 - mmengine - INFO - Epoch(train) [94][3757/3757] lr: 1.0000e-05 eta: 1:03:11 time: 0.1770 data_time: 0.0109 memory: 7124 grad_norm: 7.1556 top1_acc: 0.8571 top5_acc: 1.0000 loss_cls: 1.0852 loss: 1.0852 2022/09/07 21:15:05 - mmengine - INFO - Saving checkpoint at 94 epochs 2022/09/07 21:18:20 - mmengine - INFO - Epoch(val) [94][100/310] eta: 0:06:24 time: 1.8328 data_time: 1.5122 memory: 7627 2022/09/07 21:21:14 - mmengine - INFO - Epoch(val) [94][200/310] eta: 0:02:37 time: 1.4345 data_time: 1.1190 memory: 7627 2022/09/07 21:24:04 - mmengine - INFO - Epoch(val) [94][300/310] eta: 0:00:17 time: 1.7697 data_time: 1.4588 memory: 7627 2022/09/07 21:24:28 - mmengine - INFO - Epoch(val) [94][310/310] acc/top1: 0.7529 acc/top5: 0.9202 acc/mean1: 0.7528 2022/09/07 21:24:51 - mmengine - INFO - Epoch(train) [95][100/3757] lr: 1.0000e-05 eta: 1:02:52 time: 0.2042 data_time: 0.0153 memory: 7627 grad_norm: 7.2403 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9627 loss: 0.9627 2022/09/07 21:25:11 - mmengine - INFO - Epoch(train) [95][200/3757] lr: 1.0000e-05 eta: 1:02:35 time: 0.2029 data_time: 0.0165 memory: 7124 grad_norm: 7.3845 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2142 loss: 1.2142 2022/09/07 21:25:32 - mmengine - INFO - Epoch(train) [95][300/3757] lr: 1.0000e-05 eta: 1:02:19 time: 0.2074 data_time: 0.0177 memory: 7124 grad_norm: 6.8974 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1754 loss: 1.1754 2022/09/07 21:25:52 - mmengine - INFO - Epoch(train) [95][400/3757] lr: 1.0000e-05 eta: 1:02:02 time: 0.2049 data_time: 0.0137 memory: 7124 grad_norm: 7.2264 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9514 loss: 0.9514 2022/09/07 21:26:13 - mmengine - INFO - Epoch(train) [95][500/3757] lr: 1.0000e-05 eta: 1:01:46 time: 0.2000 data_time: 0.0151 memory: 7124 grad_norm: 6.9646 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.9003 loss: 0.9003 2022/09/07 21:26:33 - mmengine - INFO - Epoch(train) [95][600/3757] lr: 1.0000e-05 eta: 1:01:29 time: 0.2116 data_time: 0.0456 memory: 7124 grad_norm: 7.1544 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1472 loss: 1.1472 2022/09/07 21:26:54 - mmengine - INFO - Epoch(train) [95][700/3757] lr: 1.0000e-05 eta: 1:01:12 time: 0.2155 data_time: 0.0141 memory: 7124 grad_norm: 7.0360 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8855 loss: 0.8855 2022/09/07 21:27:15 - mmengine - INFO - Epoch(train) [95][800/3757] lr: 1.0000e-05 eta: 1:00:56 time: 0.2008 data_time: 0.0129 memory: 7124 grad_norm: 7.1486 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0645 loss: 1.0645 2022/09/07 21:27:23 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 21:27:35 - mmengine - INFO - Epoch(train) [95][900/3757] lr: 1.0000e-05 eta: 1:00:39 time: 0.2009 data_time: 0.0168 memory: 7124 grad_norm: 6.9257 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0672 loss: 1.0672 2022/09/07 21:27:55 - mmengine - INFO - Epoch(train) [95][1000/3757] lr: 1.0000e-05 eta: 1:00:23 time: 0.2074 data_time: 0.0206 memory: 7124 grad_norm: 7.4123 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1686 loss: 1.1686 2022/09/07 21:28:16 - mmengine - INFO - Epoch(train) [95][1100/3757] lr: 1.0000e-05 eta: 1:00:06 time: 0.1984 data_time: 0.0142 memory: 7124 grad_norm: 7.1412 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0953 loss: 1.0953 2022/09/07 21:28:37 - mmengine - INFO - Epoch(train) [95][1200/3757] lr: 1.0000e-05 eta: 0:59:49 time: 0.2025 data_time: 0.0146 memory: 7124 grad_norm: 7.2787 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8998 loss: 0.8998 2022/09/07 21:28:57 - mmengine - INFO - Epoch(train) [95][1300/3757] lr: 1.0000e-05 eta: 0:59:33 time: 0.1997 data_time: 0.0146 memory: 7124 grad_norm: 7.2168 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8554 loss: 0.8554 2022/09/07 21:29:18 - mmengine - INFO - Epoch(train) [95][1400/3757] lr: 1.0000e-05 eta: 0:59:16 time: 0.2147 data_time: 0.0144 memory: 7124 grad_norm: 7.2391 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1365 loss: 1.1365 2022/09/07 21:29:38 - mmengine - INFO - Epoch(train) [95][1500/3757] lr: 1.0000e-05 eta: 0:59:00 time: 0.1922 data_time: 0.0139 memory: 7124 grad_norm: 7.2472 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1738 loss: 1.1738 2022/09/07 21:29:58 - mmengine - INFO - Epoch(train) [95][1600/3757] lr: 1.0000e-05 eta: 0:58:43 time: 0.2012 data_time: 0.0167 memory: 7124 grad_norm: 7.3563 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.7584 loss: 0.7584 2022/09/07 21:30:18 - mmengine - INFO - Epoch(train) [95][1700/3757] lr: 1.0000e-05 eta: 0:58:26 time: 0.2126 data_time: 0.0306 memory: 7124 grad_norm: 7.1567 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8937 loss: 0.8937 2022/09/07 21:30:38 - mmengine - INFO - Epoch(train) [95][1800/3757] lr: 1.0000e-05 eta: 0:58:10 time: 0.1906 data_time: 0.0143 memory: 7124 grad_norm: 7.1744 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1048 loss: 1.1048 2022/09/07 21:30:46 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 21:30:57 - mmengine - INFO - Epoch(train) [95][1900/3757] lr: 1.0000e-05 eta: 0:57:53 time: 0.1904 data_time: 0.0146 memory: 7124 grad_norm: 7.2079 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9121 loss: 0.9121 2022/09/07 21:31:18 - mmengine - INFO - Epoch(train) [95][2000/3757] lr: 1.0000e-05 eta: 0:57:36 time: 0.1889 data_time: 0.0137 memory: 7124 grad_norm: 7.3299 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.1650 loss: 1.1650 2022/09/07 21:31:37 - mmengine - INFO - Epoch(train) [95][2100/3757] lr: 1.0000e-05 eta: 0:57:20 time: 0.1937 data_time: 0.0142 memory: 7124 grad_norm: 7.5053 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2314 loss: 1.2314 2022/09/07 21:31:56 - mmengine - INFO - Epoch(train) [95][2200/3757] lr: 1.0000e-05 eta: 0:57:03 time: 0.1825 data_time: 0.0143 memory: 7124 grad_norm: 7.0554 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.1385 loss: 1.1385 2022/09/07 21:32:16 - mmengine - INFO - Epoch(train) [95][2300/3757] lr: 1.0000e-05 eta: 0:56:46 time: 0.1799 data_time: 0.0133 memory: 7124 grad_norm: 7.2332 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0411 loss: 1.0411 2022/09/07 21:32:34 - mmengine - INFO - Epoch(train) [95][2400/3757] lr: 1.0000e-05 eta: 0:56:29 time: 0.1772 data_time: 0.0115 memory: 7124 grad_norm: 7.6280 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1996 loss: 1.1996 2022/09/07 21:32:53 - mmengine - INFO - Epoch(train) [95][2500/3757] lr: 1.0000e-05 eta: 0:56:13 time: 0.1877 data_time: 0.0146 memory: 7124 grad_norm: 6.9986 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9251 loss: 0.9251 2022/09/07 21:33:11 - mmengine - INFO - Epoch(train) [95][2600/3757] lr: 1.0000e-05 eta: 0:55:56 time: 0.1901 data_time: 0.0154 memory: 7124 grad_norm: 7.1154 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9578 loss: 0.9578 2022/09/07 21:33:32 - mmengine - INFO - Epoch(train) [95][2700/3757] lr: 1.0000e-05 eta: 0:55:39 time: 0.2034 data_time: 0.0146 memory: 7124 grad_norm: 7.1695 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9762 loss: 0.9762 2022/09/07 21:33:54 - mmengine - INFO - Epoch(train) [95][2800/3757] lr: 1.0000e-05 eta: 0:55:23 time: 0.2130 data_time: 0.0147 memory: 7124 grad_norm: 7.3770 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9286 loss: 0.9286 2022/09/07 21:34:04 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 21:34:17 - mmengine - INFO - Epoch(train) [95][2900/3757] lr: 1.0000e-05 eta: 0:55:06 time: 0.2251 data_time: 0.0146 memory: 7124 grad_norm: 7.4889 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.1224 loss: 1.1224 2022/09/07 21:34:38 - mmengine - INFO - Epoch(train) [95][3000/3757] lr: 1.0000e-05 eta: 0:54:50 time: 0.2104 data_time: 0.0170 memory: 7124 grad_norm: 7.2032 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.9232 loss: 0.9232 2022/09/07 21:35:00 - mmengine - INFO - Epoch(train) [95][3100/3757] lr: 1.0000e-05 eta: 0:54:33 time: 0.2103 data_time: 0.0154 memory: 7124 grad_norm: 7.2410 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1402 loss: 1.1402 2022/09/07 21:35:22 - mmengine - INFO - Epoch(train) [95][3200/3757] lr: 1.0000e-05 eta: 0:54:17 time: 0.2472 data_time: 0.0164 memory: 7124 grad_norm: 6.9973 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9498 loss: 0.9498 2022/09/07 21:35:44 - mmengine - INFO - Epoch(train) [95][3300/3757] lr: 1.0000e-05 eta: 0:54:00 time: 0.2062 data_time: 0.0172 memory: 7124 grad_norm: 6.9508 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0218 loss: 1.0218 2022/09/07 21:36:05 - mmengine - INFO - Epoch(train) [95][3400/3757] lr: 1.0000e-05 eta: 0:53:43 time: 0.2095 data_time: 0.0136 memory: 7124 grad_norm: 6.9636 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9183 loss: 0.9183 2022/09/07 21:36:27 - mmengine - INFO - Epoch(train) [95][3500/3757] lr: 1.0000e-05 eta: 0:53:27 time: 0.2084 data_time: 0.0160 memory: 7124 grad_norm: 7.0057 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1116 loss: 1.1116 2022/09/07 21:36:48 - mmengine - INFO - Epoch(train) [95][3600/3757] lr: 1.0000e-05 eta: 0:53:10 time: 0.2142 data_time: 0.0178 memory: 7124 grad_norm: 7.3447 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3520 loss: 1.3520 2022/09/07 21:37:10 - mmengine - INFO - Epoch(train) [95][3700/3757] lr: 1.0000e-05 eta: 0:52:54 time: 0.2131 data_time: 0.0174 memory: 7124 grad_norm: 7.4669 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.0396 loss: 1.0396 2022/09/07 21:37:21 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 21:37:21 - mmengine - INFO - Epoch(train) [95][3757/3757] lr: 1.0000e-05 eta: 0:52:47 time: 0.1792 data_time: 0.0111 memory: 7124 grad_norm: 7.3165 top1_acc: 0.5714 top5_acc: 1.0000 loss_cls: 1.1628 loss: 1.1628 2022/09/07 21:37:22 - mmengine - INFO - Saving checkpoint at 95 epochs 2022/09/07 21:40:34 - mmengine - INFO - Epoch(val) [95][100/310] eta: 0:05:41 time: 1.6245 data_time: 1.3086 memory: 7627 2022/09/07 21:43:41 - mmengine - INFO - Epoch(val) [95][200/310] eta: 0:03:04 time: 1.6814 data_time: 1.3649 memory: 7627 2022/09/07 21:46:26 - mmengine - INFO - Epoch(val) [95][300/310] eta: 0:00:15 time: 1.5073 data_time: 1.1962 memory: 7627 2022/09/07 21:46:50 - mmengine - INFO - Epoch(val) [95][310/310] acc/top1: 0.7525 acc/top5: 0.9183 acc/mean1: 0.7524 2022/09/07 21:47:10 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 21:47:13 - mmengine - INFO - Epoch(train) [96][100/3757] lr: 1.0000e-05 eta: 0:52:27 time: 0.2024 data_time: 0.0162 memory: 7627 grad_norm: 7.2352 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1012 loss: 1.1012 2022/09/07 21:47:34 - mmengine - INFO - Epoch(train) [96][200/3757] lr: 1.0000e-05 eta: 0:52:11 time: 0.2024 data_time: 0.0167 memory: 7124 grad_norm: 7.1358 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0457 loss: 1.0457 2022/09/07 21:47:54 - mmengine - INFO - Epoch(train) [96][300/3757] lr: 1.0000e-05 eta: 0:51:54 time: 0.2024 data_time: 0.0205 memory: 7124 grad_norm: 7.4855 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0261 loss: 1.0261 2022/09/07 21:48:14 - mmengine - INFO - Epoch(train) [96][400/3757] lr: 1.0000e-05 eta: 0:51:37 time: 0.2046 data_time: 0.0186 memory: 7124 grad_norm: 7.2197 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0725 loss: 1.0725 2022/09/07 21:48:34 - mmengine - INFO - Epoch(train) [96][500/3757] lr: 1.0000e-05 eta: 0:51:21 time: 0.2095 data_time: 0.0134 memory: 7124 grad_norm: 7.1765 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9271 loss: 0.9271 2022/09/07 21:48:54 - mmengine - INFO - Epoch(train) [96][600/3757] lr: 1.0000e-05 eta: 0:51:04 time: 0.1977 data_time: 0.0161 memory: 7124 grad_norm: 7.5324 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.2405 loss: 1.2405 2022/09/07 21:49:14 - mmengine - INFO - Epoch(train) [96][700/3757] lr: 1.0000e-05 eta: 0:50:47 time: 0.1921 data_time: 0.0120 memory: 7124 grad_norm: 7.1461 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0908 loss: 1.0908 2022/09/07 21:49:34 - mmengine - INFO - Epoch(train) [96][800/3757] lr: 1.0000e-05 eta: 0:50:31 time: 0.2037 data_time: 0.0151 memory: 7124 grad_norm: 7.1166 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1324 loss: 1.1324 2022/09/07 21:49:53 - mmengine - INFO - Epoch(train) [96][900/3757] lr: 1.0000e-05 eta: 0:50:14 time: 0.1905 data_time: 0.0133 memory: 7124 grad_norm: 6.9627 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9099 loss: 0.9099 2022/09/07 21:50:13 - mmengine - INFO - Epoch(train) [96][1000/3757] lr: 1.0000e-05 eta: 0:49:57 time: 0.1926 data_time: 0.0151 memory: 7124 grad_norm: 7.5380 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0611 loss: 1.0611 2022/09/07 21:50:29 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 21:50:31 - mmengine - INFO - Epoch(train) [96][1100/3757] lr: 1.0000e-05 eta: 0:49:40 time: 0.1887 data_time: 0.0144 memory: 7124 grad_norm: 7.1322 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.7516 loss: 0.7516 2022/09/07 21:50:51 - mmengine - INFO - Epoch(train) [96][1200/3757] lr: 1.0000e-05 eta: 0:49:24 time: 0.1874 data_time: 0.0135 memory: 7124 grad_norm: 7.1980 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 0.7988 loss: 0.7988 2022/09/07 21:51:10 - mmengine - INFO - Epoch(train) [96][1300/3757] lr: 1.0000e-05 eta: 0:49:07 time: 0.1862 data_time: 0.0158 memory: 7124 grad_norm: 6.7306 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9084 loss: 0.9084 2022/09/07 21:51:29 - mmengine - INFO - Epoch(train) [96][1400/3757] lr: 1.0000e-05 eta: 0:48:50 time: 0.1869 data_time: 0.0140 memory: 7124 grad_norm: 7.1924 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1465 loss: 1.1465 2022/09/07 21:51:47 - mmengine - INFO - Epoch(train) [96][1500/3757] lr: 1.0000e-05 eta: 0:48:33 time: 0.1771 data_time: 0.0119 memory: 7124 grad_norm: 7.3819 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0274 loss: 1.0274 2022/09/07 21:52:05 - mmengine - INFO - Epoch(train) [96][1600/3757] lr: 1.0000e-05 eta: 0:48:16 time: 0.1836 data_time: 0.0117 memory: 7124 grad_norm: 7.0817 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0480 loss: 1.0480 2022/09/07 21:52:24 - mmengine - INFO - Epoch(train) [96][1700/3757] lr: 1.0000e-05 eta: 0:48:00 time: 0.1844 data_time: 0.0136 memory: 7124 grad_norm: 7.2362 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8696 loss: 0.8696 2022/09/07 21:52:43 - mmengine - INFO - Epoch(train) [96][1800/3757] lr: 1.0000e-05 eta: 0:47:43 time: 0.1813 data_time: 0.0125 memory: 7124 grad_norm: 7.1447 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.8099 loss: 0.8099 2022/09/07 21:53:01 - mmengine - INFO - Epoch(train) [96][1900/3757] lr: 1.0000e-05 eta: 0:47:26 time: 0.1817 data_time: 0.0151 memory: 7124 grad_norm: 7.5812 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.0668 loss: 1.0668 2022/09/07 21:53:19 - mmengine - INFO - Epoch(train) [96][2000/3757] lr: 1.0000e-05 eta: 0:47:09 time: 0.1785 data_time: 0.0120 memory: 7124 grad_norm: 7.2144 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0503 loss: 1.0503 2022/09/07 21:53:35 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 21:53:38 - mmengine - INFO - Epoch(train) [96][2100/3757] lr: 1.0000e-05 eta: 0:46:52 time: 0.1802 data_time: 0.0132 memory: 7124 grad_norm: 7.2028 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9829 loss: 0.9829 2022/09/07 21:53:56 - mmengine - INFO - Epoch(train) [96][2200/3757] lr: 1.0000e-05 eta: 0:46:36 time: 0.1857 data_time: 0.0136 memory: 7124 grad_norm: 7.2789 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9829 loss: 0.9829 2022/09/07 21:54:15 - mmengine - INFO - Epoch(train) [96][2300/3757] lr: 1.0000e-05 eta: 0:46:19 time: 0.1805 data_time: 0.0130 memory: 7124 grad_norm: 7.3497 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1878 loss: 1.1878 2022/09/07 21:54:34 - mmengine - INFO - Epoch(train) [96][2400/3757] lr: 1.0000e-05 eta: 0:46:02 time: 0.2124 data_time: 0.0128 memory: 7124 grad_norm: 7.3950 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1526 loss: 1.1526 2022/09/07 21:54:54 - mmengine - INFO - Epoch(train) [96][2500/3757] lr: 1.0000e-05 eta: 0:45:45 time: 0.2147 data_time: 0.0155 memory: 7124 grad_norm: 7.2867 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9416 loss: 0.9416 2022/09/07 21:55:15 - mmengine - INFO - Epoch(train) [96][2600/3757] lr: 1.0000e-05 eta: 0:45:29 time: 0.2150 data_time: 0.0135 memory: 7124 grad_norm: 7.2318 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8976 loss: 0.8976 2022/09/07 21:55:36 - mmengine - INFO - Epoch(train) [96][2700/3757] lr: 1.0000e-05 eta: 0:45:12 time: 0.2084 data_time: 0.0166 memory: 7124 grad_norm: 7.3658 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.0961 loss: 1.0961 2022/09/07 21:55:57 - mmengine - INFO - Epoch(train) [96][2800/3757] lr: 1.0000e-05 eta: 0:44:55 time: 0.2056 data_time: 0.0132 memory: 7124 grad_norm: 6.8774 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0600 loss: 1.0600 2022/09/07 21:56:18 - mmengine - INFO - Epoch(train) [96][2900/3757] lr: 1.0000e-05 eta: 0:44:39 time: 0.2066 data_time: 0.0132 memory: 7124 grad_norm: 7.3438 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9244 loss: 0.9244 2022/09/07 21:56:39 - mmengine - INFO - Epoch(train) [96][3000/3757] lr: 1.0000e-05 eta: 0:44:22 time: 0.2097 data_time: 0.0145 memory: 7124 grad_norm: 6.9820 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0009 loss: 1.0009 2022/09/07 21:56:57 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 21:57:00 - mmengine - INFO - Epoch(train) [96][3100/3757] lr: 1.0000e-05 eta: 0:44:05 time: 0.2117 data_time: 0.0173 memory: 7124 grad_norm: 7.2811 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9911 loss: 0.9911 2022/09/07 21:57:22 - mmengine - INFO - Epoch(train) [96][3200/3757] lr: 1.0000e-05 eta: 0:43:49 time: 0.2146 data_time: 0.0147 memory: 7124 grad_norm: 7.2540 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1257 loss: 1.1257 2022/09/07 21:57:43 - mmengine - INFO - Epoch(train) [96][3300/3757] lr: 1.0000e-05 eta: 0:43:32 time: 0.2090 data_time: 0.0166 memory: 7124 grad_norm: 7.1221 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9938 loss: 0.9938 2022/09/07 21:58:04 - mmengine - INFO - Epoch(train) [96][3400/3757] lr: 1.0000e-05 eta: 0:43:15 time: 0.2338 data_time: 0.0155 memory: 7124 grad_norm: 7.1383 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9572 loss: 0.9572 2022/09/07 21:58:25 - mmengine - INFO - Epoch(train) [96][3500/3757] lr: 1.0000e-05 eta: 0:42:59 time: 0.2344 data_time: 0.0147 memory: 7124 grad_norm: 7.3860 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1077 loss: 1.1077 2022/09/07 21:58:46 - mmengine - INFO - Epoch(train) [96][3600/3757] lr: 1.0000e-05 eta: 0:42:42 time: 0.1922 data_time: 0.0149 memory: 7124 grad_norm: 7.1259 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.2823 loss: 1.2823 2022/09/07 21:59:07 - mmengine - INFO - Epoch(train) [96][3700/3757] lr: 1.0000e-05 eta: 0:42:25 time: 0.2076 data_time: 0.0138 memory: 7124 grad_norm: 7.2536 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 0.9958 loss: 0.9958 2022/09/07 21:59:19 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 21:59:19 - mmengine - INFO - Epoch(train) [96][3757/3757] lr: 1.0000e-05 eta: 0:42:18 time: 0.1751 data_time: 0.0101 memory: 7124 grad_norm: 6.9798 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8334 loss: 0.8334 2022/09/07 21:59:19 - mmengine - INFO - Saving checkpoint at 96 epochs 2022/09/07 22:02:21 - mmengine - INFO - Epoch(val) [96][100/310] eta: 0:05:13 time: 1.4913 data_time: 1.1791 memory: 7627 2022/09/07 22:05:28 - mmengine - INFO - Epoch(val) [96][200/310] eta: 0:03:22 time: 1.8388 data_time: 1.5269 memory: 7627 2022/09/07 22:08:03 - mmengine - INFO - Epoch(val) [96][300/310] eta: 0:00:13 time: 1.3682 data_time: 1.0617 memory: 7627 2022/09/07 22:08:19 - mmengine - INFO - Epoch(val) [96][310/310] acc/top1: 0.7516 acc/top5: 0.9188 acc/mean1: 0.7515 2022/09/07 22:08:42 - mmengine - INFO - Epoch(train) [97][100/3757] lr: 1.0000e-05 eta: 0:41:59 time: 0.1947 data_time: 0.0149 memory: 7627 grad_norm: 6.8800 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0214 loss: 1.0214 2022/09/07 22:09:02 - mmengine - INFO - Epoch(train) [97][200/3757] lr: 1.0000e-05 eta: 0:41:42 time: 0.1873 data_time: 0.0126 memory: 7124 grad_norm: 7.1851 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1023 loss: 1.1023 2022/09/07 22:09:22 - mmengine - INFO - Epoch(train) [97][300/3757] lr: 1.0000e-05 eta: 0:41:25 time: 0.2009 data_time: 0.0159 memory: 7124 grad_norm: 7.2006 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.0339 loss: 1.0339 2022/09/07 22:09:28 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 22:09:42 - mmengine - INFO - Epoch(train) [97][400/3757] lr: 1.0000e-05 eta: 0:41:09 time: 0.1933 data_time: 0.0160 memory: 7124 grad_norm: 7.0505 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9016 loss: 0.9016 2022/09/07 22:10:03 - mmengine - INFO - Epoch(train) [97][500/3757] lr: 1.0000e-05 eta: 0:40:52 time: 0.1965 data_time: 0.0137 memory: 7124 grad_norm: 6.9960 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9633 loss: 0.9633 2022/09/07 22:10:24 - mmengine - INFO - Epoch(train) [97][600/3757] lr: 1.0000e-05 eta: 0:40:35 time: 0.1909 data_time: 0.0146 memory: 7124 grad_norm: 7.4626 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1481 loss: 1.1481 2022/09/07 22:10:45 - mmengine - INFO - Epoch(train) [97][700/3757] lr: 1.0000e-05 eta: 0:40:18 time: 0.1963 data_time: 0.0157 memory: 7124 grad_norm: 7.3072 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8351 loss: 0.8351 2022/09/07 22:11:04 - mmengine - INFO - Epoch(train) [97][800/3757] lr: 1.0000e-05 eta: 0:40:02 time: 0.1938 data_time: 0.0151 memory: 7124 grad_norm: 7.2760 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.0999 loss: 1.0999 2022/09/07 22:11:24 - mmengine - INFO - Epoch(train) [97][900/3757] lr: 1.0000e-05 eta: 0:39:45 time: 0.1906 data_time: 0.0145 memory: 7124 grad_norm: 7.1138 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1301 loss: 1.1301 2022/09/07 22:11:44 - mmengine - INFO - Epoch(train) [97][1000/3757] lr: 1.0000e-05 eta: 0:39:28 time: 0.1952 data_time: 0.0144 memory: 7124 grad_norm: 7.5938 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0138 loss: 1.0138 2022/09/07 22:12:04 - mmengine - INFO - Epoch(train) [97][1100/3757] lr: 1.0000e-05 eta: 0:39:11 time: 0.1933 data_time: 0.0133 memory: 7124 grad_norm: 7.2935 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.7844 loss: 0.7844 2022/09/07 22:12:24 - mmengine - INFO - Epoch(train) [97][1200/3757] lr: 1.0000e-05 eta: 0:38:54 time: 0.1986 data_time: 0.0145 memory: 7124 grad_norm: 7.2314 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0356 loss: 1.0356 2022/09/07 22:12:44 - mmengine - INFO - Epoch(train) [97][1300/3757] lr: 1.0000e-05 eta: 0:38:38 time: 0.1881 data_time: 0.0152 memory: 7124 grad_norm: 7.3053 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9705 loss: 0.9705 2022/09/07 22:12:49 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 22:13:03 - mmengine - INFO - Epoch(train) [97][1400/3757] lr: 1.0000e-05 eta: 0:38:21 time: 0.1955 data_time: 0.0146 memory: 7124 grad_norm: 7.5071 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.0549 loss: 1.0549 2022/09/07 22:13:23 - mmengine - INFO - Epoch(train) [97][1500/3757] lr: 1.0000e-05 eta: 0:38:04 time: 0.1913 data_time: 0.0141 memory: 7124 grad_norm: 7.1376 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0090 loss: 1.0090 2022/09/07 22:13:42 - mmengine - INFO - Epoch(train) [97][1600/3757] lr: 1.0000e-05 eta: 0:37:47 time: 0.1957 data_time: 0.0144 memory: 7124 grad_norm: 7.2418 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0353 loss: 1.0353 2022/09/07 22:14:02 - mmengine - INFO - Epoch(train) [97][1700/3757] lr: 1.0000e-05 eta: 0:37:31 time: 0.1861 data_time: 0.0131 memory: 7124 grad_norm: 7.4380 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9864 loss: 0.9864 2022/09/07 22:14:21 - mmengine - INFO - Epoch(train) [97][1800/3757] lr: 1.0000e-05 eta: 0:37:14 time: 0.1897 data_time: 0.0146 memory: 7124 grad_norm: 6.9719 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8484 loss: 0.8484 2022/09/07 22:14:40 - mmengine - INFO - Epoch(train) [97][1900/3757] lr: 1.0000e-05 eta: 0:36:57 time: 0.1855 data_time: 0.0142 memory: 7124 grad_norm: 7.4097 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9692 loss: 0.9692 2022/09/07 22:15:00 - mmengine - INFO - Epoch(train) [97][2000/3757] lr: 1.0000e-05 eta: 0:36:40 time: 0.1856 data_time: 0.0150 memory: 7124 grad_norm: 7.1792 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8881 loss: 0.8881 2022/09/07 22:15:20 - mmengine - INFO - Epoch(train) [97][2100/3757] lr: 1.0000e-05 eta: 0:36:23 time: 0.1898 data_time: 0.0144 memory: 7124 grad_norm: 7.6301 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9191 loss: 0.9191 2022/09/07 22:15:39 - mmengine - INFO - Epoch(train) [97][2200/3757] lr: 1.0000e-05 eta: 0:36:06 time: 0.1868 data_time: 0.0147 memory: 7124 grad_norm: 7.2374 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1136 loss: 1.1136 2022/09/07 22:15:58 - mmengine - INFO - Epoch(train) [97][2300/3757] lr: 1.0000e-05 eta: 0:35:50 time: 0.1930 data_time: 0.0130 memory: 7124 grad_norm: 7.1495 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2687 loss: 1.2687 2022/09/07 22:16:04 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 22:16:18 - mmengine - INFO - Epoch(train) [97][2400/3757] lr: 1.0000e-05 eta: 0:35:33 time: 0.2146 data_time: 0.0125 memory: 7124 grad_norm: 7.4243 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9334 loss: 0.9334 2022/09/07 22:16:38 - mmengine - INFO - Epoch(train) [97][2500/3757] lr: 1.0000e-05 eta: 0:35:16 time: 0.2168 data_time: 0.0138 memory: 7124 grad_norm: 7.0697 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9841 loss: 0.9841 2022/09/07 22:16:59 - mmengine - INFO - Epoch(train) [97][2600/3757] lr: 1.0000e-05 eta: 0:34:59 time: 0.2120 data_time: 0.0156 memory: 7124 grad_norm: 6.9591 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.9405 loss: 0.9405 2022/09/07 22:17:21 - mmengine - INFO - Epoch(train) [97][2700/3757] lr: 1.0000e-05 eta: 0:34:43 time: 0.2161 data_time: 0.0183 memory: 7124 grad_norm: 7.4696 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0117 loss: 1.0117 2022/09/07 22:17:42 - mmengine - INFO - Epoch(train) [97][2800/3757] lr: 1.0000e-05 eta: 0:34:26 time: 0.2131 data_time: 0.0156 memory: 7124 grad_norm: 7.1751 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1898 loss: 1.1898 2022/09/07 22:18:04 - mmengine - INFO - Epoch(train) [97][2900/3757] lr: 1.0000e-05 eta: 0:34:09 time: 0.2101 data_time: 0.0176 memory: 7124 grad_norm: 6.9733 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8431 loss: 0.8431 2022/09/07 22:18:26 - mmengine - INFO - Epoch(train) [97][3000/3757] lr: 1.0000e-05 eta: 0:33:52 time: 0.2204 data_time: 0.0156 memory: 7124 grad_norm: 7.3293 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2075 loss: 1.2075 2022/09/07 22:18:48 - mmengine - INFO - Epoch(train) [97][3100/3757] lr: 1.0000e-05 eta: 0:33:36 time: 0.2214 data_time: 0.0229 memory: 7124 grad_norm: 7.0632 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.1122 loss: 1.1122 2022/09/07 22:19:09 - mmengine - INFO - Epoch(train) [97][3200/3757] lr: 1.0000e-05 eta: 0:33:19 time: 0.2040 data_time: 0.0163 memory: 7124 grad_norm: 7.6996 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1129 loss: 1.1129 2022/09/07 22:19:30 - mmengine - INFO - Epoch(train) [97][3300/3757] lr: 1.0000e-05 eta: 0:33:02 time: 0.2040 data_time: 0.0147 memory: 7124 grad_norm: 7.2048 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9089 loss: 0.9089 2022/09/07 22:19:36 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 22:19:52 - mmengine - INFO - Epoch(train) [97][3400/3757] lr: 1.0000e-05 eta: 0:32:45 time: 0.2089 data_time: 0.0158 memory: 7124 grad_norm: 7.4212 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.0823 loss: 1.0823 2022/09/07 22:20:13 - mmengine - INFO - Epoch(train) [97][3500/3757] lr: 1.0000e-05 eta: 0:32:29 time: 0.2120 data_time: 0.0148 memory: 7124 grad_norm: 7.2333 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9331 loss: 0.9331 2022/09/07 22:20:34 - mmengine - INFO - Epoch(train) [97][3600/3757] lr: 1.0000e-05 eta: 0:32:12 time: 0.2061 data_time: 0.0165 memory: 7124 grad_norm: 7.4512 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1614 loss: 1.1614 2022/09/07 22:20:55 - mmengine - INFO - Epoch(train) [97][3700/3757] lr: 1.0000e-05 eta: 0:31:55 time: 0.2078 data_time: 0.0139 memory: 7124 grad_norm: 7.1835 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0093 loss: 1.0093 2022/09/07 22:21:06 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 22:21:06 - mmengine - INFO - Epoch(train) [97][3757/3757] lr: 1.0000e-05 eta: 0:31:48 time: 0.1744 data_time: 0.0123 memory: 7124 grad_norm: 7.3340 top1_acc: 0.8571 top5_acc: 1.0000 loss_cls: 1.1110 loss: 1.1110 2022/09/07 22:21:06 - mmengine - INFO - Saving checkpoint at 97 epochs 2022/09/07 22:24:14 - mmengine - INFO - Epoch(val) [97][100/310] eta: 0:05:10 time: 1.4806 data_time: 1.1699 memory: 7627 2022/09/07 22:27:11 - mmengine - INFO - Epoch(val) [97][200/310] eta: 0:03:11 time: 1.7422 data_time: 1.4322 memory: 7627 2022/09/07 22:29:43 - mmengine - INFO - Epoch(val) [97][300/310] eta: 0:00:12 time: 1.2587 data_time: 0.9550 memory: 7627 2022/09/07 22:30:00 - mmengine - INFO - Epoch(val) [97][310/310] acc/top1: 0.7525 acc/top5: 0.9184 acc/mean1: 0.7524 2022/09/07 22:30:22 - mmengine - INFO - Epoch(train) [98][100/3757] lr: 1.0000e-05 eta: 0:31:29 time: 0.1841 data_time: 0.0137 memory: 7627 grad_norm: 7.2724 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2384 loss: 1.2384 2022/09/07 22:30:41 - mmengine - INFO - Epoch(train) [98][200/3757] lr: 1.0000e-05 eta: 0:31:12 time: 0.1882 data_time: 0.0146 memory: 7124 grad_norm: 7.1213 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9832 loss: 0.9832 2022/09/07 22:30:59 - mmengine - INFO - Epoch(train) [98][300/3757] lr: 1.0000e-05 eta: 0:30:55 time: 0.1841 data_time: 0.0135 memory: 7124 grad_norm: 7.3012 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9111 loss: 0.9111 2022/09/07 22:31:17 - mmengine - INFO - Epoch(train) [98][400/3757] lr: 1.0000e-05 eta: 0:30:38 time: 0.1789 data_time: 0.0135 memory: 7124 grad_norm: 7.1931 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.7470 loss: 0.7470 2022/09/07 22:31:36 - mmengine - INFO - Epoch(train) [98][500/3757] lr: 1.0000e-05 eta: 0:30:21 time: 0.1839 data_time: 0.0122 memory: 7124 grad_norm: 7.2185 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1625 loss: 1.1625 2022/09/07 22:31:50 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 22:31:55 - mmengine - INFO - Epoch(train) [98][600/3757] lr: 1.0000e-05 eta: 0:30:04 time: 0.1802 data_time: 0.0119 memory: 7124 grad_norm: 7.2884 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8854 loss: 0.8854 2022/09/07 22:32:13 - mmengine - INFO - Epoch(train) [98][700/3757] lr: 1.0000e-05 eta: 0:29:47 time: 0.1801 data_time: 0.0128 memory: 7124 grad_norm: 7.0621 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0209 loss: 1.0209 2022/09/07 22:32:31 - mmengine - INFO - Epoch(train) [98][800/3757] lr: 1.0000e-05 eta: 0:29:30 time: 0.1785 data_time: 0.0130 memory: 7124 grad_norm: 7.2665 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9305 loss: 0.9305 2022/09/07 22:32:50 - mmengine - INFO - Epoch(train) [98][900/3757] lr: 1.0000e-05 eta: 0:29:14 time: 0.2083 data_time: 0.0123 memory: 7124 grad_norm: 7.5663 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9353 loss: 0.9353 2022/09/07 22:33:09 - mmengine - INFO - Epoch(train) [98][1000/3757] lr: 1.0000e-05 eta: 0:28:57 time: 0.1971 data_time: 0.0130 memory: 7124 grad_norm: 6.9383 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.0055 loss: 1.0055 2022/09/07 22:33:27 - mmengine - INFO - Epoch(train) [98][1100/3757] lr: 1.0000e-05 eta: 0:28:40 time: 0.2194 data_time: 0.0134 memory: 7124 grad_norm: 7.4214 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1040 loss: 1.1040 2022/09/07 22:33:45 - mmengine - INFO - Epoch(train) [98][1200/3757] lr: 1.0000e-05 eta: 0:28:23 time: 0.1765 data_time: 0.0132 memory: 7124 grad_norm: 7.1770 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9297 loss: 0.9297 2022/09/07 22:34:04 - mmengine - INFO - Epoch(train) [98][1300/3757] lr: 1.0000e-05 eta: 0:28:06 time: 0.1757 data_time: 0.0121 memory: 7124 grad_norm: 7.1793 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0443 loss: 1.0443 2022/09/07 22:34:22 - mmengine - INFO - Epoch(train) [98][1400/3757] lr: 1.0000e-05 eta: 0:27:49 time: 0.1778 data_time: 0.0131 memory: 7124 grad_norm: 7.2502 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0438 loss: 1.0438 2022/09/07 22:34:40 - mmengine - INFO - Epoch(train) [98][1500/3757] lr: 1.0000e-05 eta: 0:27:32 time: 0.1739 data_time: 0.0119 memory: 7124 grad_norm: 7.3325 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0974 loss: 1.0974 2022/09/07 22:34:53 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 22:34:58 - mmengine - INFO - Epoch(train) [98][1600/3757] lr: 1.0000e-05 eta: 0:27:15 time: 0.1757 data_time: 0.0121 memory: 7124 grad_norm: 7.3171 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9821 loss: 0.9821 2022/09/07 22:35:16 - mmengine - INFO - Epoch(train) [98][1700/3757] lr: 1.0000e-05 eta: 0:26:59 time: 0.1778 data_time: 0.0144 memory: 7124 grad_norm: 7.2642 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.2044 loss: 1.2044 2022/09/07 22:35:35 - mmengine - INFO - Epoch(train) [98][1800/3757] lr: 1.0000e-05 eta: 0:26:42 time: 0.1757 data_time: 0.0147 memory: 7124 grad_norm: 7.3502 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0740 loss: 1.0740 2022/09/07 22:35:53 - mmengine - INFO - Epoch(train) [98][1900/3757] lr: 1.0000e-05 eta: 0:26:25 time: 0.1768 data_time: 0.0156 memory: 7124 grad_norm: 7.0361 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0082 loss: 1.0082 2022/09/07 22:36:11 - mmengine - INFO - Epoch(train) [98][2000/3757] lr: 1.0000e-05 eta: 0:26:08 time: 0.1761 data_time: 0.0128 memory: 7124 grad_norm: 7.2849 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0500 loss: 1.0500 2022/09/07 22:36:29 - mmengine - INFO - Epoch(train) [98][2100/3757] lr: 1.0000e-05 eta: 0:25:51 time: 0.1807 data_time: 0.0134 memory: 7124 grad_norm: 7.2784 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1211 loss: 1.1211 2022/09/07 22:36:47 - mmengine - INFO - Epoch(train) [98][2200/3757] lr: 1.0000e-05 eta: 0:25:34 time: 0.1833 data_time: 0.0130 memory: 7124 grad_norm: 7.4512 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0493 loss: 1.0493 2022/09/07 22:37:06 - mmengine - INFO - Epoch(train) [98][2300/3757] lr: 1.0000e-05 eta: 0:25:17 time: 0.1840 data_time: 0.0115 memory: 7124 grad_norm: 7.2722 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0108 loss: 1.0108 2022/09/07 22:37:24 - mmengine - INFO - Epoch(train) [98][2400/3757] lr: 1.0000e-05 eta: 0:25:00 time: 0.1786 data_time: 0.0139 memory: 7124 grad_norm: 7.0867 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0888 loss: 1.0888 2022/09/07 22:37:42 - mmengine - INFO - Epoch(train) [98][2500/3757] lr: 1.0000e-05 eta: 0:24:43 time: 0.1776 data_time: 0.0138 memory: 7124 grad_norm: 6.9764 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9196 loss: 0.9196 2022/09/07 22:37:55 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 22:38:00 - mmengine - INFO - Epoch(train) [98][2600/3757] lr: 1.0000e-05 eta: 0:24:27 time: 0.1896 data_time: 0.0129 memory: 7124 grad_norm: 7.1461 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.9719 loss: 0.9719 2022/09/07 22:38:18 - mmengine - INFO - Epoch(train) [98][2700/3757] lr: 1.0000e-05 eta: 0:24:10 time: 0.1798 data_time: 0.0132 memory: 7124 grad_norm: 7.3215 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.0706 loss: 1.0706 2022/09/07 22:38:36 - mmengine - INFO - Epoch(train) [98][2800/3757] lr: 1.0000e-05 eta: 0:23:53 time: 0.1672 data_time: 0.0122 memory: 7124 grad_norm: 7.4833 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9372 loss: 0.9372 2022/09/07 22:38:53 - mmengine - INFO - Epoch(train) [98][2900/3757] lr: 1.0000e-05 eta: 0:23:36 time: 0.1711 data_time: 0.0126 memory: 7124 grad_norm: 7.0796 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1680 loss: 1.1680 2022/09/07 22:39:11 - mmengine - INFO - Epoch(train) [98][3000/3757] lr: 1.0000e-05 eta: 0:23:19 time: 0.1701 data_time: 0.0132 memory: 7124 grad_norm: 6.7745 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.8958 loss: 0.8958 2022/09/07 22:39:28 - mmengine - INFO - Epoch(train) [98][3100/3757] lr: 1.0000e-05 eta: 0:23:02 time: 0.1670 data_time: 0.0110 memory: 7124 grad_norm: 7.5129 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.0512 loss: 1.0512 2022/09/07 22:39:45 - mmengine - INFO - Epoch(train) [98][3200/3757] lr: 1.0000e-05 eta: 0:22:45 time: 0.1652 data_time: 0.0120 memory: 7124 grad_norm: 6.9728 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9053 loss: 0.9053 2022/09/07 22:40:02 - mmengine - INFO - Epoch(train) [98][3300/3757] lr: 1.0000e-05 eta: 0:22:28 time: 0.1663 data_time: 0.0107 memory: 7124 grad_norm: 7.1439 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.3756 loss: 1.3756 2022/09/07 22:40:20 - mmengine - INFO - Epoch(train) [98][3400/3757] lr: 1.0000e-05 eta: 0:22:11 time: 0.1790 data_time: 0.0115 memory: 7124 grad_norm: 7.5194 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0249 loss: 1.0249 2022/09/07 22:40:37 - mmengine - INFO - Epoch(train) [98][3500/3757] lr: 1.0000e-05 eta: 0:21:54 time: 0.1691 data_time: 0.0124 memory: 7124 grad_norm: 7.3808 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0656 loss: 1.0656 2022/09/07 22:40:49 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 22:40:54 - mmengine - INFO - Epoch(train) [98][3600/3757] lr: 1.0000e-05 eta: 0:21:37 time: 0.1679 data_time: 0.0124 memory: 7124 grad_norm: 7.0205 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8202 loss: 0.8202 2022/09/07 22:41:11 - mmengine - INFO - Epoch(train) [98][3700/3757] lr: 1.0000e-05 eta: 0:21:20 time: 0.1700 data_time: 0.0121 memory: 7124 grad_norm: 7.0396 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0952 loss: 1.0952 2022/09/07 22:41:21 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 22:41:21 - mmengine - INFO - Epoch(train) [98][3757/3757] lr: 1.0000e-05 eta: 0:21:14 time: 0.1465 data_time: 0.0075 memory: 7124 grad_norm: 7.1218 top1_acc: 0.7143 top5_acc: 0.8571 loss_cls: 1.2412 loss: 1.2412 2022/09/07 22:41:21 - mmengine - INFO - Saving checkpoint at 98 epochs 2022/09/07 22:43:54 - mmengine - INFO - Epoch(val) [98][100/310] eta: 0:04:43 time: 1.3493 data_time: 1.0403 memory: 7627 2022/09/07 22:46:18 - mmengine - INFO - Epoch(val) [98][200/310] eta: 0:02:17 time: 1.2493 data_time: 0.9415 memory: 7627 2022/09/07 22:48:41 - mmengine - INFO - Epoch(val) [98][300/310] eta: 0:00:14 time: 1.4218 data_time: 1.1106 memory: 7627 2022/09/07 22:49:03 - mmengine - INFO - Epoch(val) [98][310/310] acc/top1: 0.7526 acc/top5: 0.9188 acc/mean1: 0.7525 2022/09/07 22:49:24 - mmengine - INFO - Epoch(train) [99][100/3757] lr: 1.0000e-05 eta: 0:20:54 time: 0.1725 data_time: 0.0114 memory: 7627 grad_norm: 7.4336 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1102 loss: 1.1102 2022/09/07 22:49:42 - mmengine - INFO - Epoch(train) [99][200/3757] lr: 1.0000e-05 eta: 0:20:37 time: 0.1731 data_time: 0.0125 memory: 7124 grad_norm: 7.2199 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0456 loss: 1.0456 2022/09/07 22:50:00 - mmengine - INFO - Epoch(train) [99][300/3757] lr: 1.0000e-05 eta: 0:20:20 time: 0.1759 data_time: 0.0119 memory: 7124 grad_norm: 7.3319 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.9103 loss: 0.9103 2022/09/07 22:50:18 - mmengine - INFO - Epoch(train) [99][400/3757] lr: 1.0000e-05 eta: 0:20:03 time: 0.1800 data_time: 0.0113 memory: 7124 grad_norm: 7.2385 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8472 loss: 0.8472 2022/09/07 22:50:35 - mmengine - INFO - Epoch(train) [99][500/3757] lr: 1.0000e-05 eta: 0:19:46 time: 0.1758 data_time: 0.0140 memory: 7124 grad_norm: 7.1183 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.8731 loss: 0.8731 2022/09/07 22:50:54 - mmengine - INFO - Epoch(train) [99][600/3757] lr: 1.0000e-05 eta: 0:19:29 time: 0.1755 data_time: 0.0124 memory: 7124 grad_norm: 7.0013 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0410 loss: 1.0410 2022/09/07 22:51:12 - mmengine - INFO - Epoch(train) [99][700/3757] lr: 1.0000e-05 eta: 0:19:13 time: 0.1977 data_time: 0.0316 memory: 7124 grad_norm: 7.3569 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9675 loss: 0.9675 2022/09/07 22:51:29 - mmengine - INFO - Epoch(train) [99][800/3757] lr: 1.0000e-05 eta: 0:18:56 time: 0.1738 data_time: 0.0122 memory: 7124 grad_norm: 6.9654 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9939 loss: 0.9939 2022/09/07 22:51:32 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 22:51:47 - mmengine - INFO - Epoch(train) [99][900/3757] lr: 1.0000e-05 eta: 0:18:39 time: 0.1684 data_time: 0.0122 memory: 7124 grad_norm: 7.0533 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8211 loss: 0.8211 2022/09/07 22:52:04 - mmengine - INFO - Epoch(train) [99][1000/3757] lr: 1.0000e-05 eta: 0:18:22 time: 0.1742 data_time: 0.0129 memory: 7124 grad_norm: 7.1632 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9241 loss: 0.9241 2022/09/07 22:52:22 - mmengine - INFO - Epoch(train) [99][1100/3757] lr: 1.0000e-05 eta: 0:18:05 time: 0.1791 data_time: 0.0141 memory: 7124 grad_norm: 7.2478 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8808 loss: 0.8808 2022/09/07 22:52:41 - mmengine - INFO - Epoch(train) [99][1200/3757] lr: 1.0000e-05 eta: 0:17:48 time: 0.1987 data_time: 0.0144 memory: 7124 grad_norm: 7.2004 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9881 loss: 0.9881 2022/09/07 22:53:01 - mmengine - INFO - Epoch(train) [99][1300/3757] lr: 1.0000e-05 eta: 0:17:31 time: 0.1911 data_time: 0.0146 memory: 7124 grad_norm: 7.1562 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.0043 loss: 1.0043 2022/09/07 22:53:21 - mmengine - INFO - Epoch(train) [99][1400/3757] lr: 1.0000e-05 eta: 0:17:14 time: 0.2011 data_time: 0.0143 memory: 7124 grad_norm: 7.0583 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2304 loss: 1.2304 2022/09/07 22:53:41 - mmengine - INFO - Epoch(train) [99][1500/3757] lr: 1.0000e-05 eta: 0:16:57 time: 0.1980 data_time: 0.0131 memory: 7124 grad_norm: 7.1671 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8916 loss: 0.8916 2022/09/07 22:54:02 - mmengine - INFO - Epoch(train) [99][1600/3757] lr: 1.0000e-05 eta: 0:16:40 time: 0.1928 data_time: 0.0139 memory: 7124 grad_norm: 7.3983 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.0145 loss: 1.0145 2022/09/07 22:54:22 - mmengine - INFO - Epoch(train) [99][1700/3757] lr: 1.0000e-05 eta: 0:16:24 time: 0.2001 data_time: 0.0155 memory: 7124 grad_norm: 7.5175 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8978 loss: 0.8978 2022/09/07 22:54:42 - mmengine - INFO - Epoch(train) [99][1800/3757] lr: 1.0000e-05 eta: 0:16:07 time: 0.2014 data_time: 0.0136 memory: 7124 grad_norm: 7.1224 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 0.8625 loss: 0.8625 2022/09/07 22:54:45 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 22:55:02 - mmengine - INFO - Epoch(train) [99][1900/3757] lr: 1.0000e-05 eta: 0:15:50 time: 0.1991 data_time: 0.0218 memory: 7124 grad_norm: 6.9905 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0168 loss: 1.0168 2022/09/07 22:55:22 - mmengine - INFO - Epoch(train) [99][2000/3757] lr: 1.0000e-05 eta: 0:15:33 time: 0.2204 data_time: 0.0442 memory: 7124 grad_norm: 7.3448 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2208 loss: 1.2208 2022/09/07 22:55:43 - mmengine - INFO - Epoch(train) [99][2100/3757] lr: 1.0000e-05 eta: 0:15:16 time: 0.2216 data_time: 0.0372 memory: 7124 grad_norm: 7.1947 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1318 loss: 1.1318 2022/09/07 22:56:02 - mmengine - INFO - Epoch(train) [99][2200/3757] lr: 1.0000e-05 eta: 0:14:59 time: 0.2077 data_time: 0.0147 memory: 7124 grad_norm: 7.2519 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 0.9129 loss: 0.9129 2022/09/07 22:56:22 - mmengine - INFO - Epoch(train) [99][2300/3757] lr: 1.0000e-05 eta: 0:14:42 time: 0.2059 data_time: 0.0141 memory: 7124 grad_norm: 6.9525 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8104 loss: 0.8104 2022/09/07 22:56:42 - mmengine - INFO - Epoch(train) [99][2400/3757] lr: 1.0000e-05 eta: 0:14:25 time: 0.2051 data_time: 0.0121 memory: 7124 grad_norm: 7.1582 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0082 loss: 1.0082 2022/09/07 22:57:02 - mmengine - INFO - Epoch(train) [99][2500/3757] lr: 1.0000e-05 eta: 0:14:09 time: 0.2040 data_time: 0.0384 memory: 7124 grad_norm: 7.4425 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1498 loss: 1.1498 2022/09/07 22:57:21 - mmengine - INFO - Epoch(train) [99][2600/3757] lr: 1.0000e-05 eta: 0:13:52 time: 0.1913 data_time: 0.0137 memory: 7124 grad_norm: 7.1756 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8720 loss: 0.8720 2022/09/07 22:57:40 - mmengine - INFO - Epoch(train) [99][2700/3757] lr: 1.0000e-05 eta: 0:13:35 time: 0.1907 data_time: 0.0139 memory: 7124 grad_norm: 7.0945 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0488 loss: 1.0488 2022/09/07 22:57:59 - mmengine - INFO - Epoch(train) [99][2800/3757] lr: 1.0000e-05 eta: 0:13:18 time: 0.1826 data_time: 0.0117 memory: 7124 grad_norm: 7.1987 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1587 loss: 1.1587 2022/09/07 22:58:01 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 22:58:17 - mmengine - INFO - Epoch(train) [99][2900/3757] lr: 1.0000e-05 eta: 0:13:01 time: 0.1805 data_time: 0.0134 memory: 7124 grad_norm: 7.2586 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0592 loss: 1.0592 2022/09/07 22:58:36 - mmengine - INFO - Epoch(train) [99][3000/3757] lr: 1.0000e-05 eta: 0:12:44 time: 0.1845 data_time: 0.0133 memory: 7124 grad_norm: 7.5072 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9795 loss: 0.9795 2022/09/07 22:58:55 - mmengine - INFO - Epoch(train) [99][3100/3757] lr: 1.0000e-05 eta: 0:12:27 time: 0.1799 data_time: 0.0121 memory: 7124 grad_norm: 7.0675 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9299 loss: 0.9299 2022/09/07 22:59:14 - mmengine - INFO - Epoch(train) [99][3200/3757] lr: 1.0000e-05 eta: 0:12:10 time: 0.1823 data_time: 0.0142 memory: 7124 grad_norm: 7.6174 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2153 loss: 1.2153 2022/09/07 22:59:33 - mmengine - INFO - Epoch(train) [99][3300/3757] lr: 1.0000e-05 eta: 0:11:53 time: 0.1996 data_time: 0.0141 memory: 7124 grad_norm: 7.4588 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0548 loss: 1.0548 2022/09/07 22:59:52 - mmengine - INFO - Epoch(train) [99][3400/3757] lr: 1.0000e-05 eta: 0:11:36 time: 0.2222 data_time: 0.0120 memory: 7124 grad_norm: 7.4171 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1595 loss: 1.1595 2022/09/07 23:00:10 - mmengine - INFO - Epoch(train) [99][3500/3757] lr: 1.0000e-05 eta: 0:11:19 time: 0.1823 data_time: 0.0126 memory: 7124 grad_norm: 7.1829 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.8805 loss: 0.8805 2022/09/07 23:00:29 - mmengine - INFO - Epoch(train) [99][3600/3757] lr: 1.0000e-05 eta: 0:11:02 time: 0.1885 data_time: 0.0153 memory: 7124 grad_norm: 7.0593 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1033 loss: 1.1033 2022/09/07 23:00:48 - mmengine - INFO - Epoch(train) [99][3700/3757] lr: 1.0000e-05 eta: 0:10:46 time: 0.1872 data_time: 0.0150 memory: 7124 grad_norm: 7.1159 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8814 loss: 0.8814 2022/09/07 23:00:58 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 23:00:58 - mmengine - INFO - Epoch(train) [99][3757/3757] lr: 1.0000e-05 eta: 0:10:39 time: 0.1618 data_time: 0.0096 memory: 7124 grad_norm: 7.3974 top1_acc: 0.8571 top5_acc: 1.0000 loss_cls: 1.0444 loss: 1.0444 2022/09/07 23:00:58 - mmengine - INFO - Saving checkpoint at 99 epochs 2022/09/07 23:03:50 - mmengine - INFO - Epoch(val) [99][100/310] eta: 0:05:35 time: 1.5991 data_time: 1.2844 memory: 7627 2022/09/07 23:06:49 - mmengine - INFO - Epoch(val) [99][200/310] eta: 0:03:17 time: 1.7982 data_time: 1.4830 memory: 7627 2022/09/07 23:09:29 - mmengine - INFO - Epoch(val) [99][300/310] eta: 0:00:14 time: 1.4657 data_time: 1.1533 memory: 7627 2022/09/07 23:09:50 - mmengine - INFO - Epoch(val) [99][310/310] acc/top1: 0.7532 acc/top5: 0.9186 acc/mean1: 0.7532 2022/09/07 23:10:04 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 23:10:12 - mmengine - INFO - Epoch(train) [100][100/3757] lr: 1.0000e-05 eta: 0:10:19 time: 0.1934 data_time: 0.0142 memory: 7627 grad_norm: 7.3644 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.7911 loss: 0.7911 2022/09/07 23:10:32 - mmengine - INFO - Epoch(train) [100][200/3757] lr: 1.0000e-05 eta: 0:10:02 time: 0.1930 data_time: 0.0156 memory: 7124 grad_norm: 7.4262 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9996 loss: 0.9996 2022/09/07 23:10:50 - mmengine - INFO - Epoch(train) [100][300/3757] lr: 1.0000e-05 eta: 0:09:45 time: 0.1860 data_time: 0.0134 memory: 7124 grad_norm: 7.1476 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1807 loss: 1.1807 2022/09/07 23:11:09 - mmengine - INFO - Epoch(train) [100][400/3757] lr: 1.0000e-05 eta: 0:09:28 time: 0.1847 data_time: 0.0141 memory: 7124 grad_norm: 7.0595 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0559 loss: 1.0559 2022/09/07 23:11:28 - mmengine - INFO - Epoch(train) [100][500/3757] lr: 1.0000e-05 eta: 0:09:11 time: 0.1799 data_time: 0.0133 memory: 7124 grad_norm: 7.2324 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1953 loss: 1.1953 2022/09/07 23:11:47 - mmengine - INFO - Epoch(train) [100][600/3757] lr: 1.0000e-05 eta: 0:08:54 time: 0.1851 data_time: 0.0138 memory: 7124 grad_norm: 7.0293 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0933 loss: 1.0933 2022/09/07 23:12:06 - mmengine - INFO - Epoch(train) [100][700/3757] lr: 1.0000e-05 eta: 0:08:37 time: 0.1910 data_time: 0.0139 memory: 7124 grad_norm: 7.2027 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1568 loss: 1.1568 2022/09/07 23:12:25 - mmengine - INFO - Epoch(train) [100][800/3757] lr: 1.0000e-05 eta: 0:08:20 time: 0.1865 data_time: 0.0121 memory: 7124 grad_norm: 6.9384 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9705 loss: 0.9705 2022/09/07 23:12:44 - mmengine - INFO - Epoch(train) [100][900/3757] lr: 1.0000e-05 eta: 0:08:04 time: 0.1712 data_time: 0.0118 memory: 7124 grad_norm: 7.1739 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9796 loss: 0.9796 2022/09/07 23:13:03 - mmengine - INFO - Epoch(train) [100][1000/3757] lr: 1.0000e-05 eta: 0:07:47 time: 0.2099 data_time: 0.0148 memory: 7124 grad_norm: 7.1789 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8477 loss: 0.8477 2022/09/07 23:13:14 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 23:13:22 - mmengine - INFO - Epoch(train) [100][1100/3757] lr: 1.0000e-05 eta: 0:07:30 time: 0.1987 data_time: 0.0115 memory: 7124 grad_norm: 6.9511 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.9324 loss: 0.9324 2022/09/07 23:13:41 - mmengine - INFO - Epoch(train) [100][1200/3757] lr: 1.0000e-05 eta: 0:07:13 time: 0.1893 data_time: 0.0147 memory: 7124 grad_norm: 7.1733 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9563 loss: 0.9563 2022/09/07 23:14:00 - mmengine - INFO - Epoch(train) [100][1300/3757] lr: 1.0000e-05 eta: 0:06:56 time: 0.1852 data_time: 0.0134 memory: 7124 grad_norm: 7.0391 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9322 loss: 0.9322 2022/09/07 23:14:20 - mmengine - INFO - Epoch(train) [100][1400/3757] lr: 1.0000e-05 eta: 0:06:39 time: 0.1924 data_time: 0.0158 memory: 7124 grad_norm: 7.0303 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0679 loss: 1.0679 2022/09/07 23:14:39 - mmengine - INFO - Epoch(train) [100][1500/3757] lr: 1.0000e-05 eta: 0:06:22 time: 0.1961 data_time: 0.0136 memory: 7124 grad_norm: 7.2714 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8778 loss: 0.8778 2022/09/07 23:14:59 - mmengine - INFO - Epoch(train) [100][1600/3757] lr: 1.0000e-05 eta: 0:06:05 time: 0.1898 data_time: 0.0146 memory: 7124 grad_norm: 6.9733 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0266 loss: 1.0266 2022/09/07 23:15:19 - mmengine - INFO - Epoch(train) [100][1700/3757] lr: 1.0000e-05 eta: 0:05:48 time: 0.1911 data_time: 0.0158 memory: 7124 grad_norm: 7.0600 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0956 loss: 1.0956 2022/09/07 23:15:38 - mmengine - INFO - Epoch(train) [100][1800/3757] lr: 1.0000e-05 eta: 0:05:31 time: 0.1921 data_time: 0.0154 memory: 7124 grad_norm: 7.3232 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0650 loss: 1.0650 2022/09/07 23:15:59 - mmengine - INFO - Epoch(train) [100][1900/3757] lr: 1.0000e-05 eta: 0:05:14 time: 0.1913 data_time: 0.0146 memory: 7124 grad_norm: 6.8537 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.8513 loss: 0.8513 2022/09/07 23:16:19 - mmengine - INFO - Epoch(train) [100][2000/3757] lr: 1.0000e-05 eta: 0:04:57 time: 0.1917 data_time: 0.0131 memory: 7124 grad_norm: 6.9299 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9541 loss: 0.9541 2022/09/07 23:16:30 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 23:16:38 - mmengine - INFO - Epoch(train) [100][2100/3757] lr: 1.0000e-05 eta: 0:04:40 time: 0.1839 data_time: 0.0137 memory: 7124 grad_norm: 7.1165 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1122 loss: 1.1122 2022/09/07 23:16:58 - mmengine - INFO - Epoch(train) [100][2200/3757] lr: 1.0000e-05 eta: 0:04:23 time: 0.2020 data_time: 0.0126 memory: 7124 grad_norm: 7.1886 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9133 loss: 0.9133 2022/09/07 23:17:18 - mmengine - INFO - Epoch(train) [100][2300/3757] lr: 1.0000e-05 eta: 0:04:06 time: 0.1954 data_time: 0.0139 memory: 7124 grad_norm: 7.3792 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2605 loss: 1.2605 2022/09/07 23:17:37 - mmengine - INFO - Epoch(train) [100][2400/3757] lr: 1.0000e-05 eta: 0:03:50 time: 0.1905 data_time: 0.0153 memory: 7124 grad_norm: 7.0923 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9802 loss: 0.9802 2022/09/07 23:17:56 - mmengine - INFO - Epoch(train) [100][2500/3757] lr: 1.0000e-05 eta: 0:03:33 time: 0.1848 data_time: 0.0122 memory: 7124 grad_norm: 7.4657 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1949 loss: 1.1949 2022/09/07 23:18:15 - mmengine - INFO - Epoch(train) [100][2600/3757] lr: 1.0000e-05 eta: 0:03:16 time: 0.1907 data_time: 0.0130 memory: 7124 grad_norm: 7.2881 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1841 loss: 1.1841 2022/09/07 23:18:35 - mmengine - INFO - Epoch(train) [100][2700/3757] lr: 1.0000e-05 eta: 0:02:59 time: 0.1842 data_time: 0.0154 memory: 7124 grad_norm: 7.2781 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9359 loss: 0.9359 2022/09/07 23:18:54 - mmengine - INFO - Epoch(train) [100][2800/3757] lr: 1.0000e-05 eta: 0:02:42 time: 0.2027 data_time: 0.0139 memory: 7124 grad_norm: 7.2398 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0461 loss: 1.0461 2022/09/07 23:19:12 - mmengine - INFO - Epoch(train) [100][2900/3757] lr: 1.0000e-05 eta: 0:02:25 time: 0.1827 data_time: 0.0146 memory: 7124 grad_norm: 7.0900 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9385 loss: 0.9385 2022/09/07 23:19:31 - mmengine - INFO - Epoch(train) [100][3000/3757] lr: 1.0000e-05 eta: 0:02:08 time: 0.1874 data_time: 0.0156 memory: 7124 grad_norm: 7.0742 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9091 loss: 0.9091 2022/09/07 23:19:42 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 23:19:50 - mmengine - INFO - Epoch(train) [100][3100/3757] lr: 1.0000e-05 eta: 0:01:51 time: 0.1862 data_time: 0.0136 memory: 7124 grad_norm: 7.1625 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.0444 loss: 1.0444 2022/09/07 23:20:09 - mmengine - INFO - Epoch(train) [100][3200/3757] lr: 1.0000e-05 eta: 0:01:34 time: 0.1904 data_time: 0.0141 memory: 7124 grad_norm: 7.3028 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1011 loss: 1.1011 2022/09/07 23:20:28 - mmengine - INFO - Epoch(train) [100][3300/3757] lr: 1.0000e-05 eta: 0:01:17 time: 0.1871 data_time: 0.0145 memory: 7124 grad_norm: 7.4508 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9299 loss: 0.9299 2022/09/07 23:20:48 - mmengine - INFO - Epoch(train) [100][3400/3757] lr: 1.0000e-05 eta: 0:01:00 time: 0.1982 data_time: 0.0137 memory: 7124 grad_norm: 7.2416 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8302 loss: 0.8302 2022/09/07 23:21:06 - mmengine - INFO - Epoch(train) [100][3500/3757] lr: 1.0000e-05 eta: 0:00:43 time: 0.1802 data_time: 0.0127 memory: 7124 grad_norm: 7.1308 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0681 loss: 1.0681 2022/09/07 23:21:25 - mmengine - INFO - Epoch(train) [100][3600/3757] lr: 1.0000e-05 eta: 0:00:26 time: 0.1860 data_time: 0.0151 memory: 7124 grad_norm: 6.9155 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8464 loss: 0.8464 2022/09/07 23:21:44 - mmengine - INFO - Epoch(train) [100][3700/3757] lr: 1.0000e-05 eta: 0:00:09 time: 0.1861 data_time: 0.0142 memory: 7124 grad_norm: 7.3399 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2045 loss: 1.2045 2022/09/07 23:21:54 - mmengine - INFO - Exp name: tanet_imagenet-pretrained-r50_8xb8-dense-1x1x8-100e_kinetics400-rgb_20220906_171922 2022/09/07 23:21:54 - mmengine - INFO - Epoch(train) [100][3757/3757] lr: 1.0000e-05 eta: 0:00:02 time: 0.1455 data_time: 0.0075 memory: 7124 grad_norm: 7.2744 top1_acc: 0.7143 top5_acc: 0.8571 loss_cls: 1.0999 loss: 1.0999 2022/09/07 23:21:54 - mmengine - INFO - Saving checkpoint at 100 epochs 2022/09/07 23:24:42 - mmengine - INFO - Epoch(val) [100][100/310] eta: 0:04:32 time: 1.2984 data_time: 0.9917 memory: 7627 2022/09/07 23:27:29 - mmengine - INFO - Epoch(val) [100][200/310] eta: 0:03:03 time: 1.6704 data_time: 1.3594 memory: 7627 2022/09/07 23:30:00 - mmengine - INFO - Epoch(val) [100][300/310] eta: 0:00:13 time: 1.3022 data_time: 0.9962 memory: 7627 2022/09/07 23:30:17 - mmengine - INFO - Epoch(val) [100][310/310] acc/top1: 0.7542 acc/top5: 0.9191 acc/mean1: 0.7541