2023/06/04 01:48:15 - mmengine - INFO - ------------------------------------------------------------ System environment: sys.platform: linux Python: 3.9.0 (default, Nov 15 2020, 14:28:56) [GCC 7.3.0] CUDA available: True numpy_random_seed: 694122612 GPU 0,1,2,3,4,5,6,7: NVIDIA GeForce RTX 3090 CUDA_HOME: /usr/local/cuda NVCC: Cuda compilation tools, release 11.3, V11.3.109 GCC: gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0 PyTorch: 1.12.1+cu113 PyTorch compiling details: PyTorch built with: - GCC 9.3 - C++ Version: 201402 - Intel(R) Math Kernel Library Version 2020.0.0 Product Build 20191122 for Intel(R) 64 architecture applications - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) - 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_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 - CuDNN 8.2 - Built with CuDNN 8.3.2 - Magma 2.5.2 - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -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-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 -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.12.1, 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: 0.13.1+cu102 OpenCV: 4.7.0 MMEngine: 0.7.3 Runtime environment: cudnn_benchmark: False mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} dist_cfg: {'backend': 'nccl'} seed: None diff_rank_seed: False deterministic: False Distributed launcher: pytorch Distributed training: True GPU number: 32 ------------------------------------------------------------ 2023/06/04 01:48:16 - mmengine - INFO - Config: 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=3, save_best='auto', max_keep_ckpts=5), sampler_seed=dict(type='DistSamplerSeedHook'), sync_buffers=dict(type='SyncBuffersHook')) 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 = True num_frames = 8 model = dict( type='Recognizer3D', backbone=dict( type='UniFormerV2', input_resolution=224, patch_size=16, width=768, layers=12, heads=12, t_size=8, dw_reduction=1.5, backbone_drop_path_rate=0.0, temporal_downsample=False, no_lmhra=True, double_lmhra=True, return_list=[8, 9, 10, 11], n_layers=4, n_dim=768, n_head=12, mlp_factor=4.0, drop_path_rate=0.0, mlp_dropout=[0.5, 0.5, 0.5, 0.5], clip_pretrained=True, pretrained='ViT-B/16'), cls_head=dict( type='TimeSformerHead', dropout_ratio=0.5, num_classes=710, in_channels=768, average_clips='prob'), data_preprocessor=dict( type='ActionDataPreprocessor', mean=[114.75, 114.75, 114.75], std=[57.375, 57.375, 57.375], format_shape='NCTHW')) file_client_args = dict(io_backend='disk') train_pipeline = [ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=1), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict( type='PytorchVideoWrapper', op='RandAugment', magnitude=7, num_layers=4), dict(type='RandomResizedCrop'), dict(type='Resize', scale=(224, 224), keep_ratio=False), dict(type='Flip', flip_ratio=0.5), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ] val_pipeline = [ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=1, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='CenterCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ] test_pipeline = [ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=4, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='ThreeCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ] k400_data_root = 'data/kinetics400/videos_train' k600_data_root = 'data/kinetics600/videos' k700_data_root = 'data/kinetics700/videos' k400_data_root_val = 'data/kinetics400/videos_val' k600_data_root_val = 'data/kinetics600/videos' k700_data_root_val = 'data/kinetics700/videos' k400_ann_file_train = 'data/kinetics710/k400_train_list_videos.txt' k600_ann_file_train = 'data/kinetics710/k600_train_list_videos.txt' k700_ann_file_train = 'data/kinetics710/k700_train_list_videos.txt' k400_ann_file_val = 'data/kinetics710/k400_val_list_videos.txt' k600_ann_file_val = 'data/kinetics710/k600_val_list_videos.txt' k700_ann_file_val = 'data/kinetics710/k700_val_list_videos.txt' k400_trainset = dict( type='VideoDataset', ann_file='data/kinetics710/k400_train_list_videos.txt', data_prefix=dict(video='data/kinetics400/videos_train'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=1), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict( type='PytorchVideoWrapper', op='RandAugment', magnitude=7, num_layers=4), dict(type='RandomResizedCrop'), dict(type='Resize', scale=(224, 224), keep_ratio=False), dict(type='Flip', flip_ratio=0.5), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ]) k600_trainset = dict( type='VideoDataset', ann_file='data/kinetics710/k600_train_list_videos.txt', data_prefix=dict(video='data/kinetics600/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=1), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict( type='PytorchVideoWrapper', op='RandAugment', magnitude=7, num_layers=4), dict(type='RandomResizedCrop'), dict(type='Resize', scale=(224, 224), keep_ratio=False), dict(type='Flip', flip_ratio=0.5), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ]) k700_trainset = dict( type='VideoDataset', ann_file='data/kinetics710/k700_train_list_videos.txt', data_prefix=dict(video='data/kinetics700/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=1), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict( type='PytorchVideoWrapper', op='RandAugment', magnitude=7, num_layers=4), dict(type='RandomResizedCrop'), dict(type='Resize', scale=(224, 224), keep_ratio=False), dict(type='Flip', flip_ratio=0.5), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ]) k400_valset = dict( type='VideoDataset', ann_file='data/kinetics710/k400_val_list_videos.txt', data_prefix=dict(video='data/kinetics400/videos_val'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=1, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='CenterCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True) k600_valset = dict( type='VideoDataset', ann_file='data/kinetics710/k600_val_list_videos.txt', data_prefix=dict(video='data/kinetics600/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=1, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='CenterCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True) k700_valset = dict( type='VideoDataset', ann_file='data/kinetics710/k700_val_list_videos.txt', data_prefix=dict(video='data/kinetics700/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=1, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='CenterCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True) k400_testset = dict( type='VideoDataset', ann_file='data/kinetics710/k400_val_list_videos.txt', data_prefix=dict(video='data/kinetics400/videos_val'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=4, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='ThreeCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True) k600_testset = dict( type='VideoDataset', ann_file='data/kinetics710/k600_val_list_videos.txt', data_prefix=dict(video='data/kinetics600/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=4, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='ThreeCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True) k700_testset = dict( type='VideoDataset', ann_file='data/kinetics710/k700_val_list_videos.txt', data_prefix=dict(video='data/kinetics700/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=4, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='ThreeCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True) k710_trainset = dict( type='ConcatDataset', datasets=[ dict( type='VideoDataset', ann_file='data/kinetics710/k400_train_list_videos.txt', data_prefix=dict(video='data/kinetics400/videos_train'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=1), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict( type='PytorchVideoWrapper', op='RandAugment', magnitude=7, num_layers=4), dict(type='RandomResizedCrop'), dict(type='Resize', scale=(224, 224), keep_ratio=False), dict(type='Flip', flip_ratio=0.5), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ]), dict( type='VideoDataset', ann_file='data/kinetics710/k600_train_list_videos.txt', data_prefix=dict(video='data/kinetics600/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=1), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict( type='PytorchVideoWrapper', op='RandAugment', magnitude=7, num_layers=4), dict(type='RandomResizedCrop'), dict(type='Resize', scale=(224, 224), keep_ratio=False), dict(type='Flip', flip_ratio=0.5), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ]), dict( type='VideoDataset', ann_file='data/kinetics710/k700_train_list_videos.txt', data_prefix=dict(video='data/kinetics700/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=1), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict( type='PytorchVideoWrapper', op='RandAugment', magnitude=7, num_layers=4), dict(type='RandomResizedCrop'), dict(type='Resize', scale=(224, 224), keep_ratio=False), dict(type='Flip', flip_ratio=0.5), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ]) ]) k710_valset = dict( type='ConcatDataset', datasets=[ dict( type='VideoDataset', ann_file='data/kinetics710/k400_val_list_videos.txt', data_prefix=dict(video='data/kinetics400/videos_val'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict( type='UniformSample', clip_len=8, num_clips=1, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='CenterCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True), dict( type='VideoDataset', ann_file='data/kinetics710/k600_val_list_videos.txt', data_prefix=dict(video='data/kinetics600/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict( type='UniformSample', clip_len=8, num_clips=1, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='CenterCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True), dict( type='VideoDataset', ann_file='data/kinetics710/k700_val_list_videos.txt', data_prefix=dict(video='data/kinetics700/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict( type='UniformSample', clip_len=8, num_clips=1, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='CenterCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True) ]) k710_testset = dict( type='ConcatDataset', datasets=[ dict( type='VideoDataset', ann_file='data/kinetics710/k400_val_list_videos.txt', data_prefix=dict(video='data/kinetics400/videos_val'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict( type='UniformSample', clip_len=8, num_clips=4, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='ThreeCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True), dict( type='VideoDataset', ann_file='data/kinetics710/k600_val_list_videos.txt', data_prefix=dict(video='data/kinetics600/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict( type='UniformSample', clip_len=8, num_clips=4, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='ThreeCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True), dict( type='VideoDataset', ann_file='data/kinetics710/k700_val_list_videos.txt', data_prefix=dict(video='data/kinetics700/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict( type='UniformSample', clip_len=8, num_clips=4, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='ThreeCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True) ]) train_dataloader = dict( batch_size=8, num_workers=8, persistent_workers=True, sampler=dict(type='DefaultSampler', shuffle=True), dataset=dict( type='ConcatDataset', datasets=[ dict( type='VideoDataset', ann_file='data/kinetics710/k400_train_list_videos.txt', data_prefix=dict(video='data/kinetics400/videos_train'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=1), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict( type='PytorchVideoWrapper', op='RandAugment', magnitude=7, num_layers=4), dict(type='RandomResizedCrop'), dict(type='Resize', scale=(224, 224), keep_ratio=False), dict(type='Flip', flip_ratio=0.5), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ]), dict( type='VideoDataset', ann_file='data/kinetics710/k600_train_list_videos.txt', data_prefix=dict(video='data/kinetics600/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=1), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict( type='PytorchVideoWrapper', op='RandAugment', magnitude=7, num_layers=4), dict(type='RandomResizedCrop'), dict(type='Resize', scale=(224, 224), keep_ratio=False), dict(type='Flip', flip_ratio=0.5), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ]), dict( type='VideoDataset', ann_file='data/kinetics710/k700_train_list_videos.txt', data_prefix=dict(video='data/kinetics700/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict(type='UniformSample', clip_len=8, num_clips=1), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 256)), dict( type='PytorchVideoWrapper', op='RandAugment', magnitude=7, num_layers=4), dict(type='RandomResizedCrop'), dict(type='Resize', scale=(224, 224), keep_ratio=False), dict(type='Flip', flip_ratio=0.5), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ]) ])) val_dataloader = dict( batch_size=8, num_workers=8, persistent_workers=True, sampler=dict(type='DefaultSampler', shuffle=False), dataset=dict( type='ConcatDataset', datasets=[ dict( type='VideoDataset', ann_file='data/kinetics710/k400_val_list_videos.txt', data_prefix=dict(video='data/kinetics400/videos_val'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict( type='UniformSample', clip_len=8, num_clips=1, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='CenterCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True), dict( type='VideoDataset', ann_file='data/kinetics710/k600_val_list_videos.txt', data_prefix=dict(video='data/kinetics600/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict( type='UniformSample', clip_len=8, num_clips=1, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='CenterCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True), dict( type='VideoDataset', ann_file='data/kinetics710/k700_val_list_videos.txt', data_prefix=dict(video='data/kinetics700/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict( type='UniformSample', clip_len=8, num_clips=1, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='CenterCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), 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='ConcatDataset', datasets=[ dict( type='VideoDataset', ann_file='data/kinetics710/k400_val_list_videos.txt', data_prefix=dict(video='data/kinetics400/videos_val'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict( type='UniformSample', clip_len=8, num_clips=4, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='ThreeCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True), dict( type='VideoDataset', ann_file='data/kinetics710/k600_val_list_videos.txt', data_prefix=dict(video='data/kinetics600/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict( type='UniformSample', clip_len=8, num_clips=4, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='ThreeCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True), dict( type='VideoDataset', ann_file='data/kinetics710/k700_val_list_videos.txt', data_prefix=dict(video='data/kinetics700/videos'), pipeline=[ dict(type='DecordInit', io_backend='disk'), dict( type='UniformSample', clip_len=8, num_clips=4, test_mode=True), dict(type='DecordDecode'), dict(type='Resize', scale=(-1, 224)), dict(type='ThreeCrop', crop_size=224), dict(type='FormatShape', input_format='NCTHW'), dict(type='PackActionInputs') ], test_mode=True) ])) val_evaluator = dict(type='AccMetric') test_evaluator = dict(type='AccMetric') train_cfg = dict( type='EpochBasedTrainLoop', max_epochs=55, val_begin=1, val_interval=1) val_cfg = dict(type='ValLoop') test_cfg = dict(type='TestLoop') base_lr = 1e-05 optim_wrapper = dict( optimizer=dict( type='AdamW', lr=1e-05, betas=(0.9, 0.999), weight_decay=0.05), paramwise_cfg=dict(norm_decay_mult=0.0, bias_decay_mult=0.0), clip_grad=dict(max_norm=20, norm_type=2)) param_scheduler = [ dict( type='LinearLR', start_factor=0.5, by_epoch=True, begin=0, end=5, convert_to_iter_based=True), dict( type='CosineAnnealingLR', T_max=50, eta_min_ratio=0.5, by_epoch=True, begin=5, end=55, convert_to_iter_based=True) ] auto_scale_lr = dict(enable=True, base_batch_size=256) launcher = 'pytorch' work_dir = './work_dirs/uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train' randomness = dict(seed=None, diff_rank_seed=False, deterministic=False) 2023/06/04 01:48:17 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:17 - mmengine - INFO - No L_MHRA: True 2023/06/04 01:48:17 - mmengine - INFO - Double L_MHRA: True 2023/06/04 01:48:17 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:17 - mmengine - INFO - No L_MHRA: True 2023/06/04 01:48:17 - mmengine - INFO - Double L_MHRA: True 2023/06/04 01:48:17 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:17 - mmengine - INFO - No L_MHRA: True 2023/06/04 01:48:17 - mmengine - INFO - Double L_MHRA: True 2023/06/04 01:48:17 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:17 - mmengine - INFO - No L_MHRA: True 2023/06/04 01:48:17 - mmengine - INFO - Double L_MHRA: True 2023/06/04 01:48:17 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:17 - mmengine - INFO - No L_MHRA: True 2023/06/04 01:48:17 - mmengine - INFO - Double L_MHRA: True 2023/06/04 01:48:18 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:18 - mmengine - INFO - No L_MHRA: True 2023/06/04 01:48:18 - mmengine - INFO - Double L_MHRA: True 2023/06/04 01:48:18 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:18 - mmengine - INFO - No L_MHRA: True 2023/06/04 01:48:18 - mmengine - INFO - Double L_MHRA: True 2023/06/04 01:48:18 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:18 - mmengine - INFO - No L_MHRA: True 2023/06/04 01:48:18 - mmengine - INFO - Double L_MHRA: True 2023/06/04 01:48:18 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:18 - mmengine - INFO - No L_MHRA: True 2023/06/04 01:48:18 - mmengine - INFO - Double L_MHRA: True 2023/06/04 01:48:18 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:18 - mmengine - INFO - No L_MHRA: True 2023/06/04 01:48:18 - mmengine - INFO - Double L_MHRA: True 2023/06/04 01:48:18 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:18 - mmengine - INFO - No L_MHRA: True 2023/06/04 01:48:18 - mmengine - INFO - Double L_MHRA: True 2023/06/04 01:48:18 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:18 - mmengine - INFO - No L_MHRA: True 2023/06/04 01:48:18 - mmengine - INFO - Double L_MHRA: True 2023/06/04 01:48:18 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:18 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:18 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:18 - mmengine - INFO - Drop path rate: 0.0 2023/06/04 01:48:19 - mmengine - INFO - Hooks will be executed in the following order: before_run: (VERY_HIGH ) RuntimeInfoHook (BELOW_NORMAL) LoggerHook -------------------- before_train: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (VERY_LOW ) CheckpointHook -------------------- before_train_epoch: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (NORMAL ) DistSamplerSeedHook -------------------- before_train_iter: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook -------------------- after_train_iter: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (BELOW_NORMAL) LoggerHook (LOW ) ParamSchedulerHook (VERY_LOW ) CheckpointHook -------------------- after_train_epoch: (NORMAL ) IterTimerHook (NORMAL ) SyncBuffersHook (LOW ) ParamSchedulerHook (VERY_LOW ) CheckpointHook -------------------- before_val_epoch: (NORMAL ) IterTimerHook (NORMAL ) SyncBuffersHook -------------------- before_val_iter: (NORMAL ) IterTimerHook -------------------- after_val_iter: (NORMAL ) IterTimerHook (BELOW_NORMAL) LoggerHook -------------------- after_val_epoch: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (BELOW_NORMAL) LoggerHook (LOW ) ParamSchedulerHook (VERY_LOW ) CheckpointHook -------------------- after_train: (VERY_LOW ) CheckpointHook -------------------- before_test_epoch: (NORMAL ) IterTimerHook -------------------- before_test_iter: (NORMAL ) IterTimerHook -------------------- after_test_iter: (NORMAL ) IterTimerHook (BELOW_NORMAL) LoggerHook -------------------- after_test_epoch: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (BELOW_NORMAL) LoggerHook -------------------- after_run: (BELOW_NORMAL) LoggerHook -------------------- 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.ln_pre.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.ln_pre.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.0.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.0.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.0.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.0.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.0.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.0.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.0.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.1.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.1.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.1.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.1.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.1.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.1.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.1.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.2.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.2.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.2.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.2.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.2.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.2.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.2.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.3.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.3.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.3.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.3.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.3.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.3.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.3.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.4.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.4.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.4.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.4.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.4.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.4.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.4.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.5.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.5.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.5.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.5.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.5.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.5.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.5.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.6.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.6.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.6.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.6.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.6.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.6.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.6.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.7.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.7.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.7.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.7.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.7.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.7.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.7.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.8.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.8.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.8.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.8.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.8.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.8.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.8.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.9.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.9.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.9.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.9.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.9.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.9.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.9.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.10.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.10.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.10.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.10.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.10.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.10.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.10.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.11.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.11.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.11.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:22 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.11.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.11.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.11.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.resblocks.11.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dpe.0.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dpe.1.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dpe.2.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dpe.3.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.0.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.0.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.0.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.0.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.0.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.0.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.0.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.0.ln_3.weight:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.0.ln_3.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.1.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.1.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.1.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.1.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.1.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.1.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.1.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.1.ln_3.weight:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.1.ln_3.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.2.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.2.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.2.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.2.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.2.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.2.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.2.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.2.ln_3.weight:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.2.ln_3.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.3.attn.out_proj.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.3.ln_1.weight:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.3.ln_1.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.3.mlp.c_fc.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.3.mlp.c_proj.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.3.ln_2.weight:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.3.ln_2.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.3.ln_3.weight:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.dec.3.ln_3.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.norm.weight:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- backbone.transformer.norm.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - paramwise_options -- cls_head.fc_cls.bias:weight_decay=0.0 2023/06/04 01:48:23 - mmengine - INFO - LR is set based on batch size of 256 and the current batch size is 256. Scaling the original LR by 1.0. 2023/06/04 01:48:24 - mmengine - INFO - load model from: ViT-B/16 2023/06/04 01:48:24 - mmengine - INFO - Load CLIP pretrained model from https://download.openmmlab.com/mmaction/v1.0/recognition/uniformerv2/clipVisualEncoder/vit-base-p16-res224_clip-rgb_20221219-b8a5da86.pth 2023/06/04 01:48:26 - mmengine - INFO - Inflate: conv1.weight, torch.Size([768, 3, 16, 16]) => torch.Size([768, 3, 1, 16, 16]) 2023/06/04 01:48:26 - mmengine - INFO - Init center: True Name of parameter - Initialization information backbone.class_embedding - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.positional_embedding - torch.Size([197, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.conv1.weight - torch.Size([768, 3, 1, 16, 16]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.ln_pre.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.ln_pre.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.temporal_cls_token - torch.Size([1, 1, 768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.balance - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.resblocks.0.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.0.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.0.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.0.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.0.ln_1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.0.ln_1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.0.mlp.c_fc.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.0.mlp.c_fc.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.0.mlp.c_proj.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.0.mlp.c_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.0.ln_2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.0.ln_2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.1.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.1.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.1.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.1.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.1.ln_1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.1.ln_1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.1.mlp.c_fc.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.1.mlp.c_fc.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.1.mlp.c_proj.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.1.mlp.c_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.1.ln_2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.1.ln_2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.2.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.2.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.2.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.2.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.2.ln_1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.2.ln_1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.2.mlp.c_fc.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.2.mlp.c_fc.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.2.mlp.c_proj.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.2.mlp.c_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.2.ln_2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.2.ln_2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.3.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.3.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.3.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.3.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.3.ln_1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.3.ln_1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.3.mlp.c_fc.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.3.mlp.c_fc.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.3.mlp.c_proj.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.3.mlp.c_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.3.ln_2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.3.ln_2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.4.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.4.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.4.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.4.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.4.ln_1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.4.ln_1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.4.mlp.c_fc.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.4.mlp.c_fc.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.4.mlp.c_proj.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.4.mlp.c_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.4.ln_2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.4.ln_2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.5.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.5.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.5.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.5.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.5.ln_1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.5.ln_1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.5.mlp.c_fc.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.5.mlp.c_fc.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.5.mlp.c_proj.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.5.mlp.c_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.5.ln_2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.5.ln_2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.6.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.6.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.6.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.6.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.6.ln_1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.6.ln_1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.6.mlp.c_fc.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.6.mlp.c_fc.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.6.mlp.c_proj.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.6.mlp.c_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.6.ln_2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.6.ln_2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.7.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.7.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.7.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.7.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.7.ln_1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.7.ln_1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.7.mlp.c_fc.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.7.mlp.c_fc.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.7.mlp.c_proj.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.7.mlp.c_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.7.ln_2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.7.ln_2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.8.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.8.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.8.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.8.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.8.ln_1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.8.ln_1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.8.mlp.c_fc.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.8.mlp.c_fc.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.8.mlp.c_proj.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.8.mlp.c_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.8.ln_2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.8.ln_2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.9.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.9.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.9.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.9.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.9.ln_1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.9.ln_1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.9.mlp.c_fc.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.9.mlp.c_fc.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.9.mlp.c_proj.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.9.mlp.c_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.9.ln_2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.9.ln_2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.10.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.10.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.10.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.10.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.10.ln_1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.10.ln_1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.10.mlp.c_fc.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.10.mlp.c_fc.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.10.mlp.c_proj.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.10.mlp.c_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.10.ln_2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.10.ln_2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.11.attn.in_proj_weight - torch.Size([2304, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.11.attn.in_proj_bias - torch.Size([2304]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.11.attn.out_proj.weight - torch.Size([768, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.11.attn.out_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.11.ln_1.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.11.ln_1.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.11.mlp.c_fc.weight - torch.Size([3072, 768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.11.mlp.c_fc.bias - torch.Size([3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.11.mlp.c_proj.weight - torch.Size([768, 3072]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.11.mlp.c_proj.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.11.ln_2.weight - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.resblocks.11.ln_2.bias - torch.Size([768]): Initialized by user-defined `init_weights` in UniFormerV2 backbone.transformer.dpe.0.weight - torch.Size([768, 1, 3, 3, 3]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dpe.0.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dpe.1.weight - torch.Size([768, 1, 3, 3, 3]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dpe.1.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dpe.2.weight - torch.Size([768, 1, 3, 3, 3]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dpe.2.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dpe.3.weight - torch.Size([768, 1, 3, 3, 3]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dpe.3.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.0.attn.in_proj_weight - torch.Size([2304, 768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.0.attn.in_proj_bias - torch.Size([2304]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.0.attn.out_proj.weight - torch.Size([768, 768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.0.attn.out_proj.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.0.ln_1.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.0.ln_1.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.0.mlp.c_fc.weight - torch.Size([3072, 768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.0.mlp.c_fc.bias - torch.Size([3072]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.0.mlp.c_proj.weight - torch.Size([768, 3072]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.0.mlp.c_proj.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.0.ln_2.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.0.ln_2.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.0.ln_3.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.0.ln_3.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.1.attn.in_proj_weight - torch.Size([2304, 768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.1.attn.in_proj_bias - torch.Size([2304]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.1.attn.out_proj.weight - torch.Size([768, 768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.1.attn.out_proj.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.1.ln_1.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.1.ln_1.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.1.mlp.c_fc.weight - torch.Size([3072, 768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.1.mlp.c_fc.bias - torch.Size([3072]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.1.mlp.c_proj.weight - torch.Size([768, 3072]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.1.mlp.c_proj.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.1.ln_2.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.1.ln_2.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.1.ln_3.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.1.ln_3.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.2.attn.in_proj_weight - torch.Size([2304, 768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.2.attn.in_proj_bias - torch.Size([2304]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.2.attn.out_proj.weight - torch.Size([768, 768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.2.attn.out_proj.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.2.ln_1.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.2.ln_1.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.2.mlp.c_fc.weight - torch.Size([3072, 768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.2.mlp.c_fc.bias - torch.Size([3072]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.2.mlp.c_proj.weight - torch.Size([768, 3072]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.2.mlp.c_proj.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.2.ln_2.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.2.ln_2.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.2.ln_3.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.2.ln_3.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.3.attn.in_proj_weight - torch.Size([2304, 768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.3.attn.in_proj_bias - torch.Size([2304]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.3.attn.out_proj.weight - torch.Size([768, 768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.3.attn.out_proj.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.3.ln_1.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.3.ln_1.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.3.mlp.c_fc.weight - torch.Size([3072, 768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.3.mlp.c_fc.bias - torch.Size([3072]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.3.mlp.c_proj.weight - torch.Size([768, 3072]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.3.mlp.c_proj.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.3.ln_2.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.3.ln_2.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.3.ln_3.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.dec.3.ln_3.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.norm.weight - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D backbone.transformer.norm.bias - torch.Size([768]): The value is the same before and after calling `init_weights` of Recognizer3D cls_head.fc_cls.weight - torch.Size([710, 768]): Initialized by user-defined `init_weights` in TimeSformerHead cls_head.fc_cls.bias - torch.Size([710]): Initialized by user-defined `init_weights` in TimeSformerHead 2023/06/04 01:48:26 - mmengine - INFO - Auto resumed from the latest checkpoint /mnt/data/mmact/lilin/Repos/mmaction2/work_dirs/uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train/epoch_42.pth. 2023/06/04 01:48:28 - mmengine - INFO - Load checkpoint from /mnt/data/mmact/lilin/Repos/mmaction2/work_dirs/uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train/epoch_42.pth 2023/06/04 01:48:28 - mmengine - INFO - resumed epoch: 42, iter: 107898 2023/06/04 01:48:28 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io 2023/06/04 01:48:28 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. 2023/06/04 01:48:28 - mmengine - INFO - Checkpoints will be saved to /mnt/data/mmact/lilin/Repos/mmaction2/work_dirs/uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train. 2023/06/04 01:50:13 - mmengine - INFO - Epoch(train) [43][ 100/2569] lr: 5.7842e-06 eta: 9:45:12 time: 0.9357 data_time: 0.0076 memory: 13405 grad_norm: 26.0640 loss: 1.0965 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0965 2023/06/04 01:50:15 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 01:51:44 - mmengine - INFO - Epoch(train) [43][ 200/2569] lr: 5.7798e-06 eta: 9:03:31 time: 0.9139 data_time: 0.0078 memory: 13405 grad_norm: 26.0578 loss: 1.1759 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1759 2023/06/04 01:53:16 - mmengine - INFO - Epoch(train) [43][ 300/2569] lr: 5.7754e-06 eta: 8:50:42 time: 0.9098 data_time: 0.0079 memory: 13405 grad_norm: 25.9340 loss: 1.1815 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1815 2023/06/04 01:54:49 - mmengine - INFO - Epoch(train) [43][ 400/2569] lr: 5.7709e-06 eta: 8:43:35 time: 0.9175 data_time: 0.0084 memory: 13405 grad_norm: 25.5845 loss: 0.9251 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.9251 2023/06/04 01:56:20 - mmengine - INFO - Epoch(train) [43][ 500/2569] lr: 5.7665e-06 eta: 8:38:12 time: 0.9353 data_time: 0.0080 memory: 13405 grad_norm: 26.3787 loss: 1.1125 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1125 2023/06/04 01:57:53 - mmengine - INFO - Epoch(train) [43][ 600/2569] lr: 5.7621e-06 eta: 8:35:14 time: 0.9414 data_time: 0.0081 memory: 13405 grad_norm: 26.0326 loss: 1.3601 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.3601 2023/06/04 01:59:25 - mmengine - INFO - Epoch(train) [43][ 700/2569] lr: 5.7577e-06 eta: 8:32:01 time: 0.9172 data_time: 0.0077 memory: 13405 grad_norm: 26.4677 loss: 1.1279 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1279 2023/06/04 02:00:57 - mmengine - INFO - Epoch(train) [43][ 800/2569] lr: 5.7533e-06 eta: 8:28:58 time: 0.9121 data_time: 0.0081 memory: 13405 grad_norm: 26.8986 loss: 1.0488 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0488 2023/06/04 02:02:30 - mmengine - INFO - Epoch(train) [43][ 900/2569] lr: 5.7490e-06 eta: 8:26:51 time: 0.9502 data_time: 0.0077 memory: 13405 grad_norm: 26.3492 loss: 1.3559 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.3559 2023/06/04 02:04:03 - mmengine - INFO - Epoch(train) [43][1000/2569] lr: 5.7446e-06 eta: 8:24:45 time: 0.9153 data_time: 0.0079 memory: 13405 grad_norm: 26.5087 loss: 1.2298 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2298 2023/06/04 02:05:34 - mmengine - INFO - Epoch(train) [43][1100/2569] lr: 5.7403e-06 eta: 8:22:20 time: 0.9140 data_time: 0.0089 memory: 13405 grad_norm: 27.0140 loss: 0.8141 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.8141 2023/06/04 02:05:36 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 02:07:06 - mmengine - INFO - Epoch(train) [43][1200/2569] lr: 5.7359e-06 eta: 8:20:10 time: 0.9399 data_time: 0.0081 memory: 13405 grad_norm: 26.7424 loss: 1.1084 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1084 2023/06/04 02:08:39 - mmengine - INFO - Epoch(train) [43][1300/2569] lr: 5.7316e-06 eta: 8:18:34 time: 0.9358 data_time: 0.0077 memory: 13405 grad_norm: 26.8080 loss: 1.3592 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.3592 2023/06/04 02:10:12 - mmengine - INFO - Epoch(train) [43][1400/2569] lr: 5.7273e-06 eta: 8:16:44 time: 0.9220 data_time: 0.0081 memory: 13405 grad_norm: 27.2627 loss: 1.2030 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2030 2023/06/04 02:11:45 - mmengine - INFO - Epoch(train) [43][1500/2569] lr: 5.7230e-06 eta: 8:15:04 time: 0.9373 data_time: 0.0080 memory: 13405 grad_norm: 27.0519 loss: 1.1433 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1433 2023/06/04 02:13:19 - mmengine - INFO - Epoch(train) [43][1600/2569] lr: 5.7187e-06 eta: 8:13:56 time: 0.9597 data_time: 0.0081 memory: 13405 grad_norm: 25.6243 loss: 1.2296 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2296 2023/06/04 02:14:51 - mmengine - INFO - Epoch(train) [43][1700/2569] lr: 5.7144e-06 eta: 8:11:59 time: 0.9134 data_time: 0.0080 memory: 13405 grad_norm: 25.7143 loss: 1.0287 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0287 2023/06/04 02:16:23 - mmengine - INFO - Epoch(train) [43][1800/2569] lr: 5.7101e-06 eta: 8:10:00 time: 0.9145 data_time: 0.0083 memory: 13405 grad_norm: 26.5313 loss: 1.1962 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1962 2023/06/04 02:17:55 - mmengine - INFO - Epoch(train) [43][1900/2569] lr: 5.7059e-06 eta: 8:08:16 time: 0.9463 data_time: 0.0080 memory: 13405 grad_norm: 25.6343 loss: 1.1356 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1356 2023/06/04 02:19:27 - mmengine - INFO - Epoch(train) [43][2000/2569] lr: 5.7016e-06 eta: 8:06:25 time: 0.9255 data_time: 0.0083 memory: 13405 grad_norm: 26.3425 loss: 0.9808 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9808 2023/06/04 02:21:00 - mmengine - INFO - Epoch(train) [43][2100/2569] lr: 5.6974e-06 eta: 8:04:48 time: 0.9351 data_time: 0.0079 memory: 13405 grad_norm: 26.7055 loss: 0.9832 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9832 2023/06/04 02:21:02 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 02:22:33 - mmengine - INFO - Epoch(train) [43][2200/2569] lr: 5.6931e-06 eta: 8:03:10 time: 0.9201 data_time: 0.0080 memory: 13405 grad_norm: 26.5359 loss: 1.1836 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1836 2023/06/04 02:24:04 - mmengine - INFO - Epoch(train) [43][2300/2569] lr: 5.6889e-06 eta: 8:01:21 time: 0.9114 data_time: 0.0083 memory: 13405 grad_norm: 26.9925 loss: 1.3022 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3022 2023/06/04 02:25:37 - mmengine - INFO - Epoch(train) [43][2400/2569] lr: 5.6847e-06 eta: 7:59:42 time: 0.9180 data_time: 0.0080 memory: 13405 grad_norm: 26.4819 loss: 1.2918 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.2918 2023/06/04 02:27:10 - mmengine - INFO - Epoch(train) [43][2500/2569] lr: 5.6805e-06 eta: 7:58:13 time: 0.9284 data_time: 0.0079 memory: 13405 grad_norm: 26.0738 loss: 1.3345 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.3345 2023/06/04 02:28:13 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 02:28:13 - mmengine - INFO - Epoch(train) [43][2569/2569] lr: 5.6776e-06 eta: 7:57:00 time: 0.8951 data_time: 0.0080 memory: 13405 grad_norm: 27.1413 loss: 1.0403 top1_acc: 0.8333 top5_acc: 1.0000 loss_cls: 1.0403 2023/06/04 02:28:39 - mmengine - INFO - Epoch(val) [43][100/260] eta: 0:00:41 time: 0.2423 data_time: 0.0488 memory: 2900 2023/06/04 02:29:02 - mmengine - INFO - Epoch(val) [43][200/260] eta: 0:00:14 time: 0.2023 data_time: 0.0084 memory: 2900 2023/06/04 02:29:24 - mmengine - INFO - Epoch(val) [43][260/260] acc/top1: 0.7698 acc/top5: 0.9327 acc/mean1: 0.7654 data_time: 0.0459 time: 0.2389 2023/06/04 02:31:00 - mmengine - INFO - Epoch(train) [44][ 100/2569] lr: 5.6734e-06 eta: 7:56:08 time: 0.9105 data_time: 0.0077 memory: 13405 grad_norm: 25.4331 loss: 1.4405 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.4405 2023/06/04 02:32:33 - mmengine - INFO - Epoch(train) [44][ 200/2569] lr: 5.6693e-06 eta: 7:54:31 time: 0.9145 data_time: 0.0078 memory: 13405 grad_norm: 26.2758 loss: 0.9478 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9478 2023/06/04 02:34:06 - mmengine - INFO - Epoch(train) [44][ 300/2569] lr: 5.6651e-06 eta: 7:52:56 time: 0.9123 data_time: 0.0079 memory: 13405 grad_norm: 26.0977 loss: 1.2081 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2081 2023/06/04 02:35:38 - mmengine - INFO - Epoch(train) [44][ 400/2569] lr: 5.6610e-06 eta: 7:51:18 time: 0.9307 data_time: 0.0079 memory: 13405 grad_norm: 26.6066 loss: 0.9679 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9679 2023/06/04 02:37:12 - mmengine - INFO - Epoch(train) [44][ 500/2569] lr: 5.6568e-06 eta: 7:49:49 time: 0.9403 data_time: 0.0080 memory: 13405 grad_norm: 25.4408 loss: 0.9877 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9877 2023/06/04 02:37:42 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 02:38:44 - mmengine - INFO - Epoch(train) [44][ 600/2569] lr: 5.6527e-06 eta: 7:48:10 time: 0.9169 data_time: 0.0081 memory: 13405 grad_norm: 26.6185 loss: 1.1014 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1014 2023/06/04 02:40:15 - mmengine - INFO - Epoch(train) [44][ 700/2569] lr: 5.6486e-06 eta: 7:46:20 time: 0.9134 data_time: 0.0077 memory: 13405 grad_norm: 25.5819 loss: 0.9233 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9233 2023/06/04 02:41:47 - mmengine - INFO - Epoch(train) [44][ 800/2569] lr: 5.6445e-06 eta: 7:44:39 time: 0.9211 data_time: 0.0078 memory: 13405 grad_norm: 25.6651 loss: 1.2545 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.2545 2023/06/04 02:43:19 - mmengine - INFO - Epoch(train) [44][ 900/2569] lr: 5.6404e-06 eta: 7:43:00 time: 0.9432 data_time: 0.0081 memory: 13405 grad_norm: 26.5678 loss: 1.0859 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.0859 2023/06/04 02:44:51 - mmengine - INFO - Epoch(train) [44][1000/2569] lr: 5.6363e-06 eta: 7:41:17 time: 0.9279 data_time: 0.0080 memory: 13405 grad_norm: 26.7668 loss: 1.2122 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2122 2023/06/04 02:46:25 - mmengine - INFO - Epoch(train) [44][1100/2569] lr: 5.6322e-06 eta: 7:39:56 time: 0.9136 data_time: 0.0083 memory: 13405 grad_norm: 27.3992 loss: 1.1962 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1962 2023/06/04 02:47:57 - mmengine - INFO - Epoch(train) [44][1200/2569] lr: 5.6282e-06 eta: 7:38:14 time: 0.9251 data_time: 0.0090 memory: 13405 grad_norm: 26.0082 loss: 1.0211 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0211 2023/06/04 02:49:29 - mmengine - INFO - Epoch(train) [44][1300/2569] lr: 5.6241e-06 eta: 7:36:40 time: 0.9013 data_time: 0.0079 memory: 13405 grad_norm: 27.1998 loss: 1.0663 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0663 2023/06/04 02:51:03 - mmengine - INFO - Epoch(train) [44][1400/2569] lr: 5.6201e-06 eta: 7:35:11 time: 0.9259 data_time: 0.0083 memory: 13405 grad_norm: 27.6346 loss: 1.2861 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2861 2023/06/04 02:52:36 - mmengine - INFO - Epoch(train) [44][1500/2569] lr: 5.6161e-06 eta: 7:33:39 time: 0.9101 data_time: 0.0078 memory: 13405 grad_norm: 26.3446 loss: 1.1532 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1532 2023/06/04 02:53:06 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 02:54:08 - mmengine - INFO - Epoch(train) [44][1600/2569] lr: 5.6121e-06 eta: 7:32:05 time: 0.9345 data_time: 0.0082 memory: 13405 grad_norm: 26.4866 loss: 0.8419 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8419 2023/06/04 02:55:42 - mmengine - INFO - Epoch(train) [44][1700/2569] lr: 5.6081e-06 eta: 7:30:36 time: 0.9518 data_time: 0.0079 memory: 13405 grad_norm: 26.1319 loss: 0.8893 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8893 2023/06/04 02:57:14 - mmengine - INFO - Epoch(train) [44][1800/2569] lr: 5.6041e-06 eta: 7:28:59 time: 0.9195 data_time: 0.0086 memory: 13405 grad_norm: 26.2802 loss: 1.1639 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1639 2023/06/04 02:58:46 - mmengine - INFO - Epoch(train) [44][1900/2569] lr: 5.6001e-06 eta: 7:27:21 time: 0.9268 data_time: 0.0078 memory: 13405 grad_norm: 25.5920 loss: 1.1077 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.1077 2023/06/04 03:00:19 - mmengine - INFO - Epoch(train) [44][2000/2569] lr: 5.5961e-06 eta: 7:25:49 time: 0.9037 data_time: 0.0080 memory: 13405 grad_norm: 26.5169 loss: 1.1484 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.1484 2023/06/04 03:01:51 - mmengine - INFO - Epoch(train) [44][2100/2569] lr: 5.5922e-06 eta: 7:24:14 time: 0.9414 data_time: 0.0079 memory: 13405 grad_norm: 26.1174 loss: 0.9963 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9963 2023/06/04 03:03:24 - mmengine - INFO - Epoch(train) [44][2200/2569] lr: 5.5882e-06 eta: 7:22:41 time: 0.9413 data_time: 0.0081 memory: 13405 grad_norm: 26.5770 loss: 1.4525 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.4525 2023/06/04 03:04:57 - mmengine - INFO - Epoch(train) [44][2300/2569] lr: 5.5843e-06 eta: 7:21:09 time: 0.9173 data_time: 0.0078 memory: 13405 grad_norm: 26.4542 loss: 1.1520 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1520 2023/06/04 03:06:29 - mmengine - INFO - Epoch(train) [44][2400/2569] lr: 5.5804e-06 eta: 7:19:30 time: 0.9391 data_time: 0.0078 memory: 13405 grad_norm: 26.2394 loss: 1.1809 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1809 2023/06/04 03:08:01 - mmengine - INFO - Epoch(train) [44][2500/2569] lr: 5.5764e-06 eta: 7:17:55 time: 0.9155 data_time: 0.0085 memory: 13405 grad_norm: 26.7268 loss: 1.3003 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.3003 2023/06/04 03:08:32 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 03:09:04 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 03:09:04 - mmengine - INFO - Epoch(train) [44][2569/2569] lr: 5.5738e-06 eta: 7:16:46 time: 0.9031 data_time: 0.0085 memory: 13405 grad_norm: 26.8904 loss: 1.3260 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.3260 2023/06/04 03:09:30 - mmengine - INFO - Epoch(val) [44][100/260] eta: 0:00:40 time: 0.2459 data_time: 0.0517 memory: 2900 2023/06/04 03:09:52 - mmengine - INFO - Epoch(val) [44][200/260] eta: 0:00:14 time: 0.2044 data_time: 0.0108 memory: 2900 2023/06/04 03:10:15 - mmengine - INFO - Epoch(val) [44][260/260] acc/top1: 0.7701 acc/top5: 0.9318 acc/mean1: 0.7661 data_time: 0.0414 time: 0.2346 2023/06/04 03:11:51 - mmengine - INFO - Epoch(train) [45][ 100/2569] lr: 5.5699e-06 eta: 7:15:30 time: 0.9215 data_time: 0.0080 memory: 13405 grad_norm: 26.1772 loss: 1.0614 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0614 2023/06/04 03:13:21 - mmengine - INFO - Epoch(train) [45][ 200/2569] lr: 5.5660e-06 eta: 7:13:47 time: 0.9048 data_time: 0.0080 memory: 13405 grad_norm: 26.0432 loss: 1.3999 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.3999 2023/06/04 03:14:53 - mmengine - INFO - Epoch(train) [45][ 300/2569] lr: 5.5621e-06 eta: 7:12:06 time: 0.9061 data_time: 0.0080 memory: 13405 grad_norm: 26.2601 loss: 1.1833 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1833 2023/06/04 03:16:24 - mmengine - INFO - Epoch(train) [45][ 400/2569] lr: 5.5583e-06 eta: 7:10:26 time: 0.9163 data_time: 0.0082 memory: 13405 grad_norm: 25.6542 loss: 1.2034 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2034 2023/06/04 03:17:56 - mmengine - INFO - Epoch(train) [45][ 500/2569] lr: 5.5544e-06 eta: 7:08:50 time: 0.9292 data_time: 0.0078 memory: 13405 grad_norm: 26.2613 loss: 1.3439 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.3439 2023/06/04 03:19:28 - mmengine - INFO - Epoch(train) [45][ 600/2569] lr: 5.5506e-06 eta: 7:07:13 time: 0.9189 data_time: 0.0086 memory: 13405 grad_norm: 25.4439 loss: 1.2727 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2727 2023/06/04 03:21:01 - mmengine - INFO - Epoch(train) [45][ 700/2569] lr: 5.5468e-06 eta: 7:05:41 time: 0.9518 data_time: 0.0083 memory: 13405 grad_norm: 25.4343 loss: 0.8482 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8482 2023/06/04 03:22:35 - mmengine - INFO - Epoch(train) [45][ 800/2569] lr: 5.5429e-06 eta: 7:04:16 time: 0.9322 data_time: 0.0081 memory: 13405 grad_norm: 25.5234 loss: 1.2484 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.2484 2023/06/04 03:24:07 - mmengine - INFO - Epoch(train) [45][ 900/2569] lr: 5.5391e-06 eta: 7:02:39 time: 0.9074 data_time: 0.0079 memory: 13405 grad_norm: 26.5912 loss: 1.1139 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1139 2023/06/04 03:25:06 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 03:25:39 - mmengine - INFO - Epoch(train) [45][1000/2569] lr: 5.5354e-06 eta: 7:01:03 time: 0.9120 data_time: 0.0080 memory: 13405 grad_norm: 26.6492 loss: 1.2028 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.2028 2023/06/04 03:27:12 - mmengine - INFO - Epoch(train) [45][1100/2569] lr: 5.5316e-06 eta: 6:59:31 time: 0.9529 data_time: 0.0080 memory: 13405 grad_norm: 27.5675 loss: 1.2112 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2112 2023/06/04 03:28:45 - mmengine - INFO - Epoch(train) [45][1200/2569] lr: 5.5278e-06 eta: 6:58:00 time: 0.9305 data_time: 0.0080 memory: 13405 grad_norm: 27.1269 loss: 1.2438 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2438 2023/06/04 03:30:19 - mmengine - INFO - Epoch(train) [45][1300/2569] lr: 5.5241e-06 eta: 6:56:33 time: 0.9500 data_time: 0.0080 memory: 13405 grad_norm: 26.6827 loss: 1.2296 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2296 2023/06/04 03:31:52 - mmengine - INFO - Epoch(train) [45][1400/2569] lr: 5.5203e-06 eta: 6:55:03 time: 0.9226 data_time: 0.0079 memory: 13405 grad_norm: 26.0463 loss: 1.2311 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.2311 2023/06/04 03:33:24 - mmengine - INFO - Epoch(train) [45][1500/2569] lr: 5.5166e-06 eta: 6:53:27 time: 0.9043 data_time: 0.0079 memory: 13405 grad_norm: 26.7640 loss: 1.2063 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2063 2023/06/04 03:34:56 - mmengine - INFO - Epoch(train) [45][1600/2569] lr: 5.5129e-06 eta: 6:51:52 time: 0.9201 data_time: 0.0080 memory: 13405 grad_norm: 25.8430 loss: 0.8600 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8600 2023/06/04 03:36:29 - mmengine - INFO - Epoch(train) [45][1700/2569] lr: 5.5092e-06 eta: 6:50:19 time: 0.9311 data_time: 0.0078 memory: 13405 grad_norm: 26.7277 loss: 1.1317 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1317 2023/06/04 03:38:01 - mmengine - INFO - Epoch(train) [45][1800/2569] lr: 5.5055e-06 eta: 6:48:43 time: 0.9337 data_time: 0.0084 memory: 13405 grad_norm: 26.2936 loss: 1.2372 top1_acc: 0.3750 top5_acc: 0.8750 loss_cls: 1.2372 2023/06/04 03:39:34 - mmengine - INFO - Epoch(train) [45][1900/2569] lr: 5.5018e-06 eta: 6:47:12 time: 0.9193 data_time: 0.0081 memory: 13405 grad_norm: 26.0936 loss: 1.2732 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2732 2023/06/04 03:40:32 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 03:41:06 - mmengine - INFO - Epoch(train) [45][2000/2569] lr: 5.4981e-06 eta: 6:45:37 time: 0.9561 data_time: 0.0081 memory: 13405 grad_norm: 26.3906 loss: 1.3336 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.3336 2023/06/04 03:42:39 - mmengine - INFO - Epoch(train) [45][2100/2569] lr: 5.4945e-06 eta: 6:44:06 time: 0.9202 data_time: 0.0079 memory: 13405 grad_norm: 26.1267 loss: 1.1639 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1639 2023/06/04 03:44:12 - mmengine - INFO - Epoch(train) [45][2200/2569] lr: 5.4908e-06 eta: 6:42:32 time: 0.9116 data_time: 0.0085 memory: 13405 grad_norm: 27.9307 loss: 1.2758 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2758 2023/06/04 03:45:44 - mmengine - INFO - Epoch(train) [45][2300/2569] lr: 5.4872e-06 eta: 6:40:57 time: 0.9177 data_time: 0.0077 memory: 13405 grad_norm: 25.6355 loss: 1.5371 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.5371 2023/06/04 03:47:16 - mmengine - INFO - Epoch(train) [45][2400/2569] lr: 5.4836e-06 eta: 6:39:24 time: 0.9258 data_time: 0.0086 memory: 13405 grad_norm: 26.3572 loss: 1.3089 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.3089 2023/06/04 03:48:48 - mmengine - INFO - Epoch(train) [45][2500/2569] lr: 5.4800e-06 eta: 6:37:48 time: 0.9175 data_time: 0.0081 memory: 13405 grad_norm: 26.2703 loss: 0.8637 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8637 2023/06/04 03:49:52 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 03:49:52 - mmengine - INFO - Epoch(train) [45][2569/2569] lr: 5.4775e-06 eta: 6:36:42 time: 0.8814 data_time: 0.0080 memory: 13405 grad_norm: 25.9467 loss: 1.0680 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0680 2023/06/04 03:49:52 - mmengine - INFO - Saving checkpoint at 45 epochs 2023/06/04 03:50:22 - mmengine - INFO - Epoch(val) [45][100/260] eta: 0:00:39 time: 0.2267 data_time: 0.0329 memory: 2900 2023/06/04 03:50:45 - mmengine - INFO - Epoch(val) [45][200/260] eta: 0:00:14 time: 0.2127 data_time: 0.0179 memory: 2900 2023/06/04 03:51:04 - mmengine - INFO - Epoch(val) [45][260/260] acc/top1: 0.7702 acc/top5: 0.9328 acc/mean1: 0.7661 data_time: 0.0388 time: 0.2318 2023/06/04 03:52:40 - mmengine - INFO - Epoch(train) [46][ 100/2569] lr: 5.4739e-06 eta: 6:35:19 time: 0.9391 data_time: 0.0080 memory: 13405 grad_norm: 26.3447 loss: 1.1865 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1865 2023/06/04 03:54:14 - mmengine - INFO - Epoch(train) [46][ 200/2569] lr: 5.4703e-06 eta: 6:33:51 time: 0.9672 data_time: 0.0083 memory: 13405 grad_norm: 26.9076 loss: 0.9684 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9684 2023/06/04 03:55:47 - mmengine - INFO - Epoch(train) [46][ 300/2569] lr: 5.4668e-06 eta: 6:32:18 time: 0.9190 data_time: 0.0078 memory: 13405 grad_norm: 25.6030 loss: 1.0354 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0354 2023/06/04 03:57:14 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 03:57:19 - mmengine - INFO - Epoch(train) [46][ 400/2569] lr: 5.4632e-06 eta: 6:30:43 time: 0.9444 data_time: 0.0082 memory: 13405 grad_norm: 25.7617 loss: 0.9653 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9653 2023/06/04 03:58:53 - mmengine - INFO - Epoch(train) [46][ 500/2569] lr: 5.4597e-06 eta: 6:29:14 time: 0.9048 data_time: 0.0084 memory: 13405 grad_norm: 26.8027 loss: 0.9305 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9305 2023/06/04 04:00:26 - mmengine - INFO - Epoch(train) [46][ 600/2569] lr: 5.4561e-06 eta: 6:27:43 time: 0.9586 data_time: 0.0080 memory: 13405 grad_norm: 26.3014 loss: 0.9452 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9452 2023/06/04 04:01:57 - mmengine - INFO - Epoch(train) [46][ 700/2569] lr: 5.4526e-06 eta: 6:26:05 time: 0.9210 data_time: 0.0080 memory: 13405 grad_norm: 27.1551 loss: 0.9813 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9813 2023/06/04 04:03:29 - mmengine - INFO - Epoch(train) [46][ 800/2569] lr: 5.4491e-06 eta: 6:24:31 time: 0.9368 data_time: 0.0082 memory: 13405 grad_norm: 26.5056 loss: 1.1575 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1575 2023/06/04 04:05:02 - mmengine - INFO - Epoch(train) [46][ 900/2569] lr: 5.4456e-06 eta: 6:23:00 time: 0.9519 data_time: 0.0082 memory: 13405 grad_norm: 26.5333 loss: 0.9444 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.9444 2023/06/04 04:06:35 - mmengine - INFO - Epoch(train) [46][1000/2569] lr: 5.4422e-06 eta: 6:21:27 time: 0.9206 data_time: 0.0083 memory: 13405 grad_norm: 26.3065 loss: 1.0115 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0115 2023/06/04 04:08:07 - mmengine - INFO - Epoch(train) [46][1100/2569] lr: 5.4387e-06 eta: 6:19:53 time: 0.9180 data_time: 0.0082 memory: 13405 grad_norm: 26.8044 loss: 1.1300 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1300 2023/06/04 04:09:39 - mmengine - INFO - Epoch(train) [46][1200/2569] lr: 5.4352e-06 eta: 6:18:18 time: 0.9118 data_time: 0.0080 memory: 13405 grad_norm: 25.7780 loss: 1.0061 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0061 2023/06/04 04:11:11 - mmengine - INFO - Epoch(train) [46][1300/2569] lr: 5.4318e-06 eta: 6:16:42 time: 0.9116 data_time: 0.0081 memory: 13405 grad_norm: 26.8512 loss: 1.0190 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0190 2023/06/04 04:12:38 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 04:12:43 - mmengine - INFO - Epoch(train) [46][1400/2569] lr: 5.4284e-06 eta: 6:15:08 time: 0.9153 data_time: 0.0081 memory: 13405 grad_norm: 25.5962 loss: 1.0654 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0654 2023/06/04 04:14:15 - mmengine - INFO - Epoch(train) [46][1500/2569] lr: 5.4250e-06 eta: 6:13:33 time: 0.9235 data_time: 0.0077 memory: 13405 grad_norm: 25.3668 loss: 1.1830 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1830 2023/06/04 04:15:47 - mmengine - INFO - Epoch(train) [46][1600/2569] lr: 5.4216e-06 eta: 6:11:59 time: 0.9190 data_time: 0.0079 memory: 13405 grad_norm: 25.4196 loss: 1.1333 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1333 2023/06/04 04:17:18 - mmengine - INFO - Epoch(train) [46][1700/2569] lr: 5.4182e-06 eta: 6:10:23 time: 0.9111 data_time: 0.0080 memory: 13405 grad_norm: 26.4347 loss: 1.1812 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1812 2023/06/04 04:18:50 - mmengine - INFO - Epoch(train) [46][1800/2569] lr: 5.4148e-06 eta: 6:08:47 time: 0.9142 data_time: 0.0086 memory: 13405 grad_norm: 26.5340 loss: 1.2663 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.2663 2023/06/04 04:20:23 - mmengine - INFO - Epoch(train) [46][1900/2569] lr: 5.4114e-06 eta: 6:07:15 time: 0.9132 data_time: 0.0087 memory: 13405 grad_norm: 25.5441 loss: 1.1571 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1571 2023/06/04 04:21:55 - mmengine - INFO - Epoch(train) [46][2000/2569] lr: 5.4081e-06 eta: 6:05:41 time: 0.9339 data_time: 0.0080 memory: 13405 grad_norm: 25.5504 loss: 0.9459 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9459 2023/06/04 04:23:29 - mmengine - INFO - Epoch(train) [46][2100/2569] lr: 5.4047e-06 eta: 6:04:12 time: 0.9456 data_time: 0.0079 memory: 13405 grad_norm: 27.9010 loss: 1.1672 top1_acc: 0.6250 top5_acc: 0.6250 loss_cls: 1.1672 2023/06/04 04:25:02 - mmengine - INFO - Epoch(train) [46][2200/2569] lr: 5.4014e-06 eta: 6:02:42 time: 0.9181 data_time: 0.0083 memory: 13405 grad_norm: 26.4344 loss: 1.2078 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2078 2023/06/04 04:26:35 - mmengine - INFO - Epoch(train) [46][2300/2569] lr: 5.3981e-06 eta: 6:01:08 time: 0.9274 data_time: 0.0083 memory: 13405 grad_norm: 25.7640 loss: 1.0696 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0696 2023/06/04 04:28:03 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 04:28:08 - mmengine - INFO - Epoch(train) [46][2400/2569] lr: 5.3948e-06 eta: 5:59:38 time: 0.9510 data_time: 0.0083 memory: 13405 grad_norm: 26.3318 loss: 1.0284 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0284 2023/06/04 04:29:41 - mmengine - INFO - Epoch(train) [46][2500/2569] lr: 5.3915e-06 eta: 5:58:05 time: 0.9179 data_time: 0.0083 memory: 13405 grad_norm: 26.5961 loss: 1.2851 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2851 2023/06/04 04:30:44 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 04:30:44 - mmengine - INFO - Epoch(train) [46][2569/2569] lr: 5.3892e-06 eta: 5:56:58 time: 0.8773 data_time: 0.0082 memory: 13405 grad_norm: 26.4906 loss: 0.8767 top1_acc: 0.6667 top5_acc: 0.8333 loss_cls: 0.8767 2023/06/04 04:31:09 - mmengine - INFO - Epoch(val) [46][100/260] eta: 0:00:40 time: 0.2339 data_time: 0.0401 memory: 2900 2023/06/04 04:31:31 - mmengine - INFO - Epoch(val) [46][200/260] eta: 0:00:14 time: 0.2279 data_time: 0.0341 memory: 2900 2023/06/04 04:31:54 - mmengine - INFO - Epoch(val) [46][260/260] acc/top1: 0.7711 acc/top5: 0.9319 acc/mean1: 0.7675 data_time: 0.0434 time: 0.2367 2023/06/04 04:31:57 - mmengine - INFO - The best checkpoint with 0.7711 acc/top1 at 46 epoch is saved to best_acc_top1_epoch_46.pth. 2023/06/04 04:33:31 - mmengine - INFO - Epoch(train) [47][ 100/2569] lr: 5.3859e-06 eta: 5:55:29 time: 0.9142 data_time: 0.0081 memory: 13405 grad_norm: 25.6322 loss: 1.1416 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1416 2023/06/04 04:35:06 - mmengine - INFO - Epoch(train) [47][ 200/2569] lr: 5.3827e-06 eta: 5:54:01 time: 0.9379 data_time: 0.0081 memory: 13405 grad_norm: 26.4162 loss: 0.9262 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.9262 2023/06/04 04:36:39 - mmengine - INFO - Epoch(train) [47][ 300/2569] lr: 5.3794e-06 eta: 5:52:30 time: 0.9454 data_time: 0.0080 memory: 13405 grad_norm: 25.3765 loss: 1.0022 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0022 2023/06/04 04:38:11 - mmengine - INFO - Epoch(train) [47][ 400/2569] lr: 5.3762e-06 eta: 5:50:55 time: 0.9123 data_time: 0.0084 memory: 13405 grad_norm: 26.5664 loss: 1.1169 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.1169 2023/06/04 04:39:42 - mmengine - INFO - Epoch(train) [47][ 500/2569] lr: 5.3730e-06 eta: 5:49:20 time: 0.9111 data_time: 0.0081 memory: 13405 grad_norm: 26.0318 loss: 1.3747 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.3747 2023/06/04 04:41:16 - mmengine - INFO - Epoch(train) [47][ 600/2569] lr: 5.3698e-06 eta: 5:47:48 time: 0.9233 data_time: 0.0082 memory: 13405 grad_norm: 26.0533 loss: 1.0587 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0587 2023/06/04 04:42:47 - mmengine - INFO - Epoch(train) [47][ 700/2569] lr: 5.3666e-06 eta: 5:46:13 time: 0.9127 data_time: 0.0081 memory: 13405 grad_norm: 26.0309 loss: 1.0395 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0395 2023/06/04 04:44:20 - mmengine - INFO - Epoch(train) [47][ 800/2569] lr: 5.3634e-06 eta: 5:44:41 time: 0.9586 data_time: 0.0086 memory: 13405 grad_norm: 25.3879 loss: 1.1222 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1222 2023/06/04 04:44:44 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 04:45:51 - mmengine - INFO - Epoch(train) [47][ 900/2569] lr: 5.3602e-06 eta: 5:43:05 time: 0.9133 data_time: 0.0082 memory: 13405 grad_norm: 26.4811 loss: 1.4611 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.4611 2023/06/04 04:47:23 - mmengine - INFO - Epoch(train) [47][1000/2569] lr: 5.3571e-06 eta: 5:41:31 time: 0.9140 data_time: 0.0089 memory: 13405 grad_norm: 26.7735 loss: 1.0301 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0301 2023/06/04 04:48:56 - mmengine - INFO - Epoch(train) [47][1100/2569] lr: 5.3539e-06 eta: 5:39:59 time: 0.9270 data_time: 0.0083 memory: 13405 grad_norm: 26.8671 loss: 1.1167 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1167 2023/06/04 04:50:30 - mmengine - INFO - Epoch(train) [47][1200/2569] lr: 5.3508e-06 eta: 5:38:29 time: 0.9276 data_time: 0.0080 memory: 13405 grad_norm: 25.5707 loss: 0.9670 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.9670 2023/06/04 04:52:02 - mmengine - INFO - Epoch(train) [47][1300/2569] lr: 5.3477e-06 eta: 5:36:56 time: 0.9153 data_time: 0.0080 memory: 13405 grad_norm: 26.2192 loss: 1.3044 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3044 2023/06/04 04:53:34 - mmengine - INFO - Epoch(train) [47][1400/2569] lr: 5.3446e-06 eta: 5:35:21 time: 0.9164 data_time: 0.0079 memory: 13405 grad_norm: 26.2155 loss: 1.2762 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2762 2023/06/04 04:55:08 - mmengine - INFO - Epoch(train) [47][1500/2569] lr: 5.3415e-06 eta: 5:33:50 time: 0.9134 data_time: 0.0080 memory: 13405 grad_norm: 26.1808 loss: 1.0251 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0251 2023/06/04 04:56:40 - mmengine - INFO - Epoch(train) [47][1600/2569] lr: 5.3384e-06 eta: 5:32:16 time: 0.9381 data_time: 0.0083 memory: 13405 grad_norm: 27.0275 loss: 1.1434 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1434 2023/06/04 04:58:13 - mmengine - INFO - Epoch(train) [47][1700/2569] lr: 5.3354e-06 eta: 5:30:46 time: 0.9511 data_time: 0.0081 memory: 13405 grad_norm: 26.7951 loss: 1.2308 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2308 2023/06/04 04:59:47 - mmengine - INFO - Epoch(train) [47][1800/2569] lr: 5.3323e-06 eta: 5:29:15 time: 0.9162 data_time: 0.0082 memory: 13405 grad_norm: 26.6417 loss: 0.9367 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9367 2023/06/04 05:00:11 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 05:01:19 - mmengine - INFO - Epoch(train) [47][1900/2569] lr: 5.3293e-06 eta: 5:27:40 time: 0.9339 data_time: 0.0082 memory: 13405 grad_norm: 26.9955 loss: 1.3092 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.3092 2023/06/04 05:02:53 - mmengine - INFO - Epoch(train) [47][2000/2569] lr: 5.3262e-06 eta: 5:26:10 time: 0.9314 data_time: 0.0081 memory: 13405 grad_norm: 26.3970 loss: 0.9256 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9256 2023/06/04 05:04:25 - mmengine - INFO - Epoch(train) [47][2100/2569] lr: 5.3232e-06 eta: 5:24:37 time: 0.9370 data_time: 0.0083 memory: 13405 grad_norm: 26.3222 loss: 1.0904 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0904 2023/06/04 05:05:59 - mmengine - INFO - Epoch(train) [47][2200/2569] lr: 5.3202e-06 eta: 5:23:06 time: 0.9388 data_time: 0.0081 memory: 13405 grad_norm: 25.8301 loss: 0.9435 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9435 2023/06/04 05:07:32 - mmengine - INFO - Epoch(train) [47][2300/2569] lr: 5.3172e-06 eta: 5:21:34 time: 0.9319 data_time: 0.0081 memory: 13405 grad_norm: 27.4028 loss: 1.1960 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1960 2023/06/04 05:09:05 - mmengine - INFO - Epoch(train) [47][2400/2569] lr: 5.3143e-06 eta: 5:20:01 time: 0.9310 data_time: 0.0087 memory: 13405 grad_norm: 26.2760 loss: 1.2744 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2744 2023/06/04 05:10:36 - mmengine - INFO - Epoch(train) [47][2500/2569] lr: 5.3113e-06 eta: 5:18:27 time: 0.9198 data_time: 0.0080 memory: 13405 grad_norm: 26.5263 loss: 1.2571 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.2571 2023/06/04 05:11:39 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 05:11:39 - mmengine - INFO - Epoch(train) [47][2569/2569] lr: 5.3093e-06 eta: 5:17:21 time: 0.8816 data_time: 0.0083 memory: 13405 grad_norm: 25.5284 loss: 0.8515 top1_acc: 0.8333 top5_acc: 0.8333 loss_cls: 0.8515 2023/06/04 05:12:04 - mmengine - INFO - Epoch(val) [47][100/260] eta: 0:00:39 time: 0.2447 data_time: 0.0502 memory: 2900 2023/06/04 05:12:26 - mmengine - INFO - Epoch(val) [47][200/260] eta: 0:00:14 time: 0.2030 data_time: 0.0097 memory: 2900 2023/06/04 05:12:49 - mmengine - INFO - Epoch(val) [47][260/260] acc/top1: 0.7704 acc/top5: 0.9323 acc/mean1: 0.7658 data_time: 0.0378 time: 0.2312 2023/06/04 05:14:24 - mmengine - INFO - Epoch(train) [48][ 100/2569] lr: 5.3063e-06 eta: 5:15:52 time: 0.9101 data_time: 0.0080 memory: 13405 grad_norm: 26.0164 loss: 1.2613 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2613 2023/06/04 05:15:56 - mmengine - INFO - Epoch(train) [48][ 200/2569] lr: 5.3034e-06 eta: 5:14:19 time: 0.9140 data_time: 0.0079 memory: 13405 grad_norm: 25.8882 loss: 0.8499 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8499 2023/06/04 05:16:50 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 05:17:30 - mmengine - INFO - Epoch(train) [48][ 300/2569] lr: 5.3005e-06 eta: 5:12:48 time: 0.9229 data_time: 0.0085 memory: 13405 grad_norm: 25.7251 loss: 0.9508 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 0.9508 2023/06/04 05:19:03 - mmengine - INFO - Epoch(train) [48][ 400/2569] lr: 5.2976e-06 eta: 5:11:16 time: 0.9019 data_time: 0.0086 memory: 13405 grad_norm: 26.9722 loss: 1.0094 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0094 2023/06/04 05:20:35 - mmengine - INFO - Epoch(train) [48][ 500/2569] lr: 5.2947e-06 eta: 5:09:42 time: 0.9523 data_time: 0.0080 memory: 13405 grad_norm: 26.0011 loss: 0.8503 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8503 2023/06/04 05:22:09 - mmengine - INFO - Epoch(train) [48][ 600/2569] lr: 5.2918e-06 eta: 5:08:11 time: 0.9445 data_time: 0.0080 memory: 13405 grad_norm: 26.2349 loss: 1.2858 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2858 2023/06/04 05:23:42 - mmengine - INFO - Epoch(train) [48][ 700/2569] lr: 5.2890e-06 eta: 5:06:39 time: 0.9178 data_time: 0.0084 memory: 13405 grad_norm: 25.7117 loss: 0.9280 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9280 2023/06/04 05:25:15 - mmengine - INFO - Epoch(train) [48][ 800/2569] lr: 5.2861e-06 eta: 5:05:06 time: 0.9188 data_time: 0.0084 memory: 13405 grad_norm: 26.9223 loss: 1.3131 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3131 2023/06/04 05:26:48 - mmengine - INFO - Epoch(train) [48][ 900/2569] lr: 5.2833e-06 eta: 5:03:35 time: 0.9631 data_time: 0.0082 memory: 13405 grad_norm: 25.8153 loss: 0.9696 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9696 2023/06/04 05:28:21 - mmengine - INFO - Epoch(train) [48][1000/2569] lr: 5.2805e-06 eta: 5:02:02 time: 0.9181 data_time: 0.0083 memory: 13405 grad_norm: 27.4443 loss: 1.1055 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1055 2023/06/04 05:29:54 - mmengine - INFO - Epoch(train) [48][1100/2569] lr: 5.2777e-06 eta: 5:00:30 time: 0.9208 data_time: 0.0081 memory: 13405 grad_norm: 28.0908 loss: 1.3172 top1_acc: 0.3750 top5_acc: 0.3750 loss_cls: 1.3172 2023/06/04 05:31:26 - mmengine - INFO - Epoch(train) [48][1200/2569] lr: 5.2749e-06 eta: 4:58:56 time: 0.9117 data_time: 0.0082 memory: 13405 grad_norm: 26.0023 loss: 1.1084 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1084 2023/06/04 05:32:19 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 05:32:59 - mmengine - INFO - Epoch(train) [48][1300/2569] lr: 5.2721e-06 eta: 4:57:23 time: 0.9195 data_time: 0.0081 memory: 13405 grad_norm: 27.1576 loss: 1.1654 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1654 2023/06/04 05:34:32 - mmengine - INFO - Epoch(train) [48][1400/2569] lr: 5.2693e-06 eta: 4:55:51 time: 0.9391 data_time: 0.0080 memory: 13405 grad_norm: 27.1488 loss: 0.9747 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9747 2023/06/04 05:36:05 - mmengine - INFO - Epoch(train) [48][1500/2569] lr: 5.2666e-06 eta: 4:54:19 time: 0.9376 data_time: 0.0081 memory: 13405 grad_norm: 27.2214 loss: 1.3353 top1_acc: 0.2500 top5_acc: 0.6250 loss_cls: 1.3353 2023/06/04 05:37:38 - mmengine - INFO - Epoch(train) [48][1600/2569] lr: 5.2638e-06 eta: 4:52:46 time: 0.9188 data_time: 0.0084 memory: 13405 grad_norm: 26.4688 loss: 1.2259 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.2259 2023/06/04 05:39:11 - mmengine - INFO - Epoch(train) [48][1700/2569] lr: 5.2611e-06 eta: 4:51:14 time: 0.9367 data_time: 0.0080 memory: 13405 grad_norm: 26.1857 loss: 1.1505 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.1505 2023/06/04 05:40:44 - mmengine - INFO - Epoch(train) [48][1800/2569] lr: 5.2584e-06 eta: 4:49:43 time: 0.9407 data_time: 0.0090 memory: 13405 grad_norm: 26.2443 loss: 0.8709 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8709 2023/06/04 05:42:18 - mmengine - INFO - Epoch(train) [48][1900/2569] lr: 5.2557e-06 eta: 4:48:11 time: 0.9525 data_time: 0.0082 memory: 13405 grad_norm: 26.6931 loss: 0.7650 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.7650 2023/06/04 05:43:52 - mmengine - INFO - Epoch(train) [48][2000/2569] lr: 5.2530e-06 eta: 4:46:40 time: 0.9444 data_time: 0.0086 memory: 13405 grad_norm: 26.5477 loss: 1.1000 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1000 2023/06/04 05:45:25 - mmengine - INFO - Epoch(train) [48][2100/2569] lr: 5.2503e-06 eta: 4:45:08 time: 0.9291 data_time: 0.0079 memory: 13405 grad_norm: 27.3980 loss: 1.1542 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1542 2023/06/04 05:46:57 - mmengine - INFO - Epoch(train) [48][2200/2569] lr: 5.2477e-06 eta: 4:43:34 time: 0.9207 data_time: 0.0081 memory: 13405 grad_norm: 26.3493 loss: 0.9331 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9331 2023/06/04 05:47:50 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 05:48:29 - mmengine - INFO - Epoch(train) [48][2300/2569] lr: 5.2450e-06 eta: 4:42:01 time: 0.9177 data_time: 0.0080 memory: 13405 grad_norm: 26.0311 loss: 1.0127 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0127 2023/06/04 05:50:02 - mmengine - INFO - Epoch(train) [48][2400/2569] lr: 5.2424e-06 eta: 4:40:28 time: 0.9317 data_time: 0.0084 memory: 13405 grad_norm: 25.5332 loss: 1.0269 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0269 2023/06/04 05:51:34 - mmengine - INFO - Epoch(train) [48][2500/2569] lr: 5.2398e-06 eta: 4:38:54 time: 0.9112 data_time: 0.0082 memory: 13405 grad_norm: 26.1665 loss: 1.3728 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.3728 2023/06/04 05:52:37 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 05:52:37 - mmengine - INFO - Epoch(train) [48][2569/2569] lr: 5.2380e-06 eta: 4:37:49 time: 0.8839 data_time: 0.0083 memory: 13405 grad_norm: 26.3673 loss: 1.2047 top1_acc: 0.8333 top5_acc: 0.8333 loss_cls: 1.2047 2023/06/04 05:52:37 - mmengine - INFO - Saving checkpoint at 48 epochs 2023/06/04 05:53:08 - mmengine - INFO - Epoch(val) [48][100/260] eta: 0:00:39 time: 0.2335 data_time: 0.0393 memory: 2900 2023/06/04 05:53:30 - mmengine - INFO - Epoch(val) [48][200/260] eta: 0:00:14 time: 0.2175 data_time: 0.0237 memory: 2900 2023/06/04 05:53:49 - mmengine - INFO - Epoch(val) [48][260/260] acc/top1: 0.7702 acc/top5: 0.9324 acc/mean1: 0.7661 data_time: 0.0335 time: 0.2266 2023/06/04 05:55:24 - mmengine - INFO - Epoch(train) [49][ 100/2569] lr: 5.2354e-06 eta: 4:36:20 time: 0.9252 data_time: 0.0090 memory: 13405 grad_norm: 25.7682 loss: 1.1547 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.1547 2023/06/04 05:56:58 - mmengine - INFO - Epoch(train) [49][ 200/2569] lr: 5.2328e-06 eta: 4:34:48 time: 0.9506 data_time: 0.0081 memory: 13405 grad_norm: 27.9378 loss: 0.8822 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8822 2023/06/04 05:58:31 - mmengine - INFO - Epoch(train) [49][ 300/2569] lr: 5.2302e-06 eta: 4:33:16 time: 0.9365 data_time: 0.0082 memory: 13405 grad_norm: 25.6293 loss: 1.0424 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0424 2023/06/04 06:00:05 - mmengine - INFO - Epoch(train) [49][ 400/2569] lr: 5.2277e-06 eta: 4:31:44 time: 0.9269 data_time: 0.0085 memory: 13405 grad_norm: 27.8031 loss: 0.9981 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9981 2023/06/04 06:01:37 - mmengine - INFO - Epoch(train) [49][ 500/2569] lr: 5.2251e-06 eta: 4:30:11 time: 0.9292 data_time: 0.0079 memory: 13405 grad_norm: 26.3817 loss: 1.4230 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.4230 2023/06/04 06:03:11 - mmengine - INFO - Epoch(train) [49][ 600/2569] lr: 5.2226e-06 eta: 4:28:40 time: 0.9198 data_time: 0.0085 memory: 13405 grad_norm: 25.8894 loss: 1.1341 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1341 2023/06/04 06:04:33 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 06:04:44 - mmengine - INFO - Epoch(train) [49][ 700/2569] lr: 5.2201e-06 eta: 4:27:07 time: 0.9238 data_time: 0.0077 memory: 13405 grad_norm: 26.4380 loss: 1.0892 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0892 2023/06/04 06:06:16 - mmengine - INFO - Epoch(train) [49][ 800/2569] lr: 5.2176e-06 eta: 4:25:33 time: 0.9244 data_time: 0.0081 memory: 13405 grad_norm: 25.7624 loss: 0.8352 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8352 2023/06/04 06:07:48 - mmengine - INFO - Epoch(train) [49][ 900/2569] lr: 5.2151e-06 eta: 4:24:01 time: 0.9077 data_time: 0.0082 memory: 13405 grad_norm: 25.6679 loss: 0.9996 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9996 2023/06/04 06:09:21 - mmengine - INFO - Epoch(train) [49][1000/2569] lr: 5.2126e-06 eta: 4:22:28 time: 0.9360 data_time: 0.0083 memory: 13405 grad_norm: 26.9161 loss: 1.0512 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0512 2023/06/04 06:10:54 - mmengine - INFO - Epoch(train) [49][1100/2569] lr: 5.2101e-06 eta: 4:20:55 time: 0.9491 data_time: 0.0086 memory: 13405 grad_norm: 26.2872 loss: 0.9678 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9678 2023/06/04 06:12:27 - mmengine - INFO - Epoch(train) [49][1200/2569] lr: 5.2077e-06 eta: 4:19:23 time: 0.9331 data_time: 0.0084 memory: 13405 grad_norm: 27.5649 loss: 1.2708 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2708 2023/06/04 06:13:59 - mmengine - INFO - Epoch(train) [49][1300/2569] lr: 5.2053e-06 eta: 4:17:50 time: 0.9181 data_time: 0.0084 memory: 13405 grad_norm: 26.2048 loss: 0.8678 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8678 2023/06/04 06:15:33 - mmengine - INFO - Epoch(train) [49][1400/2569] lr: 5.2028e-06 eta: 4:16:18 time: 0.9429 data_time: 0.0080 memory: 13405 grad_norm: 26.2215 loss: 1.3016 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.3016 2023/06/04 06:17:07 - mmengine - INFO - Epoch(train) [49][1500/2569] lr: 5.2004e-06 eta: 4:14:46 time: 0.9553 data_time: 0.0085 memory: 13405 grad_norm: 25.8565 loss: 1.1028 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1028 2023/06/04 06:18:40 - mmengine - INFO - Epoch(train) [49][1600/2569] lr: 5.1980e-06 eta: 4:13:13 time: 0.9236 data_time: 0.0080 memory: 13405 grad_norm: 26.5641 loss: 0.9450 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 0.9450 2023/06/04 06:20:01 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 06:20:12 - mmengine - INFO - Epoch(train) [49][1700/2569] lr: 5.1957e-06 eta: 4:11:40 time: 0.9320 data_time: 0.0087 memory: 13405 grad_norm: 25.5298 loss: 1.0481 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.0481 2023/06/04 06:21:43 - mmengine - INFO - Epoch(train) [49][1800/2569] lr: 5.1933e-06 eta: 4:10:06 time: 0.9043 data_time: 0.0085 memory: 13405 grad_norm: 26.8252 loss: 1.0854 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0854 2023/06/04 06:23:16 - mmengine - INFO - Epoch(train) [49][1900/2569] lr: 5.1910e-06 eta: 4:08:33 time: 0.9122 data_time: 0.0083 memory: 13405 grad_norm: 26.2279 loss: 1.0911 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0911 2023/06/04 06:24:48 - mmengine - INFO - Epoch(train) [49][2000/2569] lr: 5.1886e-06 eta: 4:07:00 time: 0.9234 data_time: 0.0083 memory: 13405 grad_norm: 25.8467 loss: 0.9749 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9749 2023/06/04 06:26:20 - mmengine - INFO - Epoch(train) [49][2100/2569] lr: 5.1863e-06 eta: 4:05:27 time: 0.9285 data_time: 0.0081 memory: 13405 grad_norm: 26.5030 loss: 1.1058 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1058 2023/06/04 06:27:52 - mmengine - INFO - Epoch(train) [49][2200/2569] lr: 5.1840e-06 eta: 4:03:53 time: 0.9155 data_time: 0.0084 memory: 13405 grad_norm: 26.1833 loss: 1.2783 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.2783 2023/06/04 06:29:25 - mmengine - INFO - Epoch(train) [49][2300/2569] lr: 5.1817e-06 eta: 4:02:21 time: 0.9135 data_time: 0.0081 memory: 13405 grad_norm: 25.3066 loss: 0.9601 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9601 2023/06/04 06:30:57 - mmengine - INFO - Epoch(train) [49][2400/2569] lr: 5.1794e-06 eta: 4:00:48 time: 0.9264 data_time: 0.0082 memory: 13405 grad_norm: 25.8007 loss: 0.9662 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9662 2023/06/04 06:32:31 - mmengine - INFO - Epoch(train) [49][2500/2569] lr: 5.1771e-06 eta: 3:59:15 time: 0.9423 data_time: 0.0079 memory: 13405 grad_norm: 26.3153 loss: 1.0508 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0508 2023/06/04 06:33:35 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 06:33:35 - mmengine - INFO - Epoch(train) [49][2569/2569] lr: 5.1756e-06 eta: 3:58:11 time: 0.9009 data_time: 0.0081 memory: 13405 grad_norm: 27.3527 loss: 1.1091 top1_acc: 0.6667 top5_acc: 0.6667 loss_cls: 1.1091 2023/06/04 06:33:59 - mmengine - INFO - Epoch(val) [49][100/260] eta: 0:00:39 time: 0.2427 data_time: 0.0485 memory: 2900 2023/06/04 06:34:21 - mmengine - INFO - Epoch(val) [49][200/260] eta: 0:00:13 time: 0.2049 data_time: 0.0110 memory: 2900 2023/06/04 06:34:43 - mmengine - INFO - Epoch(val) [49][260/260] acc/top1: 0.7704 acc/top5: 0.9321 acc/mean1: 0.7657 data_time: 0.0387 time: 0.2322 2023/06/04 06:36:20 - mmengine - INFO - Epoch(train) [50][ 100/2569] lr: 5.1733e-06 eta: 3:56:42 time: 0.9114 data_time: 0.0080 memory: 13405 grad_norm: 25.8138 loss: 0.9586 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 0.9586 2023/06/04 06:36:38 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 06:37:52 - mmengine - INFO - Epoch(train) [50][ 200/2569] lr: 5.1711e-06 eta: 3:55:09 time: 0.9188 data_time: 0.0082 memory: 13405 grad_norm: 26.8899 loss: 1.2835 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2835 2023/06/04 06:39:24 - mmengine - INFO - Epoch(train) [50][ 300/2569] lr: 5.1689e-06 eta: 3:53:36 time: 0.9116 data_time: 0.0086 memory: 13405 grad_norm: 27.0850 loss: 0.9712 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9712 2023/06/04 06:40:58 - mmengine - INFO - Epoch(train) [50][ 400/2569] lr: 5.1667e-06 eta: 3:52:03 time: 0.9443 data_time: 0.0084 memory: 13405 grad_norm: 26.0925 loss: 1.0824 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0824 2023/06/04 06:42:31 - mmengine - INFO - Epoch(train) [50][ 500/2569] lr: 5.1645e-06 eta: 3:50:31 time: 0.9116 data_time: 0.0081 memory: 13405 grad_norm: 26.2232 loss: 0.9924 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 0.9924 2023/06/04 06:44:03 - mmengine - INFO - Epoch(train) [50][ 600/2569] lr: 5.1623e-06 eta: 3:48:58 time: 0.9173 data_time: 0.0082 memory: 13405 grad_norm: 26.5523 loss: 1.0733 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0733 2023/06/04 06:45:36 - mmengine - INFO - Epoch(train) [50][ 700/2569] lr: 5.1602e-06 eta: 3:47:25 time: 0.9136 data_time: 0.0083 memory: 13405 grad_norm: 26.6243 loss: 1.3453 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.3453 2023/06/04 06:47:08 - mmengine - INFO - Epoch(train) [50][ 800/2569] lr: 5.1580e-06 eta: 3:45:52 time: 0.9452 data_time: 0.0084 memory: 13405 grad_norm: 26.3165 loss: 0.7815 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.7815 2023/06/04 06:48:40 - mmengine - INFO - Epoch(train) [50][ 900/2569] lr: 5.1559e-06 eta: 3:44:19 time: 0.9204 data_time: 0.0084 memory: 13405 grad_norm: 26.9977 loss: 1.0995 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0995 2023/06/04 06:50:13 - mmengine - INFO - Epoch(train) [50][1000/2569] lr: 5.1538e-06 eta: 3:42:46 time: 0.9275 data_time: 0.0084 memory: 13405 grad_norm: 26.2753 loss: 0.9218 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9218 2023/06/04 06:51:47 - mmengine - INFO - Epoch(train) [50][1100/2569] lr: 5.1517e-06 eta: 3:41:14 time: 0.9531 data_time: 0.0080 memory: 13405 grad_norm: 26.1467 loss: 1.4552 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.4552 2023/06/04 06:52:05 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 06:53:21 - mmengine - INFO - Epoch(train) [50][1200/2569] lr: 5.1496e-06 eta: 3:39:42 time: 0.9542 data_time: 0.0085 memory: 13405 grad_norm: 25.5751 loss: 1.0462 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0462 2023/06/04 06:54:53 - mmengine - INFO - Epoch(train) [50][1300/2569] lr: 5.1475e-06 eta: 3:38:09 time: 0.9227 data_time: 0.0079 memory: 13405 grad_norm: 26.2695 loss: 0.9735 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9735 2023/06/04 06:56:26 - mmengine - INFO - Epoch(train) [50][1400/2569] lr: 5.1454e-06 eta: 3:36:36 time: 0.9218 data_time: 0.0078 memory: 13405 grad_norm: 26.5043 loss: 0.9630 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.9630 2023/06/04 06:57:58 - mmengine - INFO - Epoch(train) [50][1500/2569] lr: 5.1434e-06 eta: 3:35:03 time: 0.9142 data_time: 0.0084 memory: 13405 grad_norm: 25.9082 loss: 1.0904 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0904 2023/06/04 06:59:30 - mmengine - INFO - Epoch(train) [50][1600/2569] lr: 5.1414e-06 eta: 3:33:30 time: 0.9274 data_time: 0.0084 memory: 13405 grad_norm: 26.8094 loss: 1.1852 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1852 2023/06/04 07:01:02 - mmengine - INFO - Epoch(train) [50][1700/2569] lr: 5.1393e-06 eta: 3:31:57 time: 0.9220 data_time: 0.0080 memory: 13405 grad_norm: 26.3435 loss: 1.0943 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0943 2023/06/04 07:02:37 - mmengine - INFO - Epoch(train) [50][1800/2569] lr: 5.1373e-06 eta: 3:30:25 time: 0.9490 data_time: 0.0082 memory: 13405 grad_norm: 26.1417 loss: 0.9424 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.9424 2023/06/04 07:04:09 - mmengine - INFO - Epoch(train) [50][1900/2569] lr: 5.1353e-06 eta: 3:28:52 time: 0.9136 data_time: 0.0081 memory: 13405 grad_norm: 26.8818 loss: 0.9287 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9287 2023/06/04 07:05:43 - mmengine - INFO - Epoch(train) [50][2000/2569] lr: 5.1334e-06 eta: 3:27:20 time: 0.9199 data_time: 0.0080 memory: 13405 grad_norm: 25.8381 loss: 1.2191 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.2191 2023/06/04 07:07:15 - mmengine - INFO - Epoch(train) [50][2100/2569] lr: 5.1314e-06 eta: 3:25:47 time: 0.9214 data_time: 0.0084 memory: 13405 grad_norm: 26.2021 loss: 1.0264 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0264 2023/06/04 07:07:33 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 07:08:50 - mmengine - INFO - Epoch(train) [50][2200/2569] lr: 5.1294e-06 eta: 3:24:15 time: 0.9594 data_time: 0.0084 memory: 13405 grad_norm: 26.5540 loss: 0.9507 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9507 2023/06/04 07:10:23 - mmengine - INFO - Epoch(train) [50][2300/2569] lr: 5.1275e-06 eta: 3:22:43 time: 0.9204 data_time: 0.0082 memory: 13405 grad_norm: 26.6565 loss: 1.0843 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0843 2023/06/04 07:11:54 - mmengine - INFO - Epoch(train) [50][2400/2569] lr: 5.1256e-06 eta: 3:21:09 time: 0.9141 data_time: 0.0086 memory: 13405 grad_norm: 26.2691 loss: 1.0119 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0119 2023/06/04 07:13:27 - mmengine - INFO - Epoch(train) [50][2500/2569] lr: 5.1237e-06 eta: 3:19:36 time: 0.9142 data_time: 0.0084 memory: 13405 grad_norm: 27.1942 loss: 1.0128 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.0128 2023/06/04 07:14:30 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 07:14:30 - mmengine - INFO - Epoch(train) [50][2569/2569] lr: 5.1224e-06 eta: 3:18:32 time: 0.9140 data_time: 0.0081 memory: 13405 grad_norm: 26.5644 loss: 1.2208 top1_acc: 0.8333 top5_acc: 1.0000 loss_cls: 1.2208 2023/06/04 07:14:55 - mmengine - INFO - Epoch(val) [50][100/260] eta: 0:00:39 time: 0.2417 data_time: 0.0477 memory: 2900 2023/06/04 07:15:16 - mmengine - INFO - Epoch(val) [50][200/260] eta: 0:00:13 time: 0.2151 data_time: 0.0218 memory: 2900 2023/06/04 07:15:39 - mmengine - INFO - Epoch(val) [50][260/260] acc/top1: 0.7694 acc/top5: 0.9314 acc/mean1: 0.7654 data_time: 0.0352 time: 0.2286 2023/06/04 07:17:14 - mmengine - INFO - Epoch(train) [51][ 100/2569] lr: 5.1205e-06 eta: 3:17:01 time: 0.9180 data_time: 0.0080 memory: 13405 grad_norm: 26.5470 loss: 1.3063 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.3063 2023/06/04 07:18:48 - mmengine - INFO - Epoch(train) [51][ 200/2569] lr: 5.1186e-06 eta: 3:15:29 time: 0.9114 data_time: 0.0083 memory: 13405 grad_norm: 27.1843 loss: 1.0815 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.0815 2023/06/04 07:20:23 - mmengine - INFO - Epoch(train) [51][ 300/2569] lr: 5.1168e-06 eta: 3:13:57 time: 0.9469 data_time: 0.0080 memory: 13405 grad_norm: 26.0620 loss: 1.1109 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1109 2023/06/04 07:21:56 - mmengine - INFO - Epoch(train) [51][ 400/2569] lr: 5.1149e-06 eta: 3:12:25 time: 0.9617 data_time: 0.0080 memory: 13405 grad_norm: 25.1923 loss: 0.8814 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.8814 2023/06/04 07:23:29 - mmengine - INFO - Epoch(train) [51][ 500/2569] lr: 5.1131e-06 eta: 3:10:52 time: 0.9486 data_time: 0.0083 memory: 13405 grad_norm: 26.3774 loss: 1.2265 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2265 2023/06/04 07:24:15 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 07:25:01 - mmengine - INFO - Epoch(train) [51][ 600/2569] lr: 5.1113e-06 eta: 3:09:19 time: 0.9238 data_time: 0.0084 memory: 13405 grad_norm: 26.3027 loss: 1.0492 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0492 2023/06/04 07:26:34 - mmengine - INFO - Epoch(train) [51][ 700/2569] lr: 5.1095e-06 eta: 3:07:46 time: 0.9422 data_time: 0.0081 memory: 13405 grad_norm: 25.9759 loss: 0.8305 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8305 2023/06/04 07:28:06 - mmengine - INFO - Epoch(train) [51][ 800/2569] lr: 5.1077e-06 eta: 3:06:13 time: 0.9131 data_time: 0.0078 memory: 13405 grad_norm: 26.2976 loss: 0.9694 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9694 2023/06/04 07:29:38 - mmengine - INFO - Epoch(train) [51][ 900/2569] lr: 5.1059e-06 eta: 3:04:40 time: 0.9145 data_time: 0.0079 memory: 13405 grad_norm: 26.9905 loss: 0.9503 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9503 2023/06/04 07:31:10 - mmengine - INFO - Epoch(train) [51][1000/2569] lr: 5.1042e-06 eta: 3:03:07 time: 0.9204 data_time: 0.0087 memory: 13405 grad_norm: 26.4360 loss: 0.9358 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 0.9358 2023/06/04 07:32:43 - mmengine - INFO - Epoch(train) [51][1100/2569] lr: 5.1025e-06 eta: 3:01:34 time: 0.9241 data_time: 0.0081 memory: 13405 grad_norm: 27.7239 loss: 1.2474 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2474 2023/06/04 07:34:18 - mmengine - INFO - Epoch(train) [51][1200/2569] lr: 5.1007e-06 eta: 3:00:02 time: 0.9181 data_time: 0.0088 memory: 13405 grad_norm: 26.6337 loss: 1.0216 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0216 2023/06/04 07:35:51 - mmengine - INFO - Epoch(train) [51][1300/2569] lr: 5.0990e-06 eta: 2:58:30 time: 0.9402 data_time: 0.0083 memory: 13405 grad_norm: 26.5398 loss: 1.0572 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0572 2023/06/04 07:37:25 - mmengine - INFO - Epoch(train) [51][1400/2569] lr: 5.0973e-06 eta: 2:56:57 time: 0.9195 data_time: 0.0084 memory: 13405 grad_norm: 25.9592 loss: 0.9708 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.9708 2023/06/04 07:38:57 - mmengine - INFO - Epoch(train) [51][1500/2569] lr: 5.0956e-06 eta: 2:55:24 time: 0.9122 data_time: 0.0080 memory: 13405 grad_norm: 26.8954 loss: 1.1170 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1170 2023/06/04 07:39:43 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 07:40:29 - mmengine - INFO - Epoch(train) [51][1600/2569] lr: 5.0940e-06 eta: 2:53:51 time: 0.9233 data_time: 0.0083 memory: 13405 grad_norm: 26.4130 loss: 1.2664 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2664 2023/06/04 07:42:02 - mmengine - INFO - Epoch(train) [51][1700/2569] lr: 5.0923e-06 eta: 2:52:19 time: 0.9238 data_time: 0.0079 memory: 13405 grad_norm: 26.9445 loss: 1.0660 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0660 2023/06/04 07:43:34 - mmengine - INFO - Epoch(train) [51][1800/2569] lr: 5.0907e-06 eta: 2:50:45 time: 0.9184 data_time: 0.0082 memory: 13405 grad_norm: 25.3880 loss: 0.9154 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9154 2023/06/04 07:45:05 - mmengine - INFO - Epoch(train) [51][1900/2569] lr: 5.0891e-06 eta: 2:49:12 time: 0.9136 data_time: 0.0080 memory: 13405 grad_norm: 26.5480 loss: 1.0693 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0693 2023/06/04 07:46:38 - mmengine - INFO - Epoch(train) [51][2000/2569] lr: 5.0874e-06 eta: 2:47:39 time: 0.9445 data_time: 0.0083 memory: 13405 grad_norm: 25.6025 loss: 1.2418 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2418 2023/06/04 07:48:10 - mmengine - INFO - Epoch(train) [51][2100/2569] lr: 5.0858e-06 eta: 2:46:06 time: 0.9198 data_time: 0.0082 memory: 13405 grad_norm: 26.1385 loss: 1.1154 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1154 2023/06/04 07:49:42 - mmengine - INFO - Epoch(train) [51][2200/2569] lr: 5.0843e-06 eta: 2:44:33 time: 0.9268 data_time: 0.0083 memory: 13405 grad_norm: 27.0667 loss: 1.0880 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0880 2023/06/04 07:51:16 - mmengine - INFO - Epoch(train) [51][2300/2569] lr: 5.0827e-06 eta: 2:43:01 time: 0.9367 data_time: 0.0079 memory: 13405 grad_norm: 26.6595 loss: 0.8473 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8473 2023/06/04 07:52:50 - mmengine - INFO - Epoch(train) [51][2400/2569] lr: 5.0811e-06 eta: 2:41:29 time: 0.9196 data_time: 0.0086 memory: 13405 grad_norm: 25.5069 loss: 0.8780 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 0.8780 2023/06/04 07:54:23 - mmengine - INFO - Epoch(train) [51][2500/2569] lr: 5.0796e-06 eta: 2:39:56 time: 0.9228 data_time: 0.0081 memory: 13405 grad_norm: 26.1271 loss: 0.7456 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.7456 2023/06/04 07:55:09 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 07:55:26 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 07:55:26 - mmengine - INFO - Epoch(train) [51][2569/2569] lr: 5.0786e-06 eta: 2:38:51 time: 0.9015 data_time: 0.0084 memory: 13405 grad_norm: 26.3701 loss: 0.9656 top1_acc: 0.6667 top5_acc: 0.8333 loss_cls: 0.9656 2023/06/04 07:55:26 - mmengine - INFO - Saving checkpoint at 51 epochs 2023/06/04 07:55:55 - mmengine - INFO - Epoch(val) [51][100/260] eta: 0:00:37 time: 0.2233 data_time: 0.0297 memory: 2900 2023/06/04 07:56:16 - mmengine - INFO - Epoch(val) [51][200/260] eta: 0:00:13 time: 0.2101 data_time: 0.0162 memory: 2900 2023/06/04 07:56:35 - mmengine - INFO - Epoch(val) [51][260/260] acc/top1: 0.7692 acc/top5: 0.9325 acc/mean1: 0.7650 data_time: 0.0235 time: 0.2165 2023/06/04 07:58:10 - mmengine - INFO - Epoch(train) [52][ 100/2569] lr: 5.0770e-06 eta: 2:37:20 time: 0.9195 data_time: 0.0080 memory: 13405 grad_norm: 26.7996 loss: 1.1270 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1270 2023/06/04 07:59:44 - mmengine - INFO - Epoch(train) [52][ 200/2569] lr: 5.0755e-06 eta: 2:35:47 time: 0.9214 data_time: 0.0083 memory: 13405 grad_norm: 25.1627 loss: 1.0544 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0544 2023/06/04 08:01:17 - mmengine - INFO - Epoch(train) [52][ 300/2569] lr: 5.0741e-06 eta: 2:34:15 time: 0.9444 data_time: 0.0083 memory: 13405 grad_norm: 26.8295 loss: 0.9992 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 0.9992 2023/06/04 08:02:50 - mmengine - INFO - Epoch(train) [52][ 400/2569] lr: 5.0726e-06 eta: 2:32:42 time: 0.9133 data_time: 0.0084 memory: 13405 grad_norm: 26.6286 loss: 0.9871 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9871 2023/06/04 08:04:23 - mmengine - INFO - Epoch(train) [52][ 500/2569] lr: 5.0711e-06 eta: 2:31:09 time: 0.9257 data_time: 0.0083 memory: 13405 grad_norm: 26.5358 loss: 1.1738 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1738 2023/06/04 08:05:56 - mmengine - INFO - Epoch(train) [52][ 600/2569] lr: 5.0697e-06 eta: 2:29:37 time: 0.9339 data_time: 0.0090 memory: 13405 grad_norm: 26.4171 loss: 0.8619 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8619 2023/06/04 08:07:28 - mmengine - INFO - Epoch(train) [52][ 700/2569] lr: 5.0683e-06 eta: 2:28:04 time: 0.9447 data_time: 0.0085 memory: 13405 grad_norm: 26.5057 loss: 1.1256 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1256 2023/06/04 08:09:02 - mmengine - INFO - Epoch(train) [52][ 800/2569] lr: 5.0669e-06 eta: 2:26:31 time: 0.9115 data_time: 0.0085 memory: 13405 grad_norm: 27.0470 loss: 1.2635 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.2635 2023/06/04 08:10:34 - mmengine - INFO - Epoch(train) [52][ 900/2569] lr: 5.0655e-06 eta: 2:24:58 time: 0.9171 data_time: 0.0082 memory: 13405 grad_norm: 26.2084 loss: 0.7804 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.7804 2023/06/04 08:11:49 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 08:12:06 - mmengine - INFO - Epoch(train) [52][1000/2569] lr: 5.0641e-06 eta: 2:23:25 time: 0.9236 data_time: 0.0080 memory: 13405 grad_norm: 25.7920 loss: 0.9194 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9194 2023/06/04 08:13:39 - mmengine - INFO - Epoch(train) [52][1100/2569] lr: 5.0627e-06 eta: 2:21:52 time: 0.9306 data_time: 0.0083 memory: 13405 grad_norm: 25.4364 loss: 1.0045 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0045 2023/06/04 08:15:11 - mmengine - INFO - Epoch(train) [52][1200/2569] lr: 5.0614e-06 eta: 2:20:19 time: 0.9226 data_time: 0.0082 memory: 13405 grad_norm: 26.3082 loss: 1.0402 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0402 2023/06/04 08:16:44 - mmengine - INFO - Epoch(train) [52][1300/2569] lr: 5.0600e-06 eta: 2:18:46 time: 0.9271 data_time: 0.0076 memory: 13405 grad_norm: 26.8986 loss: 1.2398 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.2398 2023/06/04 08:18:18 - mmengine - INFO - Epoch(train) [52][1400/2569] lr: 5.0587e-06 eta: 2:17:14 time: 0.9238 data_time: 0.0083 memory: 13405 grad_norm: 26.3755 loss: 1.1481 top1_acc: 0.5000 top5_acc: 0.6250 loss_cls: 1.1481 2023/06/04 08:19:50 - mmengine - INFO - Epoch(train) [52][1500/2569] lr: 5.0574e-06 eta: 2:15:41 time: 0.9445 data_time: 0.0083 memory: 13405 grad_norm: 26.4096 loss: 0.9526 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9526 2023/06/04 08:21:23 - mmengine - INFO - Epoch(train) [52][1600/2569] lr: 5.0561e-06 eta: 2:14:08 time: 0.9151 data_time: 0.0082 memory: 13405 grad_norm: 27.3995 loss: 0.8429 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 0.8429 2023/06/04 08:22:55 - mmengine - INFO - Epoch(train) [52][1700/2569] lr: 5.0548e-06 eta: 2:12:36 time: 0.9331 data_time: 0.0083 memory: 13405 grad_norm: 26.7240 loss: 1.0494 top1_acc: 0.3750 top5_acc: 0.7500 loss_cls: 1.0494 2023/06/04 08:24:30 - mmengine - INFO - Epoch(train) [52][1800/2569] lr: 5.0535e-06 eta: 2:11:03 time: 0.9572 data_time: 0.0080 memory: 13405 grad_norm: 25.8095 loss: 1.2231 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.2231 2023/06/04 08:26:03 - mmengine - INFO - Epoch(train) [52][1900/2569] lr: 5.0523e-06 eta: 2:09:31 time: 0.9361 data_time: 0.0082 memory: 13405 grad_norm: 26.1768 loss: 0.9984 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9984 2023/06/04 08:27:18 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 08:27:36 - mmengine - INFO - Epoch(train) [52][2000/2569] lr: 5.0511e-06 eta: 2:07:58 time: 0.9440 data_time: 0.0097 memory: 13405 grad_norm: 25.7397 loss: 0.8776 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8776 2023/06/04 08:29:09 - mmengine - INFO - Epoch(train) [52][2100/2569] lr: 5.0498e-06 eta: 2:06:25 time: 0.9112 data_time: 0.0085 memory: 13405 grad_norm: 26.7377 loss: 1.1080 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.1080 2023/06/04 08:30:43 - mmengine - INFO - Epoch(train) [52][2200/2569] lr: 5.0486e-06 eta: 2:04:53 time: 0.9306 data_time: 0.0082 memory: 13405 grad_norm: 26.1827 loss: 1.1719 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.1719 2023/06/04 08:32:16 - mmengine - INFO - Epoch(train) [52][2300/2569] lr: 5.0474e-06 eta: 2:03:20 time: 0.9216 data_time: 0.0080 memory: 13405 grad_norm: 26.9423 loss: 1.0000 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.0000 2023/06/04 08:33:49 - mmengine - INFO - Epoch(train) [52][2400/2569] lr: 5.0463e-06 eta: 2:01:47 time: 0.9428 data_time: 0.0083 memory: 13405 grad_norm: 26.4525 loss: 1.1116 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.1116 2023/06/04 08:35:20 - mmengine - INFO - Epoch(train) [52][2500/2569] lr: 5.0451e-06 eta: 2:00:14 time: 0.9204 data_time: 0.0082 memory: 13405 grad_norm: 26.7369 loss: 1.4239 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.4239 2023/06/04 08:36:23 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 08:36:23 - mmengine - INFO - Epoch(train) [52][2569/2569] lr: 5.0443e-06 eta: 1:59:10 time: 0.8779 data_time: 0.0083 memory: 13405 grad_norm: 25.7812 loss: 0.8333 top1_acc: 0.8333 top5_acc: 0.8333 loss_cls: 0.8333 2023/06/04 08:36:48 - mmengine - INFO - Epoch(val) [52][100/260] eta: 0:00:39 time: 0.2340 data_time: 0.0406 memory: 2900 2023/06/04 08:37:10 - mmengine - INFO - Epoch(val) [52][200/260] eta: 0:00:14 time: 0.2334 data_time: 0.0397 memory: 2900 2023/06/04 08:37:32 - mmengine - INFO - Epoch(val) [52][260/260] acc/top1: 0.7707 acc/top5: 0.9328 acc/mean1: 0.7664 data_time: 0.0357 time: 0.2290 2023/06/04 08:39:09 - mmengine - INFO - Epoch(train) [53][ 100/2569] lr: 5.0432e-06 eta: 1:57:38 time: 0.9140 data_time: 0.0079 memory: 13405 grad_norm: 25.9070 loss: 0.7538 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.7538 2023/06/04 08:40:42 - mmengine - INFO - Epoch(train) [53][ 200/2569] lr: 5.0420e-06 eta: 1:56:05 time: 0.9574 data_time: 0.0082 memory: 13405 grad_norm: 26.2035 loss: 1.1209 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1209 2023/06/04 08:42:14 - mmengine - INFO - Epoch(train) [53][ 300/2569] lr: 5.0409e-06 eta: 1:54:33 time: 0.9406 data_time: 0.0083 memory: 13405 grad_norm: 25.9806 loss: 0.8867 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8867 2023/06/04 08:43:46 - mmengine - INFO - Epoch(train) [53][ 400/2569] lr: 5.0398e-06 eta: 1:53:00 time: 0.9166 data_time: 0.0085 memory: 13405 grad_norm: 25.9344 loss: 1.3705 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.3705 2023/06/04 08:43:58 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 08:45:19 - mmengine - INFO - Epoch(train) [53][ 500/2569] lr: 5.0387e-06 eta: 1:51:27 time: 0.9443 data_time: 0.0081 memory: 13405 grad_norm: 26.4346 loss: 1.3596 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.3596 2023/06/04 08:46:53 - mmengine - INFO - Epoch(train) [53][ 600/2569] lr: 5.0377e-06 eta: 1:49:54 time: 0.9490 data_time: 0.0082 memory: 13405 grad_norm: 26.4273 loss: 0.9352 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9352 2023/06/04 08:48:26 - mmengine - INFO - Epoch(train) [53][ 700/2569] lr: 5.0366e-06 eta: 1:48:22 time: 0.9167 data_time: 0.0080 memory: 13405 grad_norm: 26.3562 loss: 1.1444 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1444 2023/06/04 08:49:59 - mmengine - INFO - Epoch(train) [53][ 800/2569] lr: 5.0356e-06 eta: 1:46:49 time: 0.9377 data_time: 0.0084 memory: 13405 grad_norm: 26.8021 loss: 1.0654 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0654 2023/06/04 08:51:31 - mmengine - INFO - Epoch(train) [53][ 900/2569] lr: 5.0346e-06 eta: 1:45:16 time: 0.9201 data_time: 0.0083 memory: 13405 grad_norm: 27.1744 loss: 1.0225 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.0225 2023/06/04 08:53:03 - mmengine - INFO - Epoch(train) [53][1000/2569] lr: 5.0336e-06 eta: 1:43:43 time: 0.9258 data_time: 0.0084 memory: 13405 grad_norm: 26.2036 loss: 0.6290 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.6290 2023/06/04 08:54:36 - mmengine - INFO - Epoch(train) [53][1100/2569] lr: 5.0326e-06 eta: 1:42:10 time: 0.9146 data_time: 0.0079 memory: 13405 grad_norm: 26.1680 loss: 1.3954 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.3954 2023/06/04 08:56:09 - mmengine - INFO - Epoch(train) [53][1200/2569] lr: 5.0316e-06 eta: 1:40:37 time: 0.9589 data_time: 0.0081 memory: 13405 grad_norm: 26.9840 loss: 1.1802 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1802 2023/06/04 08:57:42 - mmengine - INFO - Epoch(train) [53][1300/2569] lr: 5.0306e-06 eta: 1:39:05 time: 0.9467 data_time: 0.0083 memory: 13405 grad_norm: 25.9472 loss: 0.8903 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8903 2023/06/04 08:59:15 - mmengine - INFO - Epoch(train) [53][1400/2569] lr: 5.0297e-06 eta: 1:37:32 time: 0.9593 data_time: 0.0085 memory: 13405 grad_norm: 26.1136 loss: 1.1364 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1364 2023/06/04 08:59:26 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 09:00:47 - mmengine - INFO - Epoch(train) [53][1500/2569] lr: 5.0288e-06 eta: 1:35:59 time: 0.9214 data_time: 0.0082 memory: 13405 grad_norm: 26.6169 loss: 0.9939 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.9939 2023/06/04 09:02:19 - mmengine - INFO - Epoch(train) [53][1600/2569] lr: 5.0278e-06 eta: 1:34:26 time: 0.9316 data_time: 0.0082 memory: 13405 grad_norm: 26.2882 loss: 0.8419 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8419 2023/06/04 09:03:52 - mmengine - INFO - Epoch(train) [53][1700/2569] lr: 5.0269e-06 eta: 1:32:53 time: 0.9268 data_time: 0.0087 memory: 13405 grad_norm: 26.3062 loss: 1.1224 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.1224 2023/06/04 09:05:25 - mmengine - INFO - Epoch(train) [53][1800/2569] lr: 5.0261e-06 eta: 1:31:20 time: 0.9268 data_time: 0.0078 memory: 13405 grad_norm: 25.9021 loss: 0.7947 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.7947 2023/06/04 09:06:57 - mmengine - INFO - Epoch(train) [53][1900/2569] lr: 5.0252e-06 eta: 1:29:47 time: 0.9272 data_time: 0.0082 memory: 13405 grad_norm: 26.7436 loss: 1.1165 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.1165 2023/06/04 09:08:29 - mmengine - INFO - Epoch(train) [53][2000/2569] lr: 5.0243e-06 eta: 1:28:14 time: 0.9258 data_time: 0.0084 memory: 13405 grad_norm: 26.3672 loss: 1.1539 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1539 2023/06/04 09:10:01 - mmengine - INFO - Epoch(train) [53][2100/2569] lr: 5.0235e-06 eta: 1:26:42 time: 0.9455 data_time: 0.0079 memory: 13405 grad_norm: 27.0666 loss: 1.1245 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.1245 2023/06/04 09:11:34 - mmengine - INFO - Epoch(train) [53][2200/2569] lr: 5.0227e-06 eta: 1:25:09 time: 0.9174 data_time: 0.0082 memory: 13405 grad_norm: 28.7932 loss: 1.1799 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1799 2023/06/04 09:13:06 - mmengine - INFO - Epoch(train) [53][2300/2569] lr: 5.0218e-06 eta: 1:23:36 time: 0.9273 data_time: 0.0082 memory: 13405 grad_norm: 26.5394 loss: 0.9329 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 0.9329 2023/06/04 09:14:40 - mmengine - INFO - Epoch(train) [53][2400/2569] lr: 5.0210e-06 eta: 1:22:03 time: 0.9282 data_time: 0.0079 memory: 13405 grad_norm: 27.1547 loss: 1.1159 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1159 2023/06/04 09:14:51 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 09:16:12 - mmengine - INFO - Epoch(train) [53][2500/2569] lr: 5.0203e-06 eta: 1:20:30 time: 0.9196 data_time: 0.0082 memory: 13405 grad_norm: 26.1723 loss: 0.8707 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 0.8707 2023/06/04 09:17:15 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 09:17:15 - mmengine - INFO - Epoch(train) [53][2569/2569] lr: 5.0197e-06 eta: 1:19:26 time: 0.8886 data_time: 0.0085 memory: 13405 grad_norm: 27.0934 loss: 0.9112 top1_acc: 0.5000 top5_acc: 0.6667 loss_cls: 0.9112 2023/06/04 09:17:39 - mmengine - INFO - Epoch(val) [53][100/260] eta: 0:00:37 time: 0.2258 data_time: 0.0323 memory: 2900 2023/06/04 09:18:01 - mmengine - INFO - Epoch(val) [53][200/260] eta: 0:00:13 time: 0.2033 data_time: 0.0095 memory: 2900 2023/06/04 09:18:23 - mmengine - INFO - Epoch(val) [53][260/260] acc/top1: 0.7710 acc/top5: 0.9324 acc/mean1: 0.7665 data_time: 0.0309 time: 0.2242 2023/06/04 09:19:59 - mmengine - INFO - Epoch(train) [54][ 100/2569] lr: 5.0190e-06 eta: 1:17:54 time: 0.9461 data_time: 0.0082 memory: 13405 grad_norm: 25.7450 loss: 0.8763 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8763 2023/06/04 09:21:33 - mmengine - INFO - Epoch(train) [54][ 200/2569] lr: 5.0182e-06 eta: 1:16:21 time: 0.9447 data_time: 0.0083 memory: 13405 grad_norm: 27.2519 loss: 1.1096 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1096 2023/06/04 09:23:05 - mmengine - INFO - Epoch(train) [54][ 300/2569] lr: 5.0175e-06 eta: 1:14:48 time: 0.9194 data_time: 0.0079 memory: 13405 grad_norm: 26.0983 loss: 0.8856 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 0.8856 2023/06/04 09:24:38 - mmengine - INFO - Epoch(train) [54][ 400/2569] lr: 5.0168e-06 eta: 1:13:16 time: 0.9108 data_time: 0.0084 memory: 13405 grad_norm: 26.8336 loss: 0.8975 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 0.8975 2023/06/04 09:26:10 - mmengine - INFO - Epoch(train) [54][ 500/2569] lr: 5.0161e-06 eta: 1:11:43 time: 0.9475 data_time: 0.0088 memory: 13405 grad_norm: 26.3538 loss: 1.0484 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0484 2023/06/04 09:27:42 - mmengine - INFO - Epoch(train) [54][ 600/2569] lr: 5.0154e-06 eta: 1:10:10 time: 0.9194 data_time: 0.0084 memory: 13405 grad_norm: 25.9625 loss: 1.0422 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0422 2023/06/04 09:29:14 - mmengine - INFO - Epoch(train) [54][ 700/2569] lr: 5.0147e-06 eta: 1:08:37 time: 0.9149 data_time: 0.0081 memory: 13405 grad_norm: 26.7914 loss: 1.1847 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.1847 2023/06/04 09:30:46 - mmengine - INFO - Epoch(train) [54][ 800/2569] lr: 5.0141e-06 eta: 1:07:04 time: 0.9097 data_time: 0.0082 memory: 13405 grad_norm: 25.8546 loss: 1.0756 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0756 2023/06/04 09:31:27 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 09:32:20 - mmengine - INFO - Epoch(train) [54][ 900/2569] lr: 5.0134e-06 eta: 1:05:31 time: 0.9508 data_time: 0.0081 memory: 13405 grad_norm: 26.4585 loss: 0.7983 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.7983 2023/06/04 09:33:54 - mmengine - INFO - Epoch(train) [54][1000/2569] lr: 5.0128e-06 eta: 1:03:59 time: 0.9517 data_time: 0.0081 memory: 13405 grad_norm: 26.6539 loss: 1.0004 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.0004 2023/06/04 09:35:28 - mmengine - INFO - Epoch(train) [54][1100/2569] lr: 5.0122e-06 eta: 1:02:26 time: 0.9191 data_time: 0.0082 memory: 13405 grad_norm: 26.7557 loss: 1.2508 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.2508 2023/06/04 09:37:00 - mmengine - INFO - Epoch(train) [54][1200/2569] lr: 5.0116e-06 eta: 1:00:53 time: 0.9238 data_time: 0.0083 memory: 13405 grad_norm: 27.4095 loss: 1.0191 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0191 2023/06/04 09:38:32 - mmengine - INFO - Epoch(train) [54][1300/2569] lr: 5.0110e-06 eta: 0:59:20 time: 0.9253 data_time: 0.0082 memory: 13405 grad_norm: 27.4865 loss: 0.8310 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8310 2023/06/04 09:40:04 - mmengine - INFO - Epoch(train) [54][1400/2569] lr: 5.0104e-06 eta: 0:57:48 time: 0.9567 data_time: 0.0083 memory: 13405 grad_norm: 26.6440 loss: 1.2597 top1_acc: 0.5000 top5_acc: 0.5000 loss_cls: 1.2597 2023/06/04 09:41:37 - mmengine - INFO - Epoch(train) [54][1500/2569] lr: 5.0099e-06 eta: 0:56:15 time: 0.9155 data_time: 0.0077 memory: 13405 grad_norm: 26.6393 loss: 1.2969 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 1.2969 2023/06/04 09:43:08 - mmengine - INFO - Epoch(train) [54][1600/2569] lr: 5.0094e-06 eta: 0:54:42 time: 0.9083 data_time: 0.0080 memory: 13405 grad_norm: 26.4444 loss: 1.2466 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2466 2023/06/04 09:44:40 - mmengine - INFO - Epoch(train) [54][1700/2569] lr: 5.0088e-06 eta: 0:53:09 time: 0.9163 data_time: 0.0079 memory: 13405 grad_norm: 26.3473 loss: 1.0088 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0088 2023/06/04 09:46:12 - mmengine - INFO - Epoch(train) [54][1800/2569] lr: 5.0083e-06 eta: 0:51:36 time: 0.9281 data_time: 0.0081 memory: 13405 grad_norm: 26.8396 loss: 1.0198 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0198 2023/06/04 09:46:52 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 09:47:44 - mmengine - INFO - Epoch(train) [54][1900/2569] lr: 5.0078e-06 eta: 0:50:03 time: 0.9052 data_time: 0.0084 memory: 13405 grad_norm: 26.9551 loss: 1.0920 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0920 2023/06/04 09:49:16 - mmengine - INFO - Epoch(train) [54][2000/2569] lr: 5.0074e-06 eta: 0:48:30 time: 0.9143 data_time: 0.0079 memory: 13405 grad_norm: 26.1796 loss: 0.9937 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9937 2023/06/04 09:50:47 - mmengine - INFO - Epoch(train) [54][2100/2569] lr: 5.0069e-06 eta: 0:46:57 time: 0.9103 data_time: 0.0082 memory: 13405 grad_norm: 26.0189 loss: 0.7915 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.7915 2023/06/04 09:52:20 - mmengine - INFO - Epoch(train) [54][2200/2569] lr: 5.0065e-06 eta: 0:45:25 time: 0.9204 data_time: 0.0081 memory: 13405 grad_norm: 26.2256 loss: 1.0802 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.0802 2023/06/04 09:53:52 - mmengine - INFO - Epoch(train) [54][2300/2569] lr: 5.0060e-06 eta: 0:43:52 time: 0.9145 data_time: 0.0081 memory: 13405 grad_norm: 26.6474 loss: 0.9653 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9653 2023/06/04 09:55:24 - mmengine - INFO - Epoch(train) [54][2400/2569] lr: 5.0056e-06 eta: 0:42:19 time: 0.9523 data_time: 0.0082 memory: 13405 grad_norm: 26.2782 loss: 0.9446 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.9446 2023/06/04 09:56:57 - mmengine - INFO - Epoch(train) [54][2500/2569] lr: 5.0052e-06 eta: 0:40:46 time: 0.9137 data_time: 0.0081 memory: 13405 grad_norm: 26.5599 loss: 0.8835 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8835 2023/06/04 09:58:00 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 09:58:00 - mmengine - INFO - Epoch(train) [54][2569/2569] lr: 5.0049e-06 eta: 0:39:42 time: 0.8959 data_time: 0.0078 memory: 13405 grad_norm: 26.3074 loss: 1.1193 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1193 2023/06/04 09:58:00 - mmengine - INFO - Saving checkpoint at 54 epochs 2023/06/04 09:58:30 - mmengine - INFO - Epoch(val) [54][100/260] eta: 0:00:38 time: 0.2406 data_time: 0.0471 memory: 2900 2023/06/04 09:58:52 - mmengine - INFO - Epoch(val) [54][200/260] eta: 0:00:13 time: 0.2036 data_time: 0.0091 memory: 2900 2023/06/04 09:59:11 - mmengine - INFO - Epoch(val) [54][260/260] acc/top1: 0.7703 acc/top5: 0.9312 acc/mean1: 0.7656 data_time: 0.0298 time: 0.2229 2023/06/04 10:00:48 - mmengine - INFO - Epoch(train) [55][ 100/2569] lr: 5.0046e-06 eta: 0:38:10 time: 0.9388 data_time: 0.0083 memory: 13405 grad_norm: 27.2711 loss: 1.0264 top1_acc: 0.7500 top5_acc: 0.7500 loss_cls: 1.0264 2023/06/04 10:02:20 - mmengine - INFO - Epoch(train) [55][ 200/2569] lr: 5.0042e-06 eta: 0:36:37 time: 0.9405 data_time: 0.0080 memory: 13405 grad_norm: 27.1629 loss: 0.8967 top1_acc: 0.7500 top5_acc: 0.8750 loss_cls: 0.8967 2023/06/04 10:03:29 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 10:03:53 - mmengine - INFO - Epoch(train) [55][ 300/2569] lr: 5.0039e-06 eta: 0:35:04 time: 0.9127 data_time: 0.0082 memory: 13405 grad_norm: 26.8170 loss: 0.8347 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.8347 2023/06/04 10:05:25 - mmengine - INFO - Epoch(train) [55][ 400/2569] lr: 5.0035e-06 eta: 0:33:32 time: 0.9285 data_time: 0.0081 memory: 13405 grad_norm: 26.3792 loss: 1.3616 top1_acc: 0.6250 top5_acc: 1.0000 loss_cls: 1.3616 2023/06/04 10:06:57 - mmengine - INFO - Epoch(train) [55][ 500/2569] lr: 5.0032e-06 eta: 0:31:59 time: 0.9202 data_time: 0.0078 memory: 13405 grad_norm: 26.4654 loss: 1.1059 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 1.1059 2023/06/04 10:08:30 - mmengine - INFO - Epoch(train) [55][ 600/2569] lr: 5.0029e-06 eta: 0:30:26 time: 0.9319 data_time: 0.0082 memory: 13405 grad_norm: 26.0410 loss: 0.9925 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9925 2023/06/04 10:10:01 - mmengine - INFO - Epoch(train) [55][ 700/2569] lr: 5.0026e-06 eta: 0:28:53 time: 0.9158 data_time: 0.0080 memory: 13405 grad_norm: 25.6894 loss: 1.0459 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.0459 2023/06/04 10:11:34 - mmengine - INFO - Epoch(train) [55][ 800/2569] lr: 5.0023e-06 eta: 0:27:20 time: 0.9524 data_time: 0.0083 memory: 13405 grad_norm: 25.8089 loss: 1.2486 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2486 2023/06/04 10:13:06 - mmengine - INFO - Epoch(train) [55][ 900/2569] lr: 5.0021e-06 eta: 0:25:48 time: 0.9178 data_time: 0.0083 memory: 13405 grad_norm: 26.5860 loss: 1.0378 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 1.0378 2023/06/04 10:14:39 - mmengine - INFO - Epoch(train) [55][1000/2569] lr: 5.0018e-06 eta: 0:24:15 time: 0.9674 data_time: 0.0081 memory: 13405 grad_norm: 25.9445 loss: 0.8043 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.8043 2023/06/04 10:16:11 - mmengine - INFO - Epoch(train) [55][1100/2569] lr: 5.0016e-06 eta: 0:22:42 time: 0.9040 data_time: 0.0078 memory: 13405 grad_norm: 26.5169 loss: 1.1154 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 1.1154 2023/06/04 10:17:43 - mmengine - INFO - Epoch(train) [55][1200/2569] lr: 5.0014e-06 eta: 0:21:09 time: 0.9178 data_time: 0.0085 memory: 13405 grad_norm: 26.2988 loss: 1.1229 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.1229 2023/06/04 10:18:53 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 10:19:17 - mmengine - INFO - Epoch(train) [55][1300/2569] lr: 5.0012e-06 eta: 0:19:37 time: 0.9210 data_time: 0.0080 memory: 13405 grad_norm: 26.3425 loss: 1.0285 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0285 2023/06/04 10:20:49 - mmengine - INFO - Epoch(train) [55][1400/2569] lr: 5.0010e-06 eta: 0:18:04 time: 0.9332 data_time: 0.0083 memory: 13405 grad_norm: 26.9390 loss: 1.2378 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.2378 2023/06/04 10:22:21 - mmengine - INFO - Epoch(train) [55][1500/2569] lr: 5.0009e-06 eta: 0:16:31 time: 0.9121 data_time: 0.0080 memory: 13405 grad_norm: 26.9758 loss: 1.2592 top1_acc: 0.6250 top5_acc: 0.7500 loss_cls: 1.2592 2023/06/04 10:23:54 - mmengine - INFO - Epoch(train) [55][1600/2569] lr: 5.0007e-06 eta: 0:14:58 time: 0.9217 data_time: 0.0082 memory: 13405 grad_norm: 26.8235 loss: 1.2430 top1_acc: 0.5000 top5_acc: 0.7500 loss_cls: 1.2430 2023/06/04 10:25:26 - mmengine - INFO - Epoch(train) [55][1700/2569] lr: 5.0006e-06 eta: 0:13:25 time: 0.9189 data_time: 0.0082 memory: 13405 grad_norm: 26.4867 loss: 1.0604 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 1.0604 2023/06/04 10:26:59 - mmengine - INFO - Epoch(train) [55][1800/2569] lr: 5.0004e-06 eta: 0:11:53 time: 0.9202 data_time: 0.0081 memory: 13405 grad_norm: 26.0430 loss: 1.2376 top1_acc: 0.5000 top5_acc: 0.8750 loss_cls: 1.2376 2023/06/04 10:28:32 - mmengine - INFO - Epoch(train) [55][1900/2569] lr: 5.0003e-06 eta: 0:10:20 time: 0.9567 data_time: 0.0082 memory: 13405 grad_norm: 26.4004 loss: 1.1025 top1_acc: 0.3750 top5_acc: 0.6250 loss_cls: 1.1025 2023/06/04 10:30:04 - mmengine - INFO - Epoch(train) [55][2000/2569] lr: 5.0002e-06 eta: 0:08:47 time: 0.9166 data_time: 0.0082 memory: 13405 grad_norm: 27.3374 loss: 0.9775 top1_acc: 0.6250 top5_acc: 0.8750 loss_cls: 0.9775 2023/06/04 10:31:37 - mmengine - INFO - Epoch(train) [55][2100/2569] lr: 5.0002e-06 eta: 0:07:14 time: 0.9112 data_time: 0.0080 memory: 13405 grad_norm: 26.3767 loss: 1.1218 top1_acc: 0.5000 top5_acc: 1.0000 loss_cls: 1.1218 2023/06/04 10:33:11 - mmengine - INFO - Epoch(train) [55][2200/2569] lr: 5.0001e-06 eta: 0:05:42 time: 0.9334 data_time: 0.0085 memory: 13405 grad_norm: 26.7424 loss: 0.8818 top1_acc: 0.8750 top5_acc: 0.8750 loss_cls: 0.8818 2023/06/04 10:34:20 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 10:34:44 - mmengine - INFO - Epoch(train) [55][2300/2569] lr: 5.0001e-06 eta: 0:04:09 time: 0.9299 data_time: 0.0080 memory: 13405 grad_norm: 26.4666 loss: 0.9004 top1_acc: 0.8750 top5_acc: 1.0000 loss_cls: 0.9004 2023/06/04 10:36:17 - mmengine - INFO - Epoch(train) [55][2400/2569] lr: 5.0000e-06 eta: 0:02:36 time: 0.9320 data_time: 0.0080 memory: 13405 grad_norm: 27.2124 loss: 0.9207 top1_acc: 1.0000 top5_acc: 1.0000 loss_cls: 0.9207 2023/06/04 10:37:50 - mmengine - INFO - Epoch(train) [55][2500/2569] lr: 5.0000e-06 eta: 0:01:03 time: 0.9185 data_time: 0.0077 memory: 13405 grad_norm: 26.8244 loss: 1.0139 top1_acc: 0.7500 top5_acc: 1.0000 loss_cls: 1.0139 2023/06/04 10:38:53 - mmengine - INFO - Exp name: uniformerv2-base-p16-res224_clip-pre_u8_kinetics710-rgb_train_20230604_014810 2023/06/04 10:38:53 - mmengine - INFO - Epoch(train) [55][2569/2569] lr: 5.0000e-06 eta: 0:00:00 time: 0.8837 data_time: 0.0081 memory: 13405 grad_norm: 27.7346 loss: 1.2001 top1_acc: 0.5000 top5_acc: 0.8333 loss_cls: 1.2001 2023/06/04 10:38:53 - mmengine - INFO - Saving checkpoint at 55 epochs 2023/06/04 10:39:23 - mmengine - INFO - Epoch(val) [55][100/260] eta: 0:00:39 time: 0.2382 data_time: 0.0446 memory: 2900 2023/06/04 10:39:45 - mmengine - INFO - Epoch(val) [55][200/260] eta: 0:00:13 time: 0.2018 data_time: 0.0078 memory: 2900 2023/06/04 10:40:04 - mmengine - INFO - Epoch(val) [55][260/260] acc/top1: 0.7697 acc/top5: 0.9305 acc/mean1: 0.7651 data_time: 0.0333 time: 0.2264