2022/09/26 10:36:35 - mmengine - INFO - ------------------------------------------------------------ System environment: sys.platform: linux Python: 3.8.13 (default, Mar 28 2022, 11:38:47) [GCC 7.5.0] CUDA available: True numpy_random_seed: 924417172 GPU 0,1,2,3,4,5,6,7: NVIDIA A100-SXM4-80GB CUDA_HOME: /mnt/lustre/share/cuda-11.3 NVCC: Cuda compilation tools, release 11.3, V11.3.109 GCC: gcc (GCC) 5.4.0 PyTorch: 1.12.1 PyTorch compiling details: PyTorch built with: - GCC 9.3 - C++ Version: 201402 - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 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_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 - CuDNN 8.3.2 (built against CUDA 11.5) - 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 OpenCV: 4.6.0 MMEngine: 0.1.0 Runtime environment: cudnn_benchmark: False mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} dist_cfg: {'backend': 'nccl'} seed: None Distributed launcher: slurm Distributed training: True GPU number: 8 ------------------------------------------------------------ 2022/09/26 10:36:37 - mmengine - INFO - Config: default_scope = 'mmpose' default_hooks = dict( timer=dict(type='IterTimerHook'), logger=dict(type='LoggerHook', interval=50), param_scheduler=dict(type='ParamSchedulerHook'), checkpoint=dict( type='CheckpointHook', interval=10, max_keep_ckpts=1, save_best='coco/AP', rule='greater'), sampler_seed=dict(type='DistSamplerSeedHook'), visualization=dict(type='PoseVisualizationHook', enable=False)) custom_hooks = [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')) vis_backends = [dict(type='LocalVisBackend')] visualizer = dict( type='PoseLocalVisualizer', vis_backends=[dict(type='LocalVisBackend')], name='visualizer') log_processor = dict( type='LogProcessor', window_size=50, by_epoch=True, num_digits=6) log_level = 'INFO' load_from = None resume = False file_client_args = dict( backend='petrel', path_mapping=dict({ './data/': 's3://openmmlab/datasets/detection/', 'data/': 's3://openmmlab/datasets/detection/' })) train_cfg = dict(by_epoch=True, max_epochs=210, val_interval=10) val_cfg = dict() test_cfg = dict() optim_wrapper = dict(optimizer=dict(type='Adam', lr=0.0005)) param_scheduler = [ dict( type='LinearLR', begin=0, end=500, start_factor=0.001, by_epoch=False), dict( type='MultiStepLR', begin=0, end=210, milestones=[170, 200], gamma=0.1, by_epoch=True) ] auto_scale_lr = dict(base_batch_size=512) codec = dict( type='MSRAHeatmap', input_size=(192, 256), heatmap_size=(48, 64), sigma=2) model = dict( type='TopdownPoseEstimator', data_preprocessor=dict( type='PoseDataPreprocessor', mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], bgr_to_rgb=True), backbone=dict( type='ResNet', depth=101, init_cfg=dict(type='Pretrained', checkpoint='torchvision://resnet101')), head=dict( type='HeatmapHead', in_channels=2048, out_channels=17, loss=dict(type='KeypointMSELoss', use_target_weight=True), decoder=dict( type='MSRAHeatmap', input_size=(192, 256), heatmap_size=(48, 64), sigma=2)), test_cfg=dict(flip_test=True, flip_mode='heatmap', shift_heatmap=True)) dataset_type = 'CocoDataset' data_mode = 'topdown' data_root = 'data/coco/' train_pipeline = [ dict( type='LoadImage', file_client_args=dict( backend='petrel', path_mapping=dict({ './data/': 's3://openmmlab/datasets/detection/', 'data/': 's3://openmmlab/datasets/detection/' }))), dict(type='GetBBoxCenterScale'), dict(type='RandomFlip', direction='horizontal'), dict(type='RandomHalfBody'), dict(type='RandomBBoxTransform'), dict(type='TopdownAffine', input_size=(192, 256)), dict( type='GenerateTarget', target_type='heatmap', encoder=dict( type='MSRAHeatmap', input_size=(192, 256), heatmap_size=(48, 64), sigma=2)), dict(type='PackPoseInputs') ] val_pipeline = [ dict( type='LoadImage', file_client_args=dict( backend='petrel', path_mapping=dict({ './data/': 's3://openmmlab/datasets/detection/', 'data/': 's3://openmmlab/datasets/detection/' }))), dict(type='GetBBoxCenterScale'), dict(type='TopdownAffine', input_size=(192, 256)), dict(type='PackPoseInputs') ] train_dataloader = dict( batch_size=64, num_workers=2, persistent_workers=True, sampler=dict(type='DefaultSampler', shuffle=True), dataset=dict( type='CocoDataset', data_root='data/coco/', data_mode='topdown', ann_file='annotations/person_keypoints_train2017.json', data_prefix=dict(img='train2017/'), pipeline=[ dict( type='LoadImage', file_client_args=dict( backend='petrel', path_mapping=dict({ './data/': 's3://openmmlab/datasets/detection/', 'data/': 's3://openmmlab/datasets/detection/' }))), dict(type='GetBBoxCenterScale'), dict(type='RandomFlip', direction='horizontal'), dict(type='RandomHalfBody'), dict(type='RandomBBoxTransform'), dict(type='TopdownAffine', input_size=(192, 256)), dict( type='GenerateTarget', target_type='heatmap', encoder=dict( type='MSRAHeatmap', input_size=(192, 256), heatmap_size=(48, 64), sigma=2)), dict(type='PackPoseInputs') ])) val_dataloader = dict( batch_size=32, num_workers=2, persistent_workers=True, drop_last=False, sampler=dict(type='DefaultSampler', shuffle=False, round_up=False), dataset=dict( type='CocoDataset', data_root='data/coco/', data_mode='topdown', ann_file='annotations/person_keypoints_val2017.json', bbox_file= 'data/coco/person_detection_results/COCO_val2017_detections_AP_H_56_person.json', data_prefix=dict(img='val2017/'), test_mode=True, pipeline=[ dict( type='LoadImage', file_client_args=dict( backend='petrel', path_mapping=dict({ './data/': 's3://openmmlab/datasets/detection/', 'data/': 's3://openmmlab/datasets/detection/' }))), dict(type='GetBBoxCenterScale'), dict(type='TopdownAffine', input_size=(192, 256)), dict(type='PackPoseInputs') ])) test_dataloader = dict( batch_size=32, num_workers=2, persistent_workers=True, drop_last=False, sampler=dict(type='DefaultSampler', shuffle=False, round_up=False), dataset=dict( type='CocoDataset', data_root='data/coco/', data_mode='topdown', ann_file='annotations/person_keypoints_val2017.json', bbox_file= 'data/coco/person_detection_results/COCO_val2017_detections_AP_H_56_person.json', data_prefix=dict(img='val2017/'), test_mode=True, pipeline=[ dict( type='LoadImage', file_client_args=dict( backend='petrel', path_mapping=dict({ './data/': 's3://openmmlab/datasets/detection/', 'data/': 's3://openmmlab/datasets/detection/' }))), dict(type='GetBBoxCenterScale'), dict(type='TopdownAffine', input_size=(192, 256)), dict(type='PackPoseInputs') ])) val_evaluator = dict( type='CocoMetric', ann_file='data/coco/annotations/person_keypoints_val2017.json') test_evaluator = dict( type='CocoMetric', ann_file='data/coco/annotations/person_keypoints_val2017.json') launcher = 'slurm' work_dir = '/mnt/lustre/liqikai/work_dirs/20220926/r101_256/' 2022/09/26 10:37:18 - mmengine - WARNING - Failed to search registry with scope "mmpose" in the "data sampler" registry tree. As a workaround, the current "data sampler" registry in "mmengine" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "mmpose" is a correct scope, or whether the registry is initialized. 2022/09/26 10:37:18 - mmengine - WARNING - Failed to search registry with scope "mmpose" in the "optimizer wrapper constructor" registry tree. As a workaround, the current "optimizer wrapper constructor" registry in "mmengine" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "mmpose" is a correct scope, or whether the registry is initialized. 2022/09/26 10:37:18 - mmengine - WARNING - Failed to search registry with scope "mmpose" in the "optimizer" registry tree. As a workaround, the current "optimizer" registry in "mmengine" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "mmpose" is a correct scope, or whether the registry is initialized. 2022/09/26 10:37:18 - mmengine - WARNING - Failed to search registry with scope "mmpose" in the "optim_wrapper" registry tree. As a workaround, the current "optim_wrapper" registry in "mmengine" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "mmpose" is a correct scope, or whether the registry is initialized. 2022/09/26 10:37:18 - mmengine - WARNING - Failed to search registry with scope "mmpose" in the "parameter scheduler" registry tree. As a workaround, the current "parameter scheduler" registry in "mmengine" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "mmpose" is a correct scope, or whether the registry is initialized. 2022/09/26 10:37:18 - mmengine - WARNING - Failed to search registry with scope "mmpose" in the "parameter scheduler" registry tree. As a workaround, the current "parameter scheduler" registry in "mmengine" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "mmpose" is a correct scope, or whether the registry is initialized. 2022/09/26 10:37:18 - mmengine - WARNING - Failed to search registry with scope "mmpose" in the "parameter scheduler" registry tree. As a workaround, the current "parameter scheduler" registry in "mmengine" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "mmpose" is a correct scope, or whether the registry is initialized. 2022/09/26 10:37:18 - mmengine - WARNING - Failed to search registry with scope "mmpose" in the "parameter scheduler" registry tree. As a workaround, the current "parameter scheduler" registry in "mmengine" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "mmpose" is a correct scope, or whether the registry is initialized. 2022/09/26 10:37:21 - mmengine - WARNING - Failed to search registry with scope "mmpose" in the "data sampler" registry tree. As a workaround, the current "data sampler" registry in "mmengine" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "mmpose" is a correct scope, or whether the registry is initialized. 2022/09/26 10:37:23 - mmengine - WARNING - Failed to search registry with scope "mmpose" in the "weight initializer" registry tree. As a workaround, the current "weight initializer" registry in "mmengine" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "mmpose" is a correct scope, or whether the registry is initialized. 2022/09/26 10:37:26 - mmengine - WARNING - Failed to search registry with scope "mmpose" in the "weight initializer" registry tree. As a workaround, the current "weight initializer" registry in "mmengine" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "mmpose" is a correct scope, or whether the registry is initialized. 2022/09/26 10:37:26 - mmengine - WARNING - Failed to search registry with scope "mmpose" in the "weight initializer" registry tree. As a workaround, the current "weight initializer" registry in "mmengine" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "mmpose" is a correct scope, or whether the registry is initialized. Name of parameter - Initialization information backbone.conv1.weight - torch.Size([64, 3, 7, 7]): PretrainedInit: load from torchvision://resnet101 backbone.bn1.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet101 backbone.bn1.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.0.bn1.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.0.bn1.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.0.bn2.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.0.bn2.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.0.bn3.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.0.bn3.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.0.downsample.1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.0.downsample.1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.1.bn1.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.1.bn1.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.1.bn2.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.1.bn2.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.1.bn3.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.1.bn3.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.2.bn1.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.2.bn1.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.2.bn2.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.2.bn2.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.2.bn3.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer1.2.bn3.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.0.bn1.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.0.bn1.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.0.bn2.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.0.bn2.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.0.bn3.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.0.bn3.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.0.downsample.1.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.0.downsample.1.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.1.bn1.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.1.bn1.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.1.bn2.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.1.bn2.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.1.bn3.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.1.bn3.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.2.bn1.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.2.bn1.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.2.bn2.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.2.bn2.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.2.bn3.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.2.bn3.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.3.bn1.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.3.bn1.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.3.bn2.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.3.bn2.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.3.bn3.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer2.3.bn3.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.0.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.0.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.0.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.0.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.0.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.0.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.0.downsample.1.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.0.downsample.1.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.1.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.1.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.1.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.1.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.1.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.1.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.2.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.2.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.2.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.2.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.2.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.2.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.3.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.3.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.3.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.3.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.3.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.3.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.4.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.4.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.4.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.4.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.4.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.4.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.5.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.5.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.5.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.5.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.5.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.5.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.6.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.6.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.6.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.6.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.6.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.6.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.6.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.6.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.6.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.7.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.7.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.7.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.7.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.7.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.7.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.7.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.7.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.7.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.8.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.8.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.8.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.8.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.8.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.8.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.8.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.8.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.8.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.9.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.9.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.9.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.9.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.9.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.9.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.9.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.9.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.9.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.10.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.10.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.10.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.10.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.10.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.10.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.10.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.10.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.10.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.11.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.11.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.11.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.11.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.11.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.11.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.11.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.11.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.11.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.12.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.12.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.12.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.12.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.12.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.12.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.12.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.12.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.12.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.13.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.13.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.13.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.13.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.13.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.13.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.13.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.13.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.13.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.14.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.14.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.14.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.14.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.14.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.14.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.14.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.14.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.14.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.15.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.15.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.15.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.15.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.15.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.15.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.15.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.15.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.15.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.16.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.16.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.16.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.16.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.16.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.16.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.16.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.16.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.16.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.17.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.17.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.17.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.17.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.17.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.17.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.17.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.17.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.17.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.18.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.18.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.18.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.18.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.18.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.18.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.18.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.18.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.18.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.19.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.19.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.19.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.19.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.19.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.19.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.19.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.19.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.19.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.20.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.20.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.20.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.20.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.20.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.20.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.20.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.20.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.20.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.21.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.21.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.21.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.21.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.21.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.21.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.21.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.21.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.21.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.22.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.22.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.22.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.22.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.22.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.22.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.22.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.22.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer3.22.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.0.bn1.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.0.bn1.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.0.bn2.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.0.bn2.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.0.bn3.weight - torch.Size([2048]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.0.bn3.bias - torch.Size([2048]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.0.downsample.1.weight - torch.Size([2048]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.0.downsample.1.bias - torch.Size([2048]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.1.bn1.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.1.bn1.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.1.bn2.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.1.bn2.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.1.bn3.weight - torch.Size([2048]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.1.bn3.bias - torch.Size([2048]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.2.bn1.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.2.bn1.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.2.bn2.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.2.bn2.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.2.bn3.weight - torch.Size([2048]): PretrainedInit: load from torchvision://resnet101 backbone.layer4.2.bn3.bias - torch.Size([2048]): PretrainedInit: load from torchvision://resnet101 head.deconv_layers.0.weight - torch.Size([2048, 256, 4, 4]): NormalInit: mean=0, std=0.001, bias=0 head.deconv_layers.1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of TopdownPoseEstimator head.deconv_layers.1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of TopdownPoseEstimator head.deconv_layers.3.weight - torch.Size([256, 256, 4, 4]): NormalInit: mean=0, std=0.001, bias=0 head.deconv_layers.4.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of TopdownPoseEstimator head.deconv_layers.4.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of TopdownPoseEstimator head.deconv_layers.6.weight - torch.Size([256, 256, 4, 4]): NormalInit: mean=0, std=0.001, bias=0 head.deconv_layers.7.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of TopdownPoseEstimator head.deconv_layers.7.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of TopdownPoseEstimator head.final_layer.weight - torch.Size([17, 256, 1, 1]): NormalInit: mean=0, std=0.001, bias=0 head.final_layer.bias - torch.Size([17]): NormalInit: mean=0, std=0.001, bias=0 2022/09/26 10:37:26 - mmengine - INFO - Checkpoints will be saved to /mnt/lustre/liqikai/work_dirs/20220926/r101_256 by HardDiskBackend. 2022/09/26 10:38:22 - mmengine - INFO - Epoch(train) [1][50/293] lr: 4.954910e-05 eta: 18:56:58 time: 1.109604 data_time: 0.503543 memory: 9504 loss_kpt: 0.002145 acc_pose: 0.172327 loss: 0.002145 2022/09/26 10:38:53 - mmengine - INFO - Epoch(train) [1][100/293] lr: 9.959920e-05 eta: 14:52:28 time: 0.633792 data_time: 0.096152 memory: 9504 loss_kpt: 0.001744 acc_pose: 0.393162 loss: 0.001744 2022/09/26 10:39:23 - mmengine - INFO - Epoch(train) [1][150/293] lr: 1.496493e-04 eta: 13:15:29 time: 0.589424 data_time: 0.096855 memory: 9504 loss_kpt: 0.001482 acc_pose: 0.554010 loss: 0.001482 2022/09/26 10:39:53 - mmengine - INFO - Epoch(train) [1][200/293] lr: 1.996994e-04 eta: 12:29:24 time: 0.599789 data_time: 0.075173 memory: 9504 loss_kpt: 0.001333 acc_pose: 0.570047 loss: 0.001333 2022/09/26 10:40:23 - mmengine - INFO - Epoch(train) [1][250/293] lr: 2.497495e-04 eta: 12:02:21 time: 0.603740 data_time: 0.169790 memory: 9504 loss_kpt: 0.001268 acc_pose: 0.569743 loss: 0.001268 2022/09/26 10:40:47 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 10:41:17 - mmengine - INFO - Epoch(train) [2][50/293] lr: 3.428427e-04 eta: 10:14:50 time: 0.599683 data_time: 0.128484 memory: 9504 loss_kpt: 0.001207 acc_pose: 0.623991 loss: 0.001207 2022/09/26 10:41:48 - mmengine - INFO - Epoch(train) [2][100/293] lr: 3.928928e-04 eta: 10:16:07 time: 0.616674 data_time: 0.088517 memory: 9504 loss_kpt: 0.001183 acc_pose: 0.586739 loss: 0.001183 2022/09/26 10:42:18 - mmengine - INFO - Epoch(train) [2][150/293] lr: 4.429429e-04 eta: 10:16:13 time: 0.609889 data_time: 0.104803 memory: 9504 loss_kpt: 0.001153 acc_pose: 0.673663 loss: 0.001153 2022/09/26 10:42:48 - mmengine - INFO - Epoch(train) [2][200/293] lr: 4.929930e-04 eta: 10:14:10 time: 0.590218 data_time: 0.176200 memory: 9504 loss_kpt: 0.001150 acc_pose: 0.600116 loss: 0.001150 2022/09/26 10:43:16 - mmengine - INFO - Epoch(train) [2][250/293] lr: 5.000000e-04 eta: 10:10:26 time: 0.569358 data_time: 0.174133 memory: 9504 loss_kpt: 0.001134 acc_pose: 0.627119 loss: 0.001134 2022/09/26 10:43:41 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 10:44:13 - mmengine - INFO - Epoch(train) [3][50/293] lr: 5.000000e-04 eta: 9:31:57 time: 0.646327 data_time: 0.108312 memory: 9504 loss_kpt: 0.001107 acc_pose: 0.569500 loss: 0.001107 2022/09/26 10:44:42 - mmengine - INFO - Epoch(train) [3][100/293] lr: 5.000000e-04 eta: 9:32:26 time: 0.576357 data_time: 0.098267 memory: 9504 loss_kpt: 0.001090 acc_pose: 0.694603 loss: 0.001090 2022/09/26 10:45:14 - mmengine - INFO - Epoch(train) [3][150/293] lr: 5.000000e-04 eta: 9:36:39 time: 0.632713 data_time: 0.177294 memory: 9504 loss_kpt: 0.001082 acc_pose: 0.650825 loss: 0.001082 2022/09/26 10:45:44 - mmengine - INFO - Epoch(train) [3][200/293] lr: 5.000000e-04 eta: 9:38:26 time: 0.604109 data_time: 0.098773 memory: 9504 loss_kpt: 0.001060 acc_pose: 0.689841 loss: 0.001060 2022/09/26 10:46:14 - mmengine - INFO - Epoch(train) [3][250/293] lr: 5.000000e-04 eta: 9:39:36 time: 0.598690 data_time: 0.104505 memory: 9504 loss_kpt: 0.001072 acc_pose: 0.681280 loss: 0.001072 2022/09/26 10:46:39 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 10:47:10 - mmengine - INFO - Epoch(train) [4][50/293] lr: 5.000000e-04 eta: 9:14:13 time: 0.615122 data_time: 0.108450 memory: 9504 loss_kpt: 0.001023 acc_pose: 0.641309 loss: 0.001023 2022/09/26 10:47:39 - mmengine - INFO - Epoch(train) [4][100/293] lr: 5.000000e-04 eta: 9:15:16 time: 0.577817 data_time: 0.090910 memory: 9504 loss_kpt: 0.001010 acc_pose: 0.699407 loss: 0.001010 2022/09/26 10:47:51 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 10:48:07 - mmengine - INFO - Epoch(train) [4][150/293] lr: 5.000000e-04 eta: 9:15:36 time: 0.566480 data_time: 0.127230 memory: 9504 loss_kpt: 0.000990 acc_pose: 0.708103 loss: 0.000990 2022/09/26 10:48:36 - mmengine - INFO - Epoch(train) [4][200/293] lr: 5.000000e-04 eta: 9:16:32 time: 0.580878 data_time: 0.217268 memory: 9504 loss_kpt: 0.000998 acc_pose: 0.692059 loss: 0.000998 2022/09/26 10:49:05 - mmengine - INFO - Epoch(train) [4][250/293] lr: 5.000000e-04 eta: 9:17:44 time: 0.589436 data_time: 0.124167 memory: 9504 loss_kpt: 0.001009 acc_pose: 0.663458 loss: 0.001009 2022/09/26 10:49:32 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 10:50:02 - mmengine - INFO - Epoch(train) [5][50/293] lr: 5.000000e-04 eta: 8:59:23 time: 0.605380 data_time: 0.142473 memory: 9504 loss_kpt: 0.000990 acc_pose: 0.685872 loss: 0.000990 2022/09/26 10:50:32 - mmengine - INFO - Epoch(train) [5][100/293] lr: 5.000000e-04 eta: 9:01:21 time: 0.597702 data_time: 0.106074 memory: 9504 loss_kpt: 0.000999 acc_pose: 0.733084 loss: 0.000999 2022/09/26 10:51:00 - mmengine - INFO - Epoch(train) [5][150/293] lr: 5.000000e-04 eta: 9:01:47 time: 0.562140 data_time: 0.160726 memory: 9504 loss_kpt: 0.000992 acc_pose: 0.680316 loss: 0.000992 2022/09/26 10:51:29 - mmengine - INFO - Epoch(train) [5][200/293] lr: 5.000000e-04 eta: 9:02:46 time: 0.579128 data_time: 0.114940 memory: 9504 loss_kpt: 0.000982 acc_pose: 0.653903 loss: 0.000982 2022/09/26 10:51:58 - mmengine - INFO - Epoch(train) [5][250/293] lr: 5.000000e-04 eta: 9:03:36 time: 0.578163 data_time: 0.087240 memory: 9504 loss_kpt: 0.000974 acc_pose: 0.653554 loss: 0.000974 2022/09/26 10:52:28 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 10:52:59 - mmengine - INFO - Epoch(train) [6][50/293] lr: 5.000000e-04 eta: 8:49:57 time: 0.621126 data_time: 0.121850 memory: 9504 loss_kpt: 0.000961 acc_pose: 0.732643 loss: 0.000961 2022/09/26 10:53:29 - mmengine - INFO - Epoch(train) [6][100/293] lr: 5.000000e-04 eta: 8:51:46 time: 0.600595 data_time: 0.129891 memory: 9504 loss_kpt: 0.000958 acc_pose: 0.666151 loss: 0.000958 2022/09/26 10:53:58 - mmengine - INFO - Epoch(train) [6][150/293] lr: 5.000000e-04 eta: 8:52:37 time: 0.573737 data_time: 0.086707 memory: 9504 loss_kpt: 0.000944 acc_pose: 0.685127 loss: 0.000944 2022/09/26 10:54:28 - mmengine - INFO - Epoch(train) [6][200/293] lr: 5.000000e-04 eta: 8:54:00 time: 0.594654 data_time: 0.071357 memory: 9504 loss_kpt: 0.000940 acc_pose: 0.715980 loss: 0.000940 2022/09/26 10:54:57 - mmengine - INFO - Epoch(train) [6][250/293] lr: 5.000000e-04 eta: 8:54:58 time: 0.583816 data_time: 0.117079 memory: 9504 loss_kpt: 0.000938 acc_pose: 0.695890 loss: 0.000938 2022/09/26 10:55:21 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 10:55:52 - mmengine - INFO - Epoch(train) [7][50/293] lr: 5.000000e-04 eta: 8:43:38 time: 0.616737 data_time: 0.090568 memory: 9504 loss_kpt: 0.000937 acc_pose: 0.715058 loss: 0.000937 2022/09/26 10:56:22 - mmengine - INFO - Epoch(train) [7][100/293] lr: 5.000000e-04 eta: 8:45:22 time: 0.606896 data_time: 0.123973 memory: 9504 loss_kpt: 0.000940 acc_pose: 0.634057 loss: 0.000940 2022/09/26 10:56:53 - mmengine - INFO - Epoch(train) [7][150/293] lr: 5.000000e-04 eta: 8:47:05 time: 0.611043 data_time: 0.099708 memory: 9504 loss_kpt: 0.000933 acc_pose: 0.658213 loss: 0.000933 2022/09/26 10:57:23 - mmengine - INFO - Epoch(train) [7][200/293] lr: 5.000000e-04 eta: 8:48:23 time: 0.599254 data_time: 0.086790 memory: 9504 loss_kpt: 0.000915 acc_pose: 0.732316 loss: 0.000915 2022/09/26 10:57:48 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 10:57:53 - mmengine - INFO - Epoch(train) [7][250/293] lr: 5.000000e-04 eta: 8:49:36 time: 0.599297 data_time: 0.124376 memory: 9504 loss_kpt: 0.000916 acc_pose: 0.754057 loss: 0.000916 2022/09/26 10:58:19 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 10:58:50 - mmengine - INFO - Epoch(train) [8][50/293] lr: 5.000000e-04 eta: 8:40:05 time: 0.624682 data_time: 0.161736 memory: 9504 loss_kpt: 0.000934 acc_pose: 0.745471 loss: 0.000934 2022/09/26 10:59:20 - mmengine - INFO - Epoch(train) [8][100/293] lr: 5.000000e-04 eta: 8:41:19 time: 0.597568 data_time: 0.141270 memory: 9504 loss_kpt: 0.000926 acc_pose: 0.682323 loss: 0.000926 2022/09/26 10:59:49 - mmengine - INFO - Epoch(train) [8][150/293] lr: 5.000000e-04 eta: 8:42:07 time: 0.582142 data_time: 0.081702 memory: 9504 loss_kpt: 0.000898 acc_pose: 0.738439 loss: 0.000898 2022/09/26 11:00:19 - mmengine - INFO - Epoch(train) [8][200/293] lr: 5.000000e-04 eta: 8:43:21 time: 0.603911 data_time: 0.101648 memory: 9504 loss_kpt: 0.000924 acc_pose: 0.731842 loss: 0.000924 2022/09/26 11:00:49 - mmengine - INFO - Epoch(train) [8][250/293] lr: 5.000000e-04 eta: 8:44:25 time: 0.599854 data_time: 0.113829 memory: 9504 loss_kpt: 0.000899 acc_pose: 0.759959 loss: 0.000899 2022/09/26 11:01:16 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:01:47 - mmengine - INFO - Epoch(train) [9][50/293] lr: 5.000000e-04 eta: 8:35:50 time: 0.611700 data_time: 0.229842 memory: 9504 loss_kpt: 0.000891 acc_pose: 0.709193 loss: 0.000891 2022/09/26 11:02:17 - mmengine - INFO - Epoch(train) [9][100/293] lr: 5.000000e-04 eta: 8:36:55 time: 0.598776 data_time: 0.108901 memory: 9504 loss_kpt: 0.000909 acc_pose: 0.696283 loss: 0.000909 2022/09/26 11:02:47 - mmengine - INFO - Epoch(train) [9][150/293] lr: 5.000000e-04 eta: 8:37:58 time: 0.599693 data_time: 0.079190 memory: 9504 loss_kpt: 0.000891 acc_pose: 0.782617 loss: 0.000891 2022/09/26 11:03:17 - mmengine - INFO - Epoch(train) [9][200/293] lr: 5.000000e-04 eta: 8:38:55 time: 0.599082 data_time: 0.096252 memory: 9504 loss_kpt: 0.000904 acc_pose: 0.674965 loss: 0.000904 2022/09/26 11:05:29 - mmengine - INFO - Epoch(train) [9][250/293] lr: 5.000000e-04 eta: 9:18:31 time: 2.641795 data_time: 2.075252 memory: 9504 loss_kpt: 0.000884 acc_pose: 0.732320 loss: 0.000884 2022/09/26 11:05:55 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:06:25 - mmengine - INFO - Epoch(train) [10][50/293] lr: 5.000000e-04 eta: 9:09:19 time: 0.602170 data_time: 0.259026 memory: 9504 loss_kpt: 0.000887 acc_pose: 0.678104 loss: 0.000887 2022/09/26 11:06:56 - mmengine - INFO - Epoch(train) [10][100/293] lr: 5.000000e-04 eta: 9:09:55 time: 0.619879 data_time: 0.126040 memory: 9504 loss_kpt: 0.000884 acc_pose: 0.748030 loss: 0.000884 2022/09/26 11:07:26 - mmengine - INFO - Epoch(train) [10][150/293] lr: 5.000000e-04 eta: 9:10:11 time: 0.602919 data_time: 0.128632 memory: 9504 loss_kpt: 0.000885 acc_pose: 0.690970 loss: 0.000885 2022/09/26 11:07:55 - mmengine - INFO - Epoch(train) [10][200/293] lr: 5.000000e-04 eta: 9:10:11 time: 0.589442 data_time: 0.107038 memory: 9504 loss_kpt: 0.000888 acc_pose: 0.753633 loss: 0.000888 2022/09/26 11:08:26 - mmengine - INFO - Epoch(train) [10][250/293] lr: 5.000000e-04 eta: 9:10:40 time: 0.618174 data_time: 0.104717 memory: 9504 loss_kpt: 0.000884 acc_pose: 0.726155 loss: 0.000884 2022/09/26 11:08:51 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:08:51 - mmengine - INFO - Saving checkpoint at 10 epochs 2022/09/26 11:09:32 - mmengine - INFO - Epoch(val) [10][50/407] eta: 0:03:55 time: 0.660465 data_time: 0.420972 memory: 9504 2022/09/26 11:09:47 - mmengine - INFO - Epoch(val) [10][100/407] eta: 0:01:34 time: 0.307053 data_time: 0.136448 memory: 1378 2022/09/26 11:10:04 - mmengine - INFO - Epoch(val) [10][150/407] eta: 0:01:25 time: 0.333297 data_time: 0.134091 memory: 1378 2022/09/26 11:10:18 - mmengine - INFO - Epoch(val) [10][200/407] eta: 0:01:01 time: 0.295182 data_time: 0.115274 memory: 1378 2022/09/26 11:10:35 - mmengine - INFO - Epoch(val) [10][250/407] eta: 0:00:51 time: 0.328417 data_time: 0.157773 memory: 1378 2022/09/26 11:10:51 - mmengine - INFO - Epoch(val) [10][300/407] eta: 0:00:34 time: 0.319252 data_time: 0.136075 memory: 1378 2022/09/26 11:11:07 - mmengine - INFO - Epoch(val) [10][350/407] eta: 0:00:18 time: 0.324740 data_time: 0.148550 memory: 1378 2022/09/26 11:11:23 - mmengine - INFO - Epoch(val) [10][400/407] eta: 0:00:02 time: 0.313141 data_time: 0.132977 memory: 1378 2022/09/26 11:11:58 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 11:12:12 - mmengine - INFO - Epoch(val) [10][407/407] coco/AP: 0.614756 coco/AP .5: 0.859126 coco/AP .75: 0.686036 coco/AP (M): 0.586349 coco/AP (L): 0.669822 coco/AR: 0.682667 coco/AR .5: 0.905542 coco/AR .75: 0.751417 coco/AR (M): 0.643294 coco/AR (L): 0.738127 2022/09/26 11:12:16 - mmengine - INFO - The best checkpoint with 0.6148 coco/AP at 10 epoch is saved to best_coco/AP_epoch_10.pth. 2022/09/26 11:12:45 - mmengine - INFO - Epoch(train) [11][50/293] lr: 5.000000e-04 eta: 9:02:16 time: 0.588441 data_time: 0.206788 memory: 9504 loss_kpt: 0.000887 acc_pose: 0.662594 loss: 0.000887 2022/09/26 11:13:01 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:13:20 - mmengine - INFO - Epoch(train) [11][100/293] lr: 5.000000e-04 eta: 9:03:55 time: 0.687162 data_time: 0.115656 memory: 9504 loss_kpt: 0.000877 acc_pose: 0.674728 loss: 0.000877 2022/09/26 11:13:49 - mmengine - INFO - Epoch(train) [11][150/293] lr: 5.000000e-04 eta: 9:04:06 time: 0.598760 data_time: 0.117063 memory: 9504 loss_kpt: 0.000882 acc_pose: 0.758708 loss: 0.000882 2022/09/26 11:14:18 - mmengine - INFO - Epoch(train) [11][200/293] lr: 5.000000e-04 eta: 9:03:57 time: 0.579243 data_time: 0.088120 memory: 9504 loss_kpt: 0.000878 acc_pose: 0.738983 loss: 0.000878 2022/09/26 11:14:49 - mmengine - INFO - Epoch(train) [11][250/293] lr: 5.000000e-04 eta: 9:04:17 time: 0.611115 data_time: 0.126738 memory: 9504 loss_kpt: 0.000852 acc_pose: 0.755321 loss: 0.000852 2022/09/26 11:15:15 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:15:45 - mmengine - INFO - Epoch(train) [12][50/293] lr: 5.000000e-04 eta: 8:56:51 time: 0.597437 data_time: 0.130089 memory: 9504 loss_kpt: 0.000871 acc_pose: 0.757432 loss: 0.000871 2022/09/26 11:16:15 - mmengine - INFO - Epoch(train) [12][100/293] lr: 5.000000e-04 eta: 8:57:00 time: 0.594921 data_time: 0.087866 memory: 9504 loss_kpt: 0.000884 acc_pose: 0.735549 loss: 0.000884 2022/09/26 11:16:44 - mmengine - INFO - Epoch(train) [12][150/293] lr: 5.000000e-04 eta: 8:57:04 time: 0.590524 data_time: 0.087375 memory: 9504 loss_kpt: 0.000865 acc_pose: 0.789336 loss: 0.000865 2022/09/26 11:17:14 - mmengine - INFO - Epoch(train) [12][200/293] lr: 5.000000e-04 eta: 8:57:01 time: 0.583818 data_time: 0.193279 memory: 9504 loss_kpt: 0.000860 acc_pose: 0.776958 loss: 0.000860 2022/09/26 11:17:43 - mmengine - INFO - Epoch(train) [12][250/293] lr: 5.000000e-04 eta: 8:57:04 time: 0.590722 data_time: 0.179942 memory: 9504 loss_kpt: 0.000866 acc_pose: 0.734830 loss: 0.000866 2022/09/26 11:18:09 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:18:40 - mmengine - INFO - Epoch(train) [13][50/293] lr: 5.000000e-04 eta: 8:50:36 time: 0.618539 data_time: 0.172980 memory: 9504 loss_kpt: 0.000854 acc_pose: 0.715723 loss: 0.000854 2022/09/26 11:19:10 - mmengine - INFO - Epoch(train) [13][100/293] lr: 5.000000e-04 eta: 8:50:50 time: 0.601447 data_time: 0.217830 memory: 9504 loss_kpt: 0.000872 acc_pose: 0.712684 loss: 0.000872 2022/09/26 11:19:39 - mmengine - INFO - Epoch(train) [13][150/293] lr: 5.000000e-04 eta: 8:50:43 time: 0.575379 data_time: 0.157199 memory: 9504 loss_kpt: 0.000861 acc_pose: 0.742327 loss: 0.000861 2022/09/26 11:20:08 - mmengine - INFO - Epoch(train) [13][200/293] lr: 5.000000e-04 eta: 8:50:35 time: 0.576114 data_time: 0.118162 memory: 9504 loss_kpt: 0.000841 acc_pose: 0.779065 loss: 0.000841 2022/09/26 11:20:39 - mmengine - INFO - Epoch(train) [13][250/293] lr: 5.000000e-04 eta: 8:51:05 time: 0.624880 data_time: 0.198241 memory: 9504 loss_kpt: 0.000848 acc_pose: 0.688259 loss: 0.000848 2022/09/26 11:21:04 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:21:37 - mmengine - INFO - Epoch(train) [14][50/293] lr: 5.000000e-04 eta: 8:45:30 time: 0.646553 data_time: 0.103908 memory: 9504 loss_kpt: 0.000834 acc_pose: 0.768606 loss: 0.000834 2022/09/26 11:22:07 - mmengine - INFO - Epoch(train) [14][100/293] lr: 5.000000e-04 eta: 8:45:45 time: 0.604880 data_time: 0.087548 memory: 9504 loss_kpt: 0.000847 acc_pose: 0.783544 loss: 0.000847 2022/09/26 11:22:37 - mmengine - INFO - Epoch(train) [14][150/293] lr: 5.000000e-04 eta: 8:46:01 time: 0.606891 data_time: 0.084348 memory: 9504 loss_kpt: 0.000847 acc_pose: 0.732529 loss: 0.000847 2022/09/26 11:23:01 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:23:07 - mmengine - INFO - Epoch(train) [14][200/293] lr: 5.000000e-04 eta: 8:46:03 time: 0.588521 data_time: 0.096896 memory: 9504 loss_kpt: 0.000859 acc_pose: 0.741967 loss: 0.000859 2022/09/26 11:23:37 - mmengine - INFO - Epoch(train) [14][250/293] lr: 5.000000e-04 eta: 8:46:14 time: 0.603874 data_time: 0.089944 memory: 9504 loss_kpt: 0.000849 acc_pose: 0.777669 loss: 0.000849 2022/09/26 11:24:03 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:24:32 - mmengine - INFO - Epoch(train) [15][50/293] lr: 5.000000e-04 eta: 8:40:18 time: 0.579348 data_time: 0.095125 memory: 9504 loss_kpt: 0.000848 acc_pose: 0.750654 loss: 0.000848 2022/09/26 11:25:02 - mmengine - INFO - Epoch(train) [15][100/293] lr: 5.000000e-04 eta: 8:40:31 time: 0.603505 data_time: 0.116601 memory: 9504 loss_kpt: 0.000840 acc_pose: 0.773871 loss: 0.000840 2022/09/26 11:25:33 - mmengine - INFO - Epoch(train) [15][150/293] lr: 5.000000e-04 eta: 8:40:57 time: 0.623491 data_time: 0.188143 memory: 9504 loss_kpt: 0.000841 acc_pose: 0.733180 loss: 0.000841 2022/09/26 11:26:03 - mmengine - INFO - Epoch(train) [15][200/293] lr: 5.000000e-04 eta: 8:41:02 time: 0.594800 data_time: 0.250448 memory: 9504 loss_kpt: 0.000827 acc_pose: 0.786273 loss: 0.000827 2022/09/26 11:26:34 - mmengine - INFO - Epoch(train) [15][250/293] lr: 5.000000e-04 eta: 8:41:22 time: 0.618822 data_time: 0.104197 memory: 9504 loss_kpt: 0.000837 acc_pose: 0.769325 loss: 0.000837 2022/09/26 11:26:59 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:27:31 - mmengine - INFO - Epoch(train) [16][50/293] lr: 5.000000e-04 eta: 8:36:29 time: 0.640662 data_time: 0.208338 memory: 9504 loss_kpt: 0.000833 acc_pose: 0.764776 loss: 0.000833 2022/09/26 11:28:00 - mmengine - INFO - Epoch(train) [16][100/293] lr: 5.000000e-04 eta: 8:36:30 time: 0.586128 data_time: 0.198283 memory: 9504 loss_kpt: 0.000831 acc_pose: 0.707914 loss: 0.000831 2022/09/26 11:28:30 - mmengine - INFO - Epoch(train) [16][150/293] lr: 5.000000e-04 eta: 8:36:35 time: 0.594692 data_time: 0.107137 memory: 9504 loss_kpt: 0.000842 acc_pose: 0.734426 loss: 0.000842 2022/09/26 11:28:59 - mmengine - INFO - Epoch(train) [16][200/293] lr: 5.000000e-04 eta: 8:36:35 time: 0.588953 data_time: 0.087606 memory: 9504 loss_kpt: 0.000825 acc_pose: 0.746933 loss: 0.000825 2022/09/26 11:29:29 - mmengine - INFO - Epoch(train) [16][250/293] lr: 5.000000e-04 eta: 8:36:44 time: 0.602407 data_time: 0.117901 memory: 9504 loss_kpt: 0.000832 acc_pose: 0.701007 loss: 0.000832 2022/09/26 11:29:54 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:30:24 - mmengine - INFO - Epoch(train) [17][50/293] lr: 5.000000e-04 eta: 8:31:38 time: 0.589040 data_time: 0.113305 memory: 9504 loss_kpt: 0.000816 acc_pose: 0.795737 loss: 0.000816 2022/09/26 11:30:54 - mmengine - INFO - Epoch(train) [17][100/293] lr: 5.000000e-04 eta: 8:31:44 time: 0.595926 data_time: 0.091489 memory: 9504 loss_kpt: 0.000835 acc_pose: 0.717447 loss: 0.000835 2022/09/26 11:31:27 - mmengine - INFO - Epoch(train) [17][150/293] lr: 5.000000e-04 eta: 8:32:38 time: 0.678962 data_time: 0.177834 memory: 9504 loss_kpt: 0.000815 acc_pose: 0.756606 loss: 0.000815 2022/09/26 11:31:59 - mmengine - INFO - Epoch(train) [17][200/293] lr: 5.000000e-04 eta: 8:32:58 time: 0.624368 data_time: 0.107530 memory: 9504 loss_kpt: 0.000837 acc_pose: 0.751625 loss: 0.000837 2022/09/26 11:32:29 - mmengine - INFO - Epoch(train) [17][250/293] lr: 5.000000e-04 eta: 8:33:09 time: 0.609370 data_time: 0.123022 memory: 9504 loss_kpt: 0.000813 acc_pose: 0.736382 loss: 0.000813 2022/09/26 11:32:54 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:33:07 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:33:28 - mmengine - INFO - Epoch(train) [18][50/293] lr: 5.000000e-04 eta: 8:29:04 time: 0.666181 data_time: 0.105662 memory: 9504 loss_kpt: 0.000819 acc_pose: 0.755969 loss: 0.000819 2022/09/26 11:33:59 - mmengine - INFO - Epoch(train) [18][100/293] lr: 5.000000e-04 eta: 8:29:20 time: 0.617911 data_time: 0.156321 memory: 9504 loss_kpt: 0.000808 acc_pose: 0.852601 loss: 0.000808 2022/09/26 11:34:28 - mmengine - INFO - Epoch(train) [18][150/293] lr: 5.000000e-04 eta: 8:29:15 time: 0.581955 data_time: 0.123786 memory: 9504 loss_kpt: 0.000808 acc_pose: 0.742134 loss: 0.000808 2022/09/26 11:34:57 - mmengine - INFO - Epoch(train) [18][200/293] lr: 5.000000e-04 eta: 8:29:17 time: 0.594669 data_time: 0.078102 memory: 9504 loss_kpt: 0.000826 acc_pose: 0.728284 loss: 0.000826 2022/09/26 11:35:29 - mmengine - INFO - Epoch(train) [18][250/293] lr: 5.000000e-04 eta: 8:29:39 time: 0.634580 data_time: 0.113930 memory: 9504 loss_kpt: 0.000827 acc_pose: 0.769412 loss: 0.000827 2022/09/26 11:35:56 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:36:28 - mmengine - INFO - Epoch(train) [19][50/293] lr: 5.000000e-04 eta: 8:25:33 time: 0.639754 data_time: 0.120460 memory: 9504 loss_kpt: 0.000817 acc_pose: 0.739386 loss: 0.000817 2022/09/26 11:36:57 - mmengine - INFO - Epoch(train) [19][100/293] lr: 5.000000e-04 eta: 8:25:35 time: 0.594049 data_time: 0.084454 memory: 9504 loss_kpt: 0.000810 acc_pose: 0.808124 loss: 0.000810 2022/09/26 11:37:27 - mmengine - INFO - Epoch(train) [19][150/293] lr: 5.000000e-04 eta: 8:25:38 time: 0.599084 data_time: 0.081852 memory: 9504 loss_kpt: 0.000803 acc_pose: 0.821377 loss: 0.000803 2022/09/26 11:37:57 - mmengine - INFO - Epoch(train) [19][200/293] lr: 5.000000e-04 eta: 8:25:44 time: 0.604353 data_time: 0.082504 memory: 9504 loss_kpt: 0.000814 acc_pose: 0.753278 loss: 0.000814 2022/09/26 11:38:28 - mmengine - INFO - Epoch(train) [19][250/293] lr: 5.000000e-04 eta: 8:25:50 time: 0.606828 data_time: 0.071879 memory: 9504 loss_kpt: 0.000816 acc_pose: 0.752240 loss: 0.000816 2022/09/26 11:38:53 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:39:25 - mmengine - INFO - Epoch(train) [20][50/293] lr: 5.000000e-04 eta: 8:21:52 time: 0.630528 data_time: 0.096403 memory: 9504 loss_kpt: 0.000808 acc_pose: 0.767263 loss: 0.000808 2022/09/26 11:39:54 - mmengine - INFO - Epoch(train) [20][100/293] lr: 5.000000e-04 eta: 8:21:54 time: 0.598834 data_time: 0.150171 memory: 9504 loss_kpt: 0.000802 acc_pose: 0.765596 loss: 0.000802 2022/09/26 11:40:24 - mmengine - INFO - Epoch(train) [20][150/293] lr: 5.000000e-04 eta: 8:21:56 time: 0.598142 data_time: 0.084872 memory: 9504 loss_kpt: 0.000810 acc_pose: 0.793725 loss: 0.000810 2022/09/26 11:40:55 - mmengine - INFO - Epoch(train) [20][200/293] lr: 5.000000e-04 eta: 8:22:09 time: 0.621352 data_time: 0.100800 memory: 9504 loss_kpt: 0.000817 acc_pose: 0.749917 loss: 0.000817 2022/09/26 11:41:25 - mmengine - INFO - Epoch(train) [20][250/293] lr: 5.000000e-04 eta: 8:22:08 time: 0.594582 data_time: 0.075521 memory: 9504 loss_kpt: 0.000814 acc_pose: 0.704991 loss: 0.000814 2022/09/26 11:41:51 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:41:51 - mmengine - INFO - Saving checkpoint at 20 epochs 2022/09/26 11:42:17 - mmengine - INFO - Epoch(val) [20][50/407] eta: 0:02:02 time: 0.342186 data_time: 0.154744 memory: 9504 2022/09/26 11:42:32 - mmengine - INFO - Epoch(val) [20][100/407] eta: 0:01:35 time: 0.309951 data_time: 0.139688 memory: 1378 2022/09/26 11:42:48 - mmengine - INFO - Epoch(val) [20][150/407] eta: 0:01:24 time: 0.328674 data_time: 0.160473 memory: 1378 2022/09/26 11:43:04 - mmengine - INFO - Epoch(val) [20][200/407] eta: 0:01:05 time: 0.315858 data_time: 0.140054 memory: 1378 2022/09/26 11:43:20 - mmengine - INFO - Epoch(val) [20][250/407] eta: 0:00:49 time: 0.314757 data_time: 0.140065 memory: 1378 2022/09/26 11:43:36 - mmengine - INFO - Epoch(val) [20][300/407] eta: 0:00:34 time: 0.320097 data_time: 0.135948 memory: 1378 2022/09/26 11:43:52 - mmengine - INFO - Epoch(val) [20][350/407] eta: 0:00:18 time: 0.325323 data_time: 0.167511 memory: 1378 2022/09/26 11:44:03 - mmengine - INFO - Epoch(val) [20][400/407] eta: 0:00:01 time: 0.219048 data_time: 0.113482 memory: 1378 2022/09/26 11:44:37 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 11:44:50 - mmengine - INFO - Epoch(val) [20][407/407] coco/AP: 0.646337 coco/AP .5: 0.869170 coco/AP .75: 0.723402 coco/AP (M): 0.616143 coco/AP (L): 0.707303 coco/AR: 0.710438 coco/AR .5: 0.913413 coco/AR .75: 0.780699 coco/AR (M): 0.669517 coco/AR (L): 0.769305 2022/09/26 11:44:50 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_10.pth is removed 2022/09/26 11:44:54 - mmengine - INFO - The best checkpoint with 0.6463 coco/AP at 20 epoch is saved to best_coco/AP_epoch_20.pth. 2022/09/26 11:45:23 - mmengine - INFO - Epoch(train) [21][50/293] lr: 5.000000e-04 eta: 8:17:59 time: 0.583409 data_time: 0.185935 memory: 9504 loss_kpt: 0.000804 acc_pose: 0.753677 loss: 0.000804 2022/09/26 11:45:51 - mmengine - INFO - Epoch(train) [21][100/293] lr: 5.000000e-04 eta: 8:17:47 time: 0.570588 data_time: 0.076774 memory: 9504 loss_kpt: 0.000804 acc_pose: 0.818684 loss: 0.000804 2022/09/26 11:46:16 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:46:23 - mmengine - INFO - Epoch(train) [21][150/293] lr: 5.000000e-04 eta: 8:18:00 time: 0.622687 data_time: 0.120389 memory: 9504 loss_kpt: 0.000812 acc_pose: 0.699325 loss: 0.000812 2022/09/26 11:46:53 - mmengine - INFO - Epoch(train) [21][200/293] lr: 5.000000e-04 eta: 8:18:10 time: 0.618387 data_time: 0.074804 memory: 9504 loss_kpt: 0.000802 acc_pose: 0.763361 loss: 0.000802 2022/09/26 11:47:23 - mmengine - INFO - Epoch(train) [21][250/293] lr: 5.000000e-04 eta: 8:18:05 time: 0.588465 data_time: 0.075998 memory: 9504 loss_kpt: 0.000816 acc_pose: 0.791877 loss: 0.000816 2022/09/26 11:47:48 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:48:19 - mmengine - INFO - Epoch(train) [22][50/293] lr: 5.000000e-04 eta: 8:14:25 time: 0.620744 data_time: 0.172089 memory: 9504 loss_kpt: 0.000806 acc_pose: 0.807813 loss: 0.000806 2022/09/26 11:48:48 - mmengine - INFO - Epoch(train) [22][100/293] lr: 5.000000e-04 eta: 8:14:21 time: 0.588139 data_time: 0.082282 memory: 9504 loss_kpt: 0.000805 acc_pose: 0.727571 loss: 0.000805 2022/09/26 11:49:19 - mmengine - INFO - Epoch(train) [22][150/293] lr: 5.000000e-04 eta: 8:14:26 time: 0.610738 data_time: 0.088266 memory: 9504 loss_kpt: 0.000808 acc_pose: 0.797822 loss: 0.000808 2022/09/26 11:49:48 - mmengine - INFO - Epoch(train) [22][200/293] lr: 5.000000e-04 eta: 8:14:16 time: 0.575127 data_time: 0.107111 memory: 9504 loss_kpt: 0.000776 acc_pose: 0.797789 loss: 0.000776 2022/09/26 11:50:18 - mmengine - INFO - Epoch(train) [22][250/293] lr: 5.000000e-04 eta: 8:14:16 time: 0.599548 data_time: 0.179733 memory: 9504 loss_kpt: 0.000787 acc_pose: 0.825310 loss: 0.000787 2022/09/26 11:50:44 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:51:16 - mmengine - INFO - Epoch(train) [23][50/293] lr: 5.000000e-04 eta: 8:10:56 time: 0.646161 data_time: 0.138325 memory: 9504 loss_kpt: 0.000785 acc_pose: 0.759103 loss: 0.000785 2022/09/26 11:51:46 - mmengine - INFO - Epoch(train) [23][100/293] lr: 5.000000e-04 eta: 8:10:57 time: 0.602587 data_time: 0.125869 memory: 9504 loss_kpt: 0.000786 acc_pose: 0.755317 loss: 0.000786 2022/09/26 11:52:17 - mmengine - INFO - Epoch(train) [23][150/293] lr: 5.000000e-04 eta: 8:11:06 time: 0.621892 data_time: 0.087243 memory: 9504 loss_kpt: 0.000781 acc_pose: 0.703069 loss: 0.000781 2022/09/26 11:52:47 - mmengine - INFO - Epoch(train) [23][200/293] lr: 5.000000e-04 eta: 8:11:06 time: 0.600144 data_time: 0.088885 memory: 9504 loss_kpt: 0.000802 acc_pose: 0.705691 loss: 0.000802 2022/09/26 11:53:15 - mmengine - INFO - Epoch(train) [23][250/293] lr: 5.000000e-04 eta: 8:10:44 time: 0.548381 data_time: 0.073791 memory: 9504 loss_kpt: 0.000776 acc_pose: 0.695845 loss: 0.000776 2022/09/26 11:53:39 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:54:11 - mmengine - INFO - Epoch(train) [24][50/293] lr: 5.000000e-04 eta: 8:07:28 time: 0.637221 data_time: 0.128550 memory: 9504 loss_kpt: 0.000770 acc_pose: 0.763545 loss: 0.000770 2022/09/26 11:54:42 - mmengine - INFO - Epoch(train) [24][100/293] lr: 5.000000e-04 eta: 8:07:38 time: 0.625356 data_time: 0.091578 memory: 9504 loss_kpt: 0.000789 acc_pose: 0.770087 loss: 0.000789 2022/09/26 11:55:14 - mmengine - INFO - Epoch(train) [24][150/293] lr: 5.000000e-04 eta: 8:07:50 time: 0.634332 data_time: 0.106972 memory: 9504 loss_kpt: 0.000791 acc_pose: 0.769411 loss: 0.000791 2022/09/26 11:55:52 - mmengine - INFO - Epoch(train) [24][200/293] lr: 5.000000e-04 eta: 8:08:58 time: 0.774850 data_time: 0.231698 memory: 9504 loss_kpt: 0.000782 acc_pose: 0.761785 loss: 0.000782 2022/09/26 11:56:28 - mmengine - INFO - Epoch(train) [24][250/293] lr: 5.000000e-04 eta: 8:09:41 time: 0.717499 data_time: 0.218573 memory: 9504 loss_kpt: 0.000793 acc_pose: 0.782765 loss: 0.000793 2022/09/26 11:56:35 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:56:55 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 11:57:24 - mmengine - INFO - Epoch(train) [25][50/293] lr: 5.000000e-04 eta: 8:06:12 time: 0.589643 data_time: 0.101875 memory: 9504 loss_kpt: 0.000775 acc_pose: 0.744663 loss: 0.000775 2022/09/26 11:57:54 - mmengine - INFO - Epoch(train) [25][100/293] lr: 5.000000e-04 eta: 8:06:05 time: 0.585945 data_time: 0.088363 memory: 9504 loss_kpt: 0.000772 acc_pose: 0.772411 loss: 0.000772 2022/09/26 11:58:23 - mmengine - INFO - Epoch(train) [25][150/293] lr: 5.000000e-04 eta: 8:05:58 time: 0.590242 data_time: 0.085978 memory: 9504 loss_kpt: 0.000800 acc_pose: 0.783350 loss: 0.000800 2022/09/26 11:58:52 - mmengine - INFO - Epoch(train) [25][200/293] lr: 5.000000e-04 eta: 8:05:47 time: 0.578766 data_time: 0.115308 memory: 9504 loss_kpt: 0.000784 acc_pose: 0.787103 loss: 0.000784 2022/09/26 11:59:21 - mmengine - INFO - Epoch(train) [25][250/293] lr: 5.000000e-04 eta: 8:05:39 time: 0.585693 data_time: 0.113320 memory: 9504 loss_kpt: 0.000781 acc_pose: 0.776796 loss: 0.000781 2022/09/26 11:59:46 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:00:18 - mmengine - INFO - Epoch(train) [26][50/293] lr: 5.000000e-04 eta: 8:02:36 time: 0.635929 data_time: 0.114624 memory: 9504 loss_kpt: 0.000777 acc_pose: 0.803780 loss: 0.000777 2022/09/26 12:00:46 - mmengine - INFO - Epoch(train) [26][100/293] lr: 5.000000e-04 eta: 8:02:23 time: 0.573701 data_time: 0.121619 memory: 9504 loss_kpt: 0.000785 acc_pose: 0.766240 loss: 0.000785 2022/09/26 12:01:18 - mmengine - INFO - Epoch(train) [26][150/293] lr: 5.000000e-04 eta: 8:02:29 time: 0.625604 data_time: 0.110122 memory: 9504 loss_kpt: 0.000788 acc_pose: 0.765278 loss: 0.000788 2022/09/26 12:01:47 - mmengine - INFO - Epoch(train) [26][200/293] lr: 5.000000e-04 eta: 8:02:18 time: 0.580145 data_time: 0.078270 memory: 9504 loss_kpt: 0.000776 acc_pose: 0.765888 loss: 0.000776 2022/09/26 12:02:15 - mmengine - INFO - Epoch(train) [26][250/293] lr: 5.000000e-04 eta: 8:02:04 time: 0.570043 data_time: 0.072186 memory: 9504 loss_kpt: 0.000778 acc_pose: 0.769159 loss: 0.000778 2022/09/26 12:02:41 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:03:12 - mmengine - INFO - Epoch(train) [27][50/293] lr: 5.000000e-04 eta: 7:59:03 time: 0.624828 data_time: 0.177743 memory: 9504 loss_kpt: 0.000780 acc_pose: 0.781943 loss: 0.000780 2022/09/26 12:03:41 - mmengine - INFO - Epoch(train) [27][100/293] lr: 5.000000e-04 eta: 7:58:52 time: 0.577713 data_time: 0.166351 memory: 9504 loss_kpt: 0.000786 acc_pose: 0.741941 loss: 0.000786 2022/09/26 12:04:11 - mmengine - INFO - Epoch(train) [27][150/293] lr: 5.000000e-04 eta: 7:58:48 time: 0.600740 data_time: 0.182926 memory: 9504 loss_kpt: 0.000773 acc_pose: 0.793214 loss: 0.000773 2022/09/26 12:04:41 - mmengine - INFO - Epoch(train) [27][200/293] lr: 5.000000e-04 eta: 7:58:40 time: 0.587560 data_time: 0.145623 memory: 9504 loss_kpt: 0.000790 acc_pose: 0.795137 loss: 0.000790 2022/09/26 12:05:09 - mmengine - INFO - Epoch(train) [27][250/293] lr: 5.000000e-04 eta: 7:58:25 time: 0.570609 data_time: 0.098284 memory: 9504 loss_kpt: 0.000769 acc_pose: 0.717016 loss: 0.000769 2022/09/26 12:05:33 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:06:03 - mmengine - INFO - Epoch(train) [28][50/293] lr: 5.000000e-04 eta: 7:55:21 time: 0.595151 data_time: 0.104529 memory: 9504 loss_kpt: 0.000756 acc_pose: 0.748032 loss: 0.000756 2022/09/26 12:06:25 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:06:31 - mmengine - INFO - Epoch(train) [28][100/293] lr: 5.000000e-04 eta: 7:55:08 time: 0.572830 data_time: 0.084861 memory: 9504 loss_kpt: 0.000768 acc_pose: 0.740827 loss: 0.000768 2022/09/26 12:07:00 - mmengine - INFO - Epoch(train) [28][150/293] lr: 5.000000e-04 eta: 7:54:55 time: 0.575597 data_time: 0.232254 memory: 9504 loss_kpt: 0.000775 acc_pose: 0.769796 loss: 0.000775 2022/09/26 12:07:29 - mmengine - INFO - Epoch(train) [28][200/293] lr: 5.000000e-04 eta: 7:54:43 time: 0.576838 data_time: 0.089959 memory: 9504 loss_kpt: 0.000750 acc_pose: 0.744473 loss: 0.000750 2022/09/26 12:07:58 - mmengine - INFO - Epoch(train) [28][250/293] lr: 5.000000e-04 eta: 7:54:34 time: 0.587299 data_time: 0.081606 memory: 9504 loss_kpt: 0.000794 acc_pose: 0.770176 loss: 0.000794 2022/09/26 12:08:22 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:08:53 - mmengine - INFO - Epoch(train) [29][50/293] lr: 5.000000e-04 eta: 7:51:43 time: 0.615141 data_time: 0.162886 memory: 9504 loss_kpt: 0.000766 acc_pose: 0.777017 loss: 0.000766 2022/09/26 12:09:23 - mmengine - INFO - Epoch(train) [29][100/293] lr: 5.000000e-04 eta: 7:51:36 time: 0.592617 data_time: 0.083771 memory: 9504 loss_kpt: 0.000763 acc_pose: 0.739784 loss: 0.000763 2022/09/26 12:09:52 - mmengine - INFO - Epoch(train) [29][150/293] lr: 5.000000e-04 eta: 7:51:25 time: 0.581652 data_time: 0.078654 memory: 9504 loss_kpt: 0.000766 acc_pose: 0.765285 loss: 0.000766 2022/09/26 12:10:21 - mmengine - INFO - Epoch(train) [29][200/293] lr: 5.000000e-04 eta: 7:51:18 time: 0.592545 data_time: 0.097911 memory: 9504 loss_kpt: 0.000763 acc_pose: 0.771273 loss: 0.000763 2022/09/26 12:10:51 - mmengine - INFO - Epoch(train) [29][250/293] lr: 5.000000e-04 eta: 7:51:12 time: 0.599517 data_time: 0.116064 memory: 9504 loss_kpt: 0.000782 acc_pose: 0.771201 loss: 0.000782 2022/09/26 12:11:15 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:11:45 - mmengine - INFO - Epoch(train) [30][50/293] lr: 5.000000e-04 eta: 7:48:23 time: 0.605684 data_time: 0.120518 memory: 9504 loss_kpt: 0.000781 acc_pose: 0.773072 loss: 0.000781 2022/09/26 12:12:15 - mmengine - INFO - Epoch(train) [30][100/293] lr: 5.000000e-04 eta: 7:48:16 time: 0.591376 data_time: 0.079482 memory: 9504 loss_kpt: 0.000783 acc_pose: 0.788710 loss: 0.000783 2022/09/26 12:12:44 - mmengine - INFO - Epoch(train) [30][150/293] lr: 5.000000e-04 eta: 7:48:06 time: 0.585926 data_time: 0.088872 memory: 9504 loss_kpt: 0.000761 acc_pose: 0.794801 loss: 0.000761 2022/09/26 12:13:13 - mmengine - INFO - Epoch(train) [30][200/293] lr: 5.000000e-04 eta: 7:47:56 time: 0.585412 data_time: 0.192767 memory: 9504 loss_kpt: 0.000763 acc_pose: 0.771819 loss: 0.000763 2022/09/26 12:13:43 - mmengine - INFO - Epoch(train) [30][250/293] lr: 5.000000e-04 eta: 7:47:50 time: 0.598587 data_time: 0.083879 memory: 9504 loss_kpt: 0.000755 acc_pose: 0.770839 loss: 0.000755 2022/09/26 12:14:08 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:14:08 - mmengine - INFO - Saving checkpoint at 30 epochs 2022/09/26 12:14:35 - mmengine - INFO - Epoch(val) [30][50/407] eta: 0:02:00 time: 0.337163 data_time: 0.146707 memory: 9504 2022/09/26 12:14:51 - mmengine - INFO - Epoch(val) [30][100/407] eta: 0:01:39 time: 0.324092 data_time: 0.137310 memory: 1378 2022/09/26 12:15:08 - mmengine - INFO - Epoch(val) [30][150/407] eta: 0:01:26 time: 0.338335 data_time: 0.150155 memory: 1378 2022/09/26 12:15:23 - mmengine - INFO - Epoch(val) [30][200/407] eta: 0:01:02 time: 0.301415 data_time: 0.135482 memory: 1378 2022/09/26 12:15:39 - mmengine - INFO - Epoch(val) [30][250/407] eta: 0:00:49 time: 0.313483 data_time: 0.134542 memory: 1378 2022/09/26 12:15:54 - mmengine - INFO - Epoch(val) [30][300/407] eta: 0:00:32 time: 0.306014 data_time: 0.125278 memory: 1378 2022/09/26 12:16:10 - mmengine - INFO - Epoch(val) [30][350/407] eta: 0:00:18 time: 0.321652 data_time: 0.149978 memory: 1378 2022/09/26 12:16:21 - mmengine - INFO - Epoch(val) [30][400/407] eta: 0:00:01 time: 0.212628 data_time: 0.106473 memory: 1378 2022/09/26 12:16:53 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 12:17:06 - mmengine - INFO - Epoch(val) [30][407/407] coco/AP: 0.668628 coco/AP .5: 0.880559 coco/AP .75: 0.747478 coco/AP (M): 0.634850 coco/AP (L): 0.731261 coco/AR: 0.728275 coco/AR .5: 0.920655 coco/AR .75: 0.799591 coco/AR (M): 0.686479 coco/AR (L): 0.788294 2022/09/26 12:17:06 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_20.pth is removed 2022/09/26 12:17:09 - mmengine - INFO - The best checkpoint with 0.6686 coco/AP at 30 epoch is saved to best_coco/AP_epoch_30.pth. 2022/09/26 12:17:38 - mmengine - INFO - Epoch(train) [31][50/293] lr: 5.000000e-04 eta: 7:44:58 time: 0.578951 data_time: 0.192561 memory: 9504 loss_kpt: 0.000756 acc_pose: 0.749263 loss: 0.000756 2022/09/26 12:18:08 - mmengine - INFO - Epoch(train) [31][100/293] lr: 5.000000e-04 eta: 7:44:48 time: 0.585650 data_time: 0.080986 memory: 9504 loss_kpt: 0.000768 acc_pose: 0.786091 loss: 0.000768 2022/09/26 12:18:37 - mmengine - INFO - Epoch(train) [31][150/293] lr: 5.000000e-04 eta: 7:44:39 time: 0.590479 data_time: 0.151300 memory: 9504 loss_kpt: 0.000758 acc_pose: 0.773599 loss: 0.000758 2022/09/26 12:19:09 - mmengine - INFO - Epoch(train) [31][200/293] lr: 5.000000e-04 eta: 7:44:42 time: 0.630233 data_time: 0.075468 memory: 9504 loss_kpt: 0.000749 acc_pose: 0.813168 loss: 0.000749 2022/09/26 12:19:14 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:19:37 - mmengine - INFO - Epoch(train) [31][250/293] lr: 5.000000e-04 eta: 7:44:28 time: 0.571563 data_time: 0.083497 memory: 9504 loss_kpt: 0.000762 acc_pose: 0.779075 loss: 0.000762 2022/09/26 12:20:03 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:20:34 - mmengine - INFO - Epoch(train) [32][50/293] lr: 5.000000e-04 eta: 7:41:53 time: 0.619365 data_time: 0.099824 memory: 9504 loss_kpt: 0.000763 acc_pose: 0.735288 loss: 0.000763 2022/09/26 12:21:02 - mmengine - INFO - Epoch(train) [32][100/293] lr: 5.000000e-04 eta: 7:41:37 time: 0.568004 data_time: 0.152559 memory: 9504 loss_kpt: 0.000751 acc_pose: 0.824004 loss: 0.000751 2022/09/26 12:21:33 - mmengine - INFO - Epoch(train) [32][150/293] lr: 5.000000e-04 eta: 7:41:34 time: 0.612347 data_time: 0.119495 memory: 9504 loss_kpt: 0.000741 acc_pose: 0.794367 loss: 0.000741 2022/09/26 12:22:03 - mmengine - INFO - Epoch(train) [32][200/293] lr: 5.000000e-04 eta: 7:41:26 time: 0.595283 data_time: 0.161527 memory: 9504 loss_kpt: 0.000763 acc_pose: 0.783066 loss: 0.000763 2022/09/26 12:22:32 - mmengine - INFO - Epoch(train) [32][250/293] lr: 5.000000e-04 eta: 7:41:15 time: 0.584192 data_time: 0.076192 memory: 9504 loss_kpt: 0.000751 acc_pose: 0.821724 loss: 0.000751 2022/09/26 12:22:56 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:23:26 - mmengine - INFO - Epoch(train) [33][50/293] lr: 5.000000e-04 eta: 7:38:39 time: 0.601003 data_time: 0.156341 memory: 9504 loss_kpt: 0.000754 acc_pose: 0.770013 loss: 0.000754 2022/09/26 12:23:57 - mmengine - INFO - Epoch(train) [33][100/293] lr: 5.000000e-04 eta: 7:38:36 time: 0.612200 data_time: 0.253605 memory: 9504 loss_kpt: 0.000757 acc_pose: 0.752187 loss: 0.000757 2022/09/26 12:24:26 - mmengine - INFO - Epoch(train) [33][150/293] lr: 5.000000e-04 eta: 7:38:21 time: 0.569232 data_time: 0.174115 memory: 9504 loss_kpt: 0.000748 acc_pose: 0.777882 loss: 0.000748 2022/09/26 12:24:55 - mmengine - INFO - Epoch(train) [33][200/293] lr: 5.000000e-04 eta: 7:38:09 time: 0.583951 data_time: 0.095172 memory: 9504 loss_kpt: 0.000757 acc_pose: 0.764586 loss: 0.000757 2022/09/26 12:25:24 - mmengine - INFO - Epoch(train) [33][250/293] lr: 5.000000e-04 eta: 7:37:57 time: 0.582098 data_time: 0.078137 memory: 9504 loss_kpt: 0.000760 acc_pose: 0.818859 loss: 0.000760 2022/09/26 12:25:49 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:26:20 - mmengine - INFO - Epoch(train) [34][50/293] lr: 5.000000e-04 eta: 7:35:30 time: 0.619540 data_time: 0.103655 memory: 9504 loss_kpt: 0.000759 acc_pose: 0.816721 loss: 0.000759 2022/09/26 12:26:48 - mmengine - INFO - Epoch(train) [34][100/293] lr: 5.000000e-04 eta: 7:35:13 time: 0.560098 data_time: 0.079718 memory: 9504 loss_kpt: 0.000748 acc_pose: 0.750561 loss: 0.000748 2022/09/26 12:27:18 - mmengine - INFO - Epoch(train) [34][150/293] lr: 5.000000e-04 eta: 7:35:06 time: 0.601853 data_time: 0.080386 memory: 9504 loss_kpt: 0.000740 acc_pose: 0.795686 loss: 0.000740 2022/09/26 12:27:46 - mmengine - INFO - Epoch(train) [34][200/293] lr: 5.000000e-04 eta: 7:34:50 time: 0.570244 data_time: 0.083582 memory: 9504 loss_kpt: 0.000767 acc_pose: 0.738265 loss: 0.000767 2022/09/26 12:28:15 - mmengine - INFO - Epoch(train) [34][250/293] lr: 5.000000e-04 eta: 7:34:37 time: 0.578587 data_time: 0.112853 memory: 9504 loss_kpt: 0.000753 acc_pose: 0.790665 loss: 0.000753 2022/09/26 12:28:40 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:29:05 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:29:11 - mmengine - INFO - Epoch(train) [35][50/293] lr: 5.000000e-04 eta: 7:32:14 time: 0.617230 data_time: 0.225161 memory: 9504 loss_kpt: 0.000743 acc_pose: 0.794622 loss: 0.000743 2022/09/26 12:29:41 - mmengine - INFO - Epoch(train) [35][100/293] lr: 5.000000e-04 eta: 7:32:04 time: 0.591977 data_time: 0.219353 memory: 9504 loss_kpt: 0.000739 acc_pose: 0.838869 loss: 0.000739 2022/09/26 12:30:11 - mmengine - INFO - Epoch(train) [35][150/293] lr: 5.000000e-04 eta: 7:31:57 time: 0.602319 data_time: 0.166538 memory: 9504 loss_kpt: 0.000744 acc_pose: 0.764154 loss: 0.000744 2022/09/26 12:30:39 - mmengine - INFO - Epoch(train) [35][200/293] lr: 5.000000e-04 eta: 7:31:40 time: 0.564543 data_time: 0.105405 memory: 9504 loss_kpt: 0.000759 acc_pose: 0.839888 loss: 0.000759 2022/09/26 12:31:08 - mmengine - INFO - Epoch(train) [35][250/293] lr: 5.000000e-04 eta: 7:31:27 time: 0.581793 data_time: 0.078505 memory: 9504 loss_kpt: 0.000744 acc_pose: 0.807735 loss: 0.000744 2022/09/26 12:31:33 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:32:04 - mmengine - INFO - Epoch(train) [36][50/293] lr: 5.000000e-04 eta: 7:29:08 time: 0.620869 data_time: 0.096288 memory: 9504 loss_kpt: 0.000733 acc_pose: 0.744709 loss: 0.000733 2022/09/26 12:32:33 - mmengine - INFO - Epoch(train) [36][100/293] lr: 5.000000e-04 eta: 7:28:54 time: 0.575179 data_time: 0.137907 memory: 9504 loss_kpt: 0.000741 acc_pose: 0.782641 loss: 0.000741 2022/09/26 12:33:03 - mmengine - INFO - Epoch(train) [36][150/293] lr: 5.000000e-04 eta: 7:28:46 time: 0.599821 data_time: 0.096506 memory: 9504 loss_kpt: 0.000746 acc_pose: 0.762679 loss: 0.000746 2022/09/26 12:33:32 - mmengine - INFO - Epoch(train) [36][200/293] lr: 5.000000e-04 eta: 7:28:33 time: 0.583006 data_time: 0.114865 memory: 9504 loss_kpt: 0.000744 acc_pose: 0.819792 loss: 0.000744 2022/09/26 12:34:02 - mmengine - INFO - Epoch(train) [36][250/293] lr: 5.000000e-04 eta: 7:28:22 time: 0.590070 data_time: 0.126655 memory: 9504 loss_kpt: 0.000740 acc_pose: 0.767253 loss: 0.000740 2022/09/26 12:34:27 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:34:56 - mmengine - INFO - Epoch(train) [37][50/293] lr: 5.000000e-04 eta: 7:26:01 time: 0.596425 data_time: 0.102047 memory: 9504 loss_kpt: 0.000739 acc_pose: 0.787818 loss: 0.000739 2022/09/26 12:35:26 - mmengine - INFO - Epoch(train) [37][100/293] lr: 5.000000e-04 eta: 7:25:50 time: 0.590675 data_time: 0.085524 memory: 9504 loss_kpt: 0.000746 acc_pose: 0.803583 loss: 0.000746 2022/09/26 12:35:56 - mmengine - INFO - Epoch(train) [37][150/293] lr: 5.000000e-04 eta: 7:25:41 time: 0.596014 data_time: 0.158578 memory: 9504 loss_kpt: 0.000741 acc_pose: 0.777700 loss: 0.000741 2022/09/26 12:36:25 - mmengine - INFO - Epoch(train) [37][200/293] lr: 5.000000e-04 eta: 7:25:27 time: 0.577165 data_time: 0.112797 memory: 9504 loss_kpt: 0.000742 acc_pose: 0.697808 loss: 0.000742 2022/09/26 12:36:55 - mmengine - INFO - Epoch(train) [37][250/293] lr: 5.000000e-04 eta: 7:25:21 time: 0.615337 data_time: 0.092261 memory: 9504 loss_kpt: 0.000754 acc_pose: 0.791963 loss: 0.000754 2022/09/26 12:37:19 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:37:49 - mmengine - INFO - Epoch(train) [38][50/293] lr: 5.000000e-04 eta: 7:23:05 time: 0.606189 data_time: 0.098170 memory: 9504 loss_kpt: 0.000740 acc_pose: 0.786896 loss: 0.000740 2022/09/26 12:38:20 - mmengine - INFO - Epoch(train) [38][100/293] lr: 5.000000e-04 eta: 7:22:58 time: 0.606132 data_time: 0.081540 memory: 9504 loss_kpt: 0.000739 acc_pose: 0.809544 loss: 0.000739 2022/09/26 12:38:49 - mmengine - INFO - Epoch(train) [38][150/293] lr: 5.000000e-04 eta: 7:22:47 time: 0.592906 data_time: 0.077722 memory: 9504 loss_kpt: 0.000753 acc_pose: 0.811052 loss: 0.000753 2022/09/26 12:38:55 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:39:19 - mmengine - INFO - Epoch(train) [38][200/293] lr: 5.000000e-04 eta: 7:22:36 time: 0.592012 data_time: 0.066279 memory: 9504 loss_kpt: 0.000731 acc_pose: 0.788417 loss: 0.000731 2022/09/26 12:39:48 - mmengine - INFO - Epoch(train) [38][250/293] lr: 5.000000e-04 eta: 7:22:21 time: 0.576471 data_time: 0.123178 memory: 9504 loss_kpt: 0.000749 acc_pose: 0.769864 loss: 0.000749 2022/09/26 12:40:14 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:40:45 - mmengine - INFO - Epoch(train) [39][50/293] lr: 5.000000e-04 eta: 7:20:08 time: 0.605998 data_time: 0.232740 memory: 9504 loss_kpt: 0.000737 acc_pose: 0.778456 loss: 0.000737 2022/09/26 12:41:15 - mmengine - INFO - Epoch(train) [39][100/293] lr: 5.000000e-04 eta: 7:20:00 time: 0.604010 data_time: 0.100409 memory: 9504 loss_kpt: 0.000732 acc_pose: 0.860836 loss: 0.000732 2022/09/26 12:41:44 - mmengine - INFO - Epoch(train) [39][150/293] lr: 5.000000e-04 eta: 7:19:47 time: 0.583949 data_time: 0.089377 memory: 9504 loss_kpt: 0.000744 acc_pose: 0.823592 loss: 0.000744 2022/09/26 12:42:14 - mmengine - INFO - Epoch(train) [39][200/293] lr: 5.000000e-04 eta: 7:19:37 time: 0.601327 data_time: 0.106862 memory: 9504 loss_kpt: 0.000737 acc_pose: 0.826135 loss: 0.000737 2022/09/26 12:42:43 - mmengine - INFO - Epoch(train) [39][250/293] lr: 5.000000e-04 eta: 7:19:24 time: 0.584006 data_time: 0.106672 memory: 9504 loss_kpt: 0.000744 acc_pose: 0.759432 loss: 0.000744 2022/09/26 12:43:08 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:43:40 - mmengine - INFO - Epoch(train) [40][50/293] lr: 5.000000e-04 eta: 7:17:19 time: 0.627549 data_time: 0.265822 memory: 9504 loss_kpt: 0.000733 acc_pose: 0.746635 loss: 0.000733 2022/09/26 12:44:09 - mmengine - INFO - Epoch(train) [40][100/293] lr: 5.000000e-04 eta: 7:17:05 time: 0.583678 data_time: 0.231243 memory: 9504 loss_kpt: 0.000729 acc_pose: 0.799883 loss: 0.000729 2022/09/26 12:44:39 - mmengine - INFO - Epoch(train) [40][150/293] lr: 5.000000e-04 eta: 7:16:54 time: 0.594777 data_time: 0.076552 memory: 9504 loss_kpt: 0.000735 acc_pose: 0.760314 loss: 0.000735 2022/09/26 12:45:09 - mmengine - INFO - Epoch(train) [40][200/293] lr: 5.000000e-04 eta: 7:16:45 time: 0.601864 data_time: 0.081268 memory: 9504 loss_kpt: 0.000740 acc_pose: 0.744996 loss: 0.000740 2022/09/26 12:45:39 - mmengine - INFO - Epoch(train) [40][250/293] lr: 5.000000e-04 eta: 7:16:34 time: 0.597034 data_time: 0.096552 memory: 9504 loss_kpt: 0.000732 acc_pose: 0.778063 loss: 0.000732 2022/09/26 12:46:04 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:46:04 - mmengine - INFO - Saving checkpoint at 40 epochs 2022/09/26 12:46:29 - mmengine - INFO - Epoch(val) [40][50/407] eta: 0:01:59 time: 0.335423 data_time: 0.154629 memory: 9504 2022/09/26 12:46:45 - mmengine - INFO - Epoch(val) [40][100/407] eta: 0:01:40 time: 0.328210 data_time: 0.157023 memory: 1378 2022/09/26 12:47:01 - mmengine - INFO - Epoch(val) [40][150/407] eta: 0:01:23 time: 0.325232 data_time: 0.155956 memory: 1378 2022/09/26 12:47:18 - mmengine - INFO - Epoch(val) [40][200/407] eta: 0:01:10 time: 0.341092 data_time: 0.175108 memory: 1378 2022/09/26 12:47:34 - mmengine - INFO - Epoch(val) [40][250/407] eta: 0:00:47 time: 0.301722 data_time: 0.118905 memory: 1378 2022/09/26 12:47:50 - mmengine - INFO - Epoch(val) [40][300/407] eta: 0:00:34 time: 0.326626 data_time: 0.147483 memory: 1378 2022/09/26 12:48:06 - mmengine - INFO - Epoch(val) [40][350/407] eta: 0:00:18 time: 0.325965 data_time: 0.144351 memory: 1378 2022/09/26 12:48:17 - mmengine - INFO - Epoch(val) [40][400/407] eta: 0:00:01 time: 0.207825 data_time: 0.103176 memory: 1378 2022/09/26 12:48:49 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 12:49:03 - mmengine - INFO - Epoch(val) [40][407/407] coco/AP: 0.679695 coco/AP .5: 0.885720 coco/AP .75: 0.759713 coco/AP (M): 0.642920 coco/AP (L): 0.746485 coco/AR: 0.739956 coco/AR .5: 0.927739 coco/AR .75: 0.812028 coco/AR (M): 0.696203 coco/AR (L): 0.802155 2022/09/26 12:49:03 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_30.pth is removed 2022/09/26 12:49:06 - mmengine - INFO - The best checkpoint with 0.6797 coco/AP at 40 epoch is saved to best_coco/AP_epoch_40.pth. 2022/09/26 12:49:34 - mmengine - INFO - Epoch(train) [41][50/293] lr: 5.000000e-04 eta: 7:14:20 time: 0.574261 data_time: 0.244879 memory: 9504 loss_kpt: 0.000733 acc_pose: 0.790409 loss: 0.000733 2022/09/26 12:50:04 - mmengine - INFO - Epoch(train) [41][100/293] lr: 5.000000e-04 eta: 7:14:10 time: 0.602700 data_time: 0.129688 memory: 9504 loss_kpt: 0.000729 acc_pose: 0.827820 loss: 0.000729 2022/09/26 12:50:35 - mmengine - INFO - Epoch(train) [41][150/293] lr: 5.000000e-04 eta: 7:14:00 time: 0.600726 data_time: 0.121451 memory: 9504 loss_kpt: 0.000731 acc_pose: 0.812243 loss: 0.000731 2022/09/26 12:51:04 - mmengine - INFO - Epoch(train) [41][200/293] lr: 5.000000e-04 eta: 7:13:46 time: 0.582795 data_time: 0.225524 memory: 9504 loss_kpt: 0.000737 acc_pose: 0.805270 loss: 0.000737 2022/09/26 12:51:33 - mmengine - INFO - Epoch(train) [41][250/293] lr: 5.000000e-04 eta: 7:13:34 time: 0.594069 data_time: 0.101291 memory: 9504 loss_kpt: 0.000728 acc_pose: 0.780663 loss: 0.000728 2022/09/26 12:51:51 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:51:59 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:52:30 - mmengine - INFO - Epoch(train) [42][50/293] lr: 5.000000e-04 eta: 7:11:32 time: 0.618255 data_time: 0.223579 memory: 9504 loss_kpt: 0.000709 acc_pose: 0.810295 loss: 0.000709 2022/09/26 12:52:59 - mmengine - INFO - Epoch(train) [42][100/293] lr: 5.000000e-04 eta: 7:11:21 time: 0.597578 data_time: 0.143604 memory: 9504 loss_kpt: 0.000738 acc_pose: 0.811825 loss: 0.000738 2022/09/26 12:53:29 - mmengine - INFO - Epoch(train) [42][150/293] lr: 5.000000e-04 eta: 7:11:08 time: 0.590305 data_time: 0.080945 memory: 9504 loss_kpt: 0.000733 acc_pose: 0.779797 loss: 0.000733 2022/09/26 12:53:59 - mmengine - INFO - Epoch(train) [42][200/293] lr: 5.000000e-04 eta: 7:10:57 time: 0.596925 data_time: 0.146608 memory: 9504 loss_kpt: 0.000739 acc_pose: 0.759936 loss: 0.000739 2022/09/26 12:54:28 - mmengine - INFO - Epoch(train) [42][250/293] lr: 5.000000e-04 eta: 7:10:43 time: 0.584759 data_time: 0.168444 memory: 9504 loss_kpt: 0.000724 acc_pose: 0.841086 loss: 0.000724 2022/09/26 12:54:53 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:55:25 - mmengine - INFO - Epoch(train) [43][50/293] lr: 5.000000e-04 eta: 7:08:46 time: 0.635492 data_time: 0.116898 memory: 9504 loss_kpt: 0.000723 acc_pose: 0.750858 loss: 0.000723 2022/09/26 12:55:55 - mmengine - INFO - Epoch(train) [43][100/293] lr: 5.000000e-04 eta: 7:08:36 time: 0.602738 data_time: 0.080989 memory: 9504 loss_kpt: 0.000727 acc_pose: 0.759770 loss: 0.000727 2022/09/26 12:56:26 - mmengine - INFO - Epoch(train) [43][150/293] lr: 5.000000e-04 eta: 7:08:30 time: 0.627796 data_time: 0.139236 memory: 9504 loss_kpt: 0.000717 acc_pose: 0.817674 loss: 0.000717 2022/09/26 12:56:55 - mmengine - INFO - Epoch(train) [43][200/293] lr: 5.000000e-04 eta: 7:08:13 time: 0.571915 data_time: 0.075289 memory: 9504 loss_kpt: 0.000724 acc_pose: 0.746546 loss: 0.000724 2022/09/26 12:57:24 - mmengine - INFO - Epoch(train) [43][250/293] lr: 5.000000e-04 eta: 7:08:00 time: 0.593172 data_time: 0.080164 memory: 9504 loss_kpt: 0.000718 acc_pose: 0.800908 loss: 0.000718 2022/09/26 12:57:50 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 12:58:20 - mmengine - INFO - Epoch(train) [44][50/293] lr: 5.000000e-04 eta: 7:06:01 time: 0.611740 data_time: 0.160614 memory: 9504 loss_kpt: 0.000719 acc_pose: 0.833237 loss: 0.000719 2022/09/26 12:58:51 - mmengine - INFO - Epoch(train) [44][100/293] lr: 5.000000e-04 eta: 7:05:51 time: 0.604828 data_time: 0.107958 memory: 9504 loss_kpt: 0.000726 acc_pose: 0.758041 loss: 0.000726 2022/09/26 12:59:21 - mmengine - INFO - Epoch(train) [44][150/293] lr: 5.000000e-04 eta: 7:05:39 time: 0.597307 data_time: 0.073901 memory: 9504 loss_kpt: 0.000734 acc_pose: 0.768114 loss: 0.000734 2022/09/26 12:59:51 - mmengine - INFO - Epoch(train) [44][200/293] lr: 5.000000e-04 eta: 7:05:30 time: 0.613542 data_time: 0.088081 memory: 9504 loss_kpt: 0.000712 acc_pose: 0.747857 loss: 0.000712 2022/09/26 13:00:20 - mmengine - INFO - Epoch(train) [44][250/293] lr: 5.000000e-04 eta: 7:05:15 time: 0.584981 data_time: 0.071694 memory: 9504 loss_kpt: 0.000728 acc_pose: 0.812623 loss: 0.000728 2022/09/26 13:00:44 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:01:14 - mmengine - INFO - Epoch(train) [45][50/293] lr: 5.000000e-04 eta: 7:03:17 time: 0.605224 data_time: 0.199640 memory: 9504 loss_kpt: 0.000723 acc_pose: 0.831527 loss: 0.000723 2022/09/26 13:01:44 - mmengine - INFO - Epoch(train) [45][100/293] lr: 5.000000e-04 eta: 7:03:06 time: 0.601905 data_time: 0.092354 memory: 9504 loss_kpt: 0.000729 acc_pose: 0.819068 loss: 0.000729 2022/09/26 13:01:49 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:03:37 - mmengine - INFO - Epoch(train) [45][150/293] lr: 5.000000e-04 eta: 7:08:00 time: 2.244226 data_time: 0.120196 memory: 9504 loss_kpt: 0.000729 acc_pose: 0.772980 loss: 0.000729 2022/09/26 13:04:06 - mmengine - INFO - Epoch(train) [45][200/293] lr: 5.000000e-04 eta: 7:07:43 time: 0.583116 data_time: 0.091130 memory: 9504 loss_kpt: 0.000717 acc_pose: 0.797313 loss: 0.000717 2022/09/26 13:04:38 - mmengine - INFO - Epoch(train) [45][250/293] lr: 5.000000e-04 eta: 7:07:36 time: 0.634454 data_time: 0.104253 memory: 9504 loss_kpt: 0.000726 acc_pose: 0.809918 loss: 0.000726 2022/09/26 13:05:02 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:05:34 - mmengine - INFO - Epoch(train) [46][50/293] lr: 5.000000e-04 eta: 7:05:43 time: 0.636148 data_time: 0.135731 memory: 9504 loss_kpt: 0.000708 acc_pose: 0.792238 loss: 0.000708 2022/09/26 13:06:04 - mmengine - INFO - Epoch(train) [46][100/293] lr: 5.000000e-04 eta: 7:05:31 time: 0.612309 data_time: 0.095447 memory: 9504 loss_kpt: 0.000736 acc_pose: 0.824805 loss: 0.000736 2022/09/26 13:06:33 - mmengine - INFO - Epoch(train) [46][150/293] lr: 5.000000e-04 eta: 7:05:14 time: 0.579284 data_time: 0.078776 memory: 9504 loss_kpt: 0.000711 acc_pose: 0.802369 loss: 0.000711 2022/09/26 13:07:02 - mmengine - INFO - Epoch(train) [46][200/293] lr: 5.000000e-04 eta: 7:04:55 time: 0.571314 data_time: 0.100886 memory: 9504 loss_kpt: 0.000721 acc_pose: 0.700554 loss: 0.000721 2022/09/26 13:07:30 - mmengine - INFO - Epoch(train) [46][250/293] lr: 5.000000e-04 eta: 7:04:36 time: 0.569306 data_time: 0.118668 memory: 9504 loss_kpt: 0.000729 acc_pose: 0.805681 loss: 0.000729 2022/09/26 13:07:55 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:08:26 - mmengine - INFO - Epoch(train) [47][50/293] lr: 5.000000e-04 eta: 7:02:42 time: 0.622721 data_time: 0.104927 memory: 9504 loss_kpt: 0.000717 acc_pose: 0.821804 loss: 0.000717 2022/09/26 13:08:55 - mmengine - INFO - Epoch(train) [47][100/293] lr: 5.000000e-04 eta: 7:02:24 time: 0.576434 data_time: 0.077891 memory: 9504 loss_kpt: 0.000730 acc_pose: 0.791421 loss: 0.000730 2022/09/26 13:09:24 - mmengine - INFO - Epoch(train) [47][150/293] lr: 5.000000e-04 eta: 7:02:09 time: 0.592087 data_time: 0.140888 memory: 9504 loss_kpt: 0.000733 acc_pose: 0.849030 loss: 0.000733 2022/09/26 13:09:54 - mmengine - INFO - Epoch(train) [47][200/293] lr: 5.000000e-04 eta: 7:01:55 time: 0.598498 data_time: 0.107631 memory: 9504 loss_kpt: 0.000719 acc_pose: 0.780253 loss: 0.000719 2022/09/26 13:10:24 - mmengine - INFO - Epoch(train) [47][250/293] lr: 5.000000e-04 eta: 7:01:39 time: 0.592744 data_time: 0.193821 memory: 9504 loss_kpt: 0.000720 acc_pose: 0.751239 loss: 0.000720 2022/09/26 13:10:49 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:11:20 - mmengine - INFO - Epoch(train) [48][50/293] lr: 5.000000e-04 eta: 6:59:47 time: 0.620687 data_time: 0.144081 memory: 9504 loss_kpt: 0.000727 acc_pose: 0.772437 loss: 0.000727 2022/09/26 13:11:51 - mmengine - INFO - Epoch(train) [48][100/293] lr: 5.000000e-04 eta: 6:59:35 time: 0.611214 data_time: 0.084984 memory: 9504 loss_kpt: 0.000721 acc_pose: 0.767184 loss: 0.000721 2022/09/26 13:12:21 - mmengine - INFO - Epoch(train) [48][150/293] lr: 5.000000e-04 eta: 6:59:20 time: 0.595606 data_time: 0.102844 memory: 9504 loss_kpt: 0.000726 acc_pose: 0.781635 loss: 0.000726 2022/09/26 13:12:51 - mmengine - INFO - Epoch(train) [48][200/293] lr: 5.000000e-04 eta: 6:59:06 time: 0.596797 data_time: 0.106670 memory: 9504 loss_kpt: 0.000713 acc_pose: 0.824249 loss: 0.000713 2022/09/26 13:13:07 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:13:20 - mmengine - INFO - Epoch(train) [48][250/293] lr: 5.000000e-04 eta: 6:58:48 time: 0.579279 data_time: 0.140391 memory: 9504 loss_kpt: 0.000717 acc_pose: 0.811111 loss: 0.000717 2022/09/26 13:13:46 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:14:18 - mmengine - INFO - Epoch(train) [49][50/293] lr: 5.000000e-04 eta: 6:57:01 time: 0.641143 data_time: 0.158474 memory: 9504 loss_kpt: 0.000719 acc_pose: 0.778746 loss: 0.000719 2022/09/26 13:14:47 - mmengine - INFO - Epoch(train) [49][100/293] lr: 5.000000e-04 eta: 6:56:44 time: 0.585137 data_time: 0.070175 memory: 9504 loss_kpt: 0.000721 acc_pose: 0.779238 loss: 0.000721 2022/09/26 13:15:16 - mmengine - INFO - Epoch(train) [49][150/293] lr: 5.000000e-04 eta: 6:56:27 time: 0.583149 data_time: 0.070376 memory: 9504 loss_kpt: 0.000717 acc_pose: 0.808260 loss: 0.000717 2022/09/26 13:15:45 - mmengine - INFO - Epoch(train) [49][200/293] lr: 5.000000e-04 eta: 6:56:08 time: 0.574417 data_time: 0.082082 memory: 9504 loss_kpt: 0.000715 acc_pose: 0.809985 loss: 0.000715 2022/09/26 13:16:15 - mmengine - INFO - Epoch(train) [49][250/293] lr: 5.000000e-04 eta: 6:55:54 time: 0.603179 data_time: 0.146716 memory: 9504 loss_kpt: 0.000714 acc_pose: 0.792593 loss: 0.000714 2022/09/26 13:16:39 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:17:09 - mmengine - INFO - Epoch(train) [50][50/293] lr: 5.000000e-04 eta: 6:54:03 time: 0.602230 data_time: 0.154965 memory: 9504 loss_kpt: 0.000707 acc_pose: 0.818599 loss: 0.000707 2022/09/26 13:17:39 - mmengine - INFO - Epoch(train) [50][100/293] lr: 5.000000e-04 eta: 6:53:47 time: 0.595123 data_time: 0.086317 memory: 9504 loss_kpt: 0.000706 acc_pose: 0.762054 loss: 0.000706 2022/09/26 13:18:08 - mmengine - INFO - Epoch(train) [50][150/293] lr: 5.000000e-04 eta: 6:53:29 time: 0.578527 data_time: 0.077757 memory: 9504 loss_kpt: 0.000692 acc_pose: 0.775106 loss: 0.000692 2022/09/26 13:18:38 - mmengine - INFO - Epoch(train) [50][200/293] lr: 5.000000e-04 eta: 6:53:14 time: 0.597769 data_time: 0.178455 memory: 9504 loss_kpt: 0.000715 acc_pose: 0.811894 loss: 0.000715 2022/09/26 13:19:08 - mmengine - INFO - Epoch(train) [50][250/293] lr: 5.000000e-04 eta: 6:53:00 time: 0.603869 data_time: 0.090867 memory: 9504 loss_kpt: 0.000730 acc_pose: 0.741679 loss: 0.000730 2022/09/26 13:19:33 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:19:33 - mmengine - INFO - Saving checkpoint at 50 epochs 2022/09/26 13:19:59 - mmengine - INFO - Epoch(val) [50][50/407] eta: 0:02:00 time: 0.338160 data_time: 0.161278 memory: 9504 2022/09/26 13:20:15 - mmengine - INFO - Epoch(val) [50][100/407] eta: 0:01:36 time: 0.313657 data_time: 0.128203 memory: 1378 2022/09/26 13:20:30 - mmengine - INFO - Epoch(val) [50][150/407] eta: 0:01:18 time: 0.306651 data_time: 0.145701 memory: 1378 2022/09/26 13:20:46 - mmengine - INFO - Epoch(val) [50][200/407] eta: 0:01:03 time: 0.307501 data_time: 0.135182 memory: 1378 2022/09/26 13:21:01 - mmengine - INFO - Epoch(val) [50][250/407] eta: 0:00:48 time: 0.309771 data_time: 0.146510 memory: 1378 2022/09/26 13:21:17 - mmengine - INFO - Epoch(val) [50][300/407] eta: 0:00:34 time: 0.321038 data_time: 0.151849 memory: 1378 2022/09/26 13:21:33 - mmengine - INFO - Epoch(val) [50][350/407] eta: 0:00:18 time: 0.325055 data_time: 0.136467 memory: 1378 2022/09/26 13:21:45 - mmengine - INFO - Epoch(val) [50][400/407] eta: 0:00:01 time: 0.224089 data_time: 0.102193 memory: 1378 2022/09/26 13:22:17 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 13:22:30 - mmengine - INFO - Epoch(val) [50][407/407] coco/AP: 0.684395 coco/AP .5: 0.884361 coco/AP .75: 0.762660 coco/AP (M): 0.649739 coco/AP (L): 0.751017 coco/AR: 0.743687 coco/AR .5: 0.926795 coco/AR .75: 0.813917 coco/AR (M): 0.700219 coco/AR (L): 0.805797 2022/09/26 13:22:30 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_40.pth is removed 2022/09/26 13:22:33 - mmengine - INFO - The best checkpoint with 0.6844 coco/AP at 50 epoch is saved to best_coco/AP_epoch_50.pth. 2022/09/26 13:23:02 - mmengine - INFO - Epoch(train) [51][50/293] lr: 5.000000e-04 eta: 6:51:09 time: 0.595995 data_time: 0.147972 memory: 9504 loss_kpt: 0.000705 acc_pose: 0.782871 loss: 0.000705 2022/09/26 13:23:31 - mmengine - INFO - Epoch(train) [51][100/293] lr: 5.000000e-04 eta: 6:50:50 time: 0.573300 data_time: 0.081197 memory: 9504 loss_kpt: 0.000707 acc_pose: 0.816794 loss: 0.000707 2022/09/26 13:24:01 - mmengine - INFO - Epoch(train) [51][150/293] lr: 5.000000e-04 eta: 6:50:36 time: 0.605087 data_time: 0.092677 memory: 9504 loss_kpt: 0.000718 acc_pose: 0.748421 loss: 0.000718 2022/09/26 13:24:30 - mmengine - INFO - Epoch(train) [51][200/293] lr: 5.000000e-04 eta: 6:50:19 time: 0.580945 data_time: 0.180173 memory: 9504 loss_kpt: 0.000712 acc_pose: 0.781639 loss: 0.000712 2022/09/26 13:24:59 - mmengine - INFO - Epoch(train) [51][250/293] lr: 5.000000e-04 eta: 6:50:00 time: 0.576100 data_time: 0.078691 memory: 9504 loss_kpt: 0.000693 acc_pose: 0.786043 loss: 0.000693 2022/09/26 13:25:26 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:25:58 - mmengine - INFO - Epoch(train) [52][50/293] lr: 5.000000e-04 eta: 6:48:17 time: 0.635408 data_time: 0.203508 memory: 9504 loss_kpt: 0.000702 acc_pose: 0.759519 loss: 0.000702 2022/09/26 13:26:02 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:26:27 - mmengine - INFO - Epoch(train) [52][100/293] lr: 5.000000e-04 eta: 6:47:57 time: 0.567314 data_time: 0.196361 memory: 9504 loss_kpt: 0.000714 acc_pose: 0.803080 loss: 0.000714 2022/09/26 13:26:55 - mmengine - INFO - Epoch(train) [52][150/293] lr: 5.000000e-04 eta: 6:47:38 time: 0.576656 data_time: 0.193524 memory: 9504 loss_kpt: 0.000721 acc_pose: 0.786486 loss: 0.000721 2022/09/26 13:27:24 - mmengine - INFO - Epoch(train) [52][200/293] lr: 5.000000e-04 eta: 6:47:18 time: 0.562744 data_time: 0.074529 memory: 9504 loss_kpt: 0.000702 acc_pose: 0.811913 loss: 0.000702 2022/09/26 13:27:53 - mmengine - INFO - Epoch(train) [52][250/293] lr: 5.000000e-04 eta: 6:47:00 time: 0.585765 data_time: 0.074142 memory: 9504 loss_kpt: 0.000702 acc_pose: 0.761972 loss: 0.000702 2022/09/26 13:28:18 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:28:48 - mmengine - INFO - Epoch(train) [53][50/293] lr: 5.000000e-04 eta: 6:45:16 time: 0.614937 data_time: 0.137899 memory: 9504 loss_kpt: 0.000704 acc_pose: 0.803020 loss: 0.000704 2022/09/26 13:29:19 - mmengine - INFO - Epoch(train) [53][100/293] lr: 5.000000e-04 eta: 6:45:01 time: 0.603332 data_time: 0.156027 memory: 9504 loss_kpt: 0.000704 acc_pose: 0.781337 loss: 0.000704 2022/09/26 13:29:49 - mmengine - INFO - Epoch(train) [53][150/293] lr: 5.000000e-04 eta: 6:44:46 time: 0.600510 data_time: 0.072603 memory: 9504 loss_kpt: 0.000725 acc_pose: 0.830107 loss: 0.000725 2022/09/26 13:30:18 - mmengine - INFO - Epoch(train) [53][200/293] lr: 5.000000e-04 eta: 6:44:29 time: 0.587052 data_time: 0.093364 memory: 9504 loss_kpt: 0.000712 acc_pose: 0.784735 loss: 0.000712 2022/09/26 13:30:48 - mmengine - INFO - Epoch(train) [53][250/293] lr: 5.000000e-04 eta: 6:44:15 time: 0.609731 data_time: 0.118193 memory: 9504 loss_kpt: 0.000707 acc_pose: 0.836611 loss: 0.000707 2022/09/26 13:31:13 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:31:44 - mmengine - INFO - Epoch(train) [54][50/293] lr: 5.000000e-04 eta: 6:42:34 time: 0.623581 data_time: 0.093427 memory: 9504 loss_kpt: 0.000699 acc_pose: 0.778801 loss: 0.000699 2022/09/26 13:32:14 - mmengine - INFO - Epoch(train) [54][100/293] lr: 5.000000e-04 eta: 6:42:18 time: 0.600857 data_time: 0.101985 memory: 9504 loss_kpt: 0.000721 acc_pose: 0.799117 loss: 0.000721 2022/09/26 13:32:44 - mmengine - INFO - Epoch(train) [54][150/293] lr: 5.000000e-04 eta: 6:42:00 time: 0.581405 data_time: 0.084166 memory: 9504 loss_kpt: 0.000713 acc_pose: 0.804179 loss: 0.000713 2022/09/26 13:33:12 - mmengine - INFO - Epoch(train) [54][200/293] lr: 5.000000e-04 eta: 6:41:41 time: 0.578498 data_time: 0.189657 memory: 9504 loss_kpt: 0.000707 acc_pose: 0.847658 loss: 0.000707 2022/09/26 13:33:41 - mmengine - INFO - Epoch(train) [54][250/293] lr: 5.000000e-04 eta: 6:41:22 time: 0.571927 data_time: 0.182362 memory: 9504 loss_kpt: 0.000706 acc_pose: 0.784235 loss: 0.000706 2022/09/26 13:34:06 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:34:36 - mmengine - INFO - Epoch(train) [55][50/293] lr: 5.000000e-04 eta: 6:39:39 time: 0.602214 data_time: 0.103177 memory: 9504 loss_kpt: 0.000694 acc_pose: 0.823828 loss: 0.000694 2022/09/26 13:35:06 - mmengine - INFO - Epoch(train) [55][100/293] lr: 5.000000e-04 eta: 6:39:21 time: 0.583503 data_time: 0.074651 memory: 9504 loss_kpt: 0.000706 acc_pose: 0.821315 loss: 0.000706 2022/09/26 13:35:36 - mmengine - INFO - Epoch(train) [55][150/293] lr: 5.000000e-04 eta: 6:39:06 time: 0.607784 data_time: 0.084993 memory: 9504 loss_kpt: 0.000691 acc_pose: 0.758982 loss: 0.000691 2022/09/26 13:35:52 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:36:06 - mmengine - INFO - Epoch(train) [55][200/293] lr: 5.000000e-04 eta: 6:38:51 time: 0.602178 data_time: 0.083941 memory: 9504 loss_kpt: 0.000697 acc_pose: 0.787546 loss: 0.000697 2022/09/26 13:36:36 - mmengine - INFO - Epoch(train) [55][250/293] lr: 5.000000e-04 eta: 6:38:35 time: 0.598811 data_time: 0.176876 memory: 9504 loss_kpt: 0.000694 acc_pose: 0.792399 loss: 0.000694 2022/09/26 13:37:01 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:37:33 - mmengine - INFO - Epoch(train) [56][50/293] lr: 5.000000e-04 eta: 6:36:56 time: 0.624484 data_time: 0.118427 memory: 9504 loss_kpt: 0.000707 acc_pose: 0.820059 loss: 0.000707 2022/09/26 13:38:03 - mmengine - INFO - Epoch(train) [56][100/293] lr: 5.000000e-04 eta: 6:36:42 time: 0.608133 data_time: 0.168554 memory: 9504 loss_kpt: 0.000709 acc_pose: 0.774748 loss: 0.000709 2022/09/26 13:38:33 - mmengine - INFO - Epoch(train) [56][150/293] lr: 5.000000e-04 eta: 6:36:25 time: 0.595776 data_time: 0.144029 memory: 9504 loss_kpt: 0.000696 acc_pose: 0.797192 loss: 0.000696 2022/09/26 13:39:03 - mmengine - INFO - Epoch(train) [56][200/293] lr: 5.000000e-04 eta: 6:36:10 time: 0.605877 data_time: 0.077818 memory: 9504 loss_kpt: 0.000700 acc_pose: 0.780762 loss: 0.000700 2022/09/26 13:39:32 - mmengine - INFO - Epoch(train) [56][250/293] lr: 5.000000e-04 eta: 6:35:50 time: 0.572329 data_time: 0.085328 memory: 9504 loss_kpt: 0.000702 acc_pose: 0.838530 loss: 0.000702 2022/09/26 13:39:57 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:40:29 - mmengine - INFO - Epoch(train) [57][50/293] lr: 5.000000e-04 eta: 6:34:12 time: 0.622179 data_time: 0.142335 memory: 9504 loss_kpt: 0.000703 acc_pose: 0.796022 loss: 0.000703 2022/09/26 13:40:58 - mmengine - INFO - Epoch(train) [57][100/293] lr: 5.000000e-04 eta: 6:33:55 time: 0.586565 data_time: 0.112557 memory: 9504 loss_kpt: 0.000705 acc_pose: 0.823357 loss: 0.000705 2022/09/26 13:41:29 - mmengine - INFO - Epoch(train) [57][150/293] lr: 5.000000e-04 eta: 6:33:43 time: 0.631979 data_time: 0.229225 memory: 9504 loss_kpt: 0.000693 acc_pose: 0.804924 loss: 0.000693 2022/09/26 13:41:59 - mmengine - INFO - Epoch(train) [57][200/293] lr: 5.000000e-04 eta: 6:33:25 time: 0.589819 data_time: 0.073545 memory: 9504 loss_kpt: 0.000715 acc_pose: 0.777293 loss: 0.000715 2022/09/26 13:42:28 - mmengine - INFO - Epoch(train) [57][250/293] lr: 5.000000e-04 eta: 6:33:06 time: 0.578293 data_time: 0.079829 memory: 9504 loss_kpt: 0.000704 acc_pose: 0.773826 loss: 0.000704 2022/09/26 13:42:53 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:43:24 - mmengine - INFO - Epoch(train) [58][50/293] lr: 5.000000e-04 eta: 6:31:29 time: 0.618966 data_time: 0.148140 memory: 9504 loss_kpt: 0.000692 acc_pose: 0.770833 loss: 0.000692 2022/09/26 13:43:53 - mmengine - INFO - Epoch(train) [58][100/293] lr: 5.000000e-04 eta: 6:31:11 time: 0.585990 data_time: 0.081141 memory: 9504 loss_kpt: 0.000705 acc_pose: 0.818251 loss: 0.000705 2022/09/26 13:44:22 - mmengine - INFO - Epoch(train) [58][150/293] lr: 5.000000e-04 eta: 6:30:53 time: 0.581914 data_time: 0.083834 memory: 9504 loss_kpt: 0.000718 acc_pose: 0.787007 loss: 0.000718 2022/09/26 13:44:52 - mmengine - INFO - Epoch(train) [58][200/293] lr: 5.000000e-04 eta: 6:30:37 time: 0.606870 data_time: 0.200503 memory: 9504 loss_kpt: 0.000695 acc_pose: 0.822066 loss: 0.000695 2022/09/26 13:45:21 - mmengine - INFO - Epoch(train) [58][250/293] lr: 5.000000e-04 eta: 6:30:19 time: 0.583159 data_time: 0.076735 memory: 9504 loss_kpt: 0.000703 acc_pose: 0.764109 loss: 0.000703 2022/09/26 13:45:46 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:45:50 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:46:16 - mmengine - INFO - Epoch(train) [59][50/293] lr: 5.000000e-04 eta: 6:28:40 time: 0.599614 data_time: 0.097454 memory: 9504 loss_kpt: 0.000692 acc_pose: 0.839690 loss: 0.000692 2022/09/26 13:46:45 - mmengine - INFO - Epoch(train) [59][100/293] lr: 5.000000e-04 eta: 6:28:22 time: 0.584158 data_time: 0.099165 memory: 9504 loss_kpt: 0.000694 acc_pose: 0.789174 loss: 0.000694 2022/09/26 13:47:15 - mmengine - INFO - Epoch(train) [59][150/293] lr: 5.000000e-04 eta: 6:28:06 time: 0.601290 data_time: 0.151575 memory: 9504 loss_kpt: 0.000701 acc_pose: 0.865332 loss: 0.000701 2022/09/26 13:47:45 - mmengine - INFO - Epoch(train) [59][200/293] lr: 5.000000e-04 eta: 6:27:48 time: 0.591986 data_time: 0.086153 memory: 9504 loss_kpt: 0.000691 acc_pose: 0.830580 loss: 0.000691 2022/09/26 13:48:14 - mmengine - INFO - Epoch(train) [59][250/293] lr: 5.000000e-04 eta: 6:27:30 time: 0.586245 data_time: 0.144922 memory: 9504 loss_kpt: 0.000708 acc_pose: 0.836505 loss: 0.000708 2022/09/26 13:48:39 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:49:11 - mmengine - INFO - Epoch(train) [60][50/293] lr: 5.000000e-04 eta: 6:25:57 time: 0.631385 data_time: 0.241010 memory: 9504 loss_kpt: 0.000691 acc_pose: 0.815496 loss: 0.000691 2022/09/26 13:49:41 - mmengine - INFO - Epoch(train) [60][100/293] lr: 5.000000e-04 eta: 6:25:40 time: 0.594808 data_time: 0.234034 memory: 9504 loss_kpt: 0.000699 acc_pose: 0.781645 loss: 0.000699 2022/09/26 13:50:10 - mmengine - INFO - Epoch(train) [60][150/293] lr: 5.000000e-04 eta: 6:25:20 time: 0.575658 data_time: 0.217916 memory: 9504 loss_kpt: 0.000682 acc_pose: 0.841057 loss: 0.000682 2022/09/26 13:50:39 - mmengine - INFO - Epoch(train) [60][200/293] lr: 5.000000e-04 eta: 6:25:02 time: 0.590088 data_time: 0.080613 memory: 9504 loss_kpt: 0.000694 acc_pose: 0.746485 loss: 0.000694 2022/09/26 13:51:08 - mmengine - INFO - Epoch(train) [60][250/293] lr: 5.000000e-04 eta: 6:24:42 time: 0.569940 data_time: 0.082012 memory: 9504 loss_kpt: 0.000704 acc_pose: 0.803246 loss: 0.000704 2022/09/26 13:51:32 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:51:32 - mmengine - INFO - Saving checkpoint at 60 epochs 2022/09/26 13:51:59 - mmengine - INFO - Epoch(val) [60][50/407] eta: 0:02:00 time: 0.337599 data_time: 0.151708 memory: 9504 2022/09/26 13:52:15 - mmengine - INFO - Epoch(val) [60][100/407] eta: 0:01:40 time: 0.326332 data_time: 0.173879 memory: 1378 2022/09/26 13:52:31 - mmengine - INFO - Epoch(val) [60][150/407] eta: 0:01:22 time: 0.321530 data_time: 0.144982 memory: 1378 2022/09/26 13:52:47 - mmengine - INFO - Epoch(val) [60][200/407] eta: 0:01:05 time: 0.314254 data_time: 0.139883 memory: 1378 2022/09/26 13:53:03 - mmengine - INFO - Epoch(val) [60][250/407] eta: 0:00:48 time: 0.311072 data_time: 0.141288 memory: 1378 2022/09/26 13:53:19 - mmengine - INFO - Epoch(val) [60][300/407] eta: 0:00:34 time: 0.321157 data_time: 0.161093 memory: 1378 2022/09/26 13:53:35 - mmengine - INFO - Epoch(val) [60][350/407] eta: 0:00:18 time: 0.322709 data_time: 0.142458 memory: 1378 2022/09/26 13:53:44 - mmengine - INFO - Epoch(val) [60][400/407] eta: 0:00:01 time: 0.176629 data_time: 0.084355 memory: 1378 2022/09/26 13:54:16 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 13:54:29 - mmengine - INFO - Epoch(val) [60][407/407] coco/AP: 0.690639 coco/AP .5: 0.888741 coco/AP .75: 0.771455 coco/AP (M): 0.656877 coco/AP (L): 0.754247 coco/AR: 0.748615 coco/AR .5: 0.929786 coco/AR .75: 0.821788 coco/AR (M): 0.707566 coco/AR (L): 0.807804 2022/09/26 13:54:29 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_50.pth is removed 2022/09/26 13:54:32 - mmengine - INFO - The best checkpoint with 0.6906 coco/AP at 60 epoch is saved to best_coco/AP_epoch_60.pth. 2022/09/26 13:55:00 - mmengine - INFO - Epoch(train) [61][50/293] lr: 5.000000e-04 eta: 6:23:01 time: 0.561498 data_time: 0.151714 memory: 9504 loss_kpt: 0.000684 acc_pose: 0.862234 loss: 0.000684 2022/09/26 13:55:29 - mmengine - INFO - Epoch(train) [61][100/293] lr: 5.000000e-04 eta: 6:22:41 time: 0.569569 data_time: 0.083191 memory: 9504 loss_kpt: 0.000687 acc_pose: 0.799607 loss: 0.000687 2022/09/26 13:55:58 - mmengine - INFO - Epoch(train) [61][150/293] lr: 5.000000e-04 eta: 6:22:23 time: 0.591594 data_time: 0.093351 memory: 9504 loss_kpt: 0.000693 acc_pose: 0.756819 loss: 0.000693 2022/09/26 13:56:27 - mmengine - INFO - Epoch(train) [61][200/293] lr: 5.000000e-04 eta: 6:22:04 time: 0.584180 data_time: 0.099113 memory: 9504 loss_kpt: 0.000695 acc_pose: 0.809564 loss: 0.000695 2022/09/26 13:56:56 - mmengine - INFO - Epoch(train) [61][250/293] lr: 5.000000e-04 eta: 6:21:44 time: 0.573123 data_time: 0.115958 memory: 9504 loss_kpt: 0.000694 acc_pose: 0.820311 loss: 0.000694 2022/09/26 13:57:20 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:57:52 - mmengine - INFO - Epoch(train) [62][50/293] lr: 5.000000e-04 eta: 6:20:14 time: 0.635648 data_time: 0.108872 memory: 9504 loss_kpt: 0.000695 acc_pose: 0.801067 loss: 0.000695 2022/09/26 13:58:21 - mmengine - INFO - Epoch(train) [62][100/293] lr: 5.000000e-04 eta: 6:19:54 time: 0.575029 data_time: 0.137694 memory: 9504 loss_kpt: 0.000705 acc_pose: 0.853572 loss: 0.000705 2022/09/26 13:58:36 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 13:58:50 - mmengine - INFO - Epoch(train) [62][150/293] lr: 5.000000e-04 eta: 6:19:35 time: 0.583394 data_time: 0.188404 memory: 9504 loss_kpt: 0.000696 acc_pose: 0.797274 loss: 0.000696 2022/09/26 13:59:20 - mmengine - INFO - Epoch(train) [62][200/293] lr: 5.000000e-04 eta: 6:19:17 time: 0.591160 data_time: 0.103622 memory: 9504 loss_kpt: 0.000679 acc_pose: 0.807873 loss: 0.000679 2022/09/26 13:59:49 - mmengine - INFO - Epoch(train) [62][250/293] lr: 5.000000e-04 eta: 6:18:58 time: 0.582779 data_time: 0.078659 memory: 9504 loss_kpt: 0.000688 acc_pose: 0.777224 loss: 0.000688 2022/09/26 14:00:14 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:00:46 - mmengine - INFO - Epoch(train) [63][50/293] lr: 5.000000e-04 eta: 6:17:28 time: 0.622727 data_time: 0.147464 memory: 9504 loss_kpt: 0.000694 acc_pose: 0.823778 loss: 0.000694 2022/09/26 14:01:16 - mmengine - INFO - Epoch(train) [63][100/293] lr: 5.000000e-04 eta: 6:17:11 time: 0.603423 data_time: 0.083093 memory: 9504 loss_kpt: 0.000687 acc_pose: 0.793791 loss: 0.000687 2022/09/26 14:01:46 - mmengine - INFO - Epoch(train) [63][150/293] lr: 5.000000e-04 eta: 6:16:56 time: 0.613736 data_time: 0.132117 memory: 9504 loss_kpt: 0.000679 acc_pose: 0.803683 loss: 0.000679 2022/09/26 14:02:16 - mmengine - INFO - Epoch(train) [63][200/293] lr: 5.000000e-04 eta: 6:16:37 time: 0.590644 data_time: 0.124038 memory: 9504 loss_kpt: 0.000687 acc_pose: 0.785994 loss: 0.000687 2022/09/26 14:02:45 - mmengine - INFO - Epoch(train) [63][250/293] lr: 5.000000e-04 eta: 6:16:17 time: 0.578049 data_time: 0.148483 memory: 9504 loss_kpt: 0.000699 acc_pose: 0.795037 loss: 0.000699 2022/09/26 14:03:09 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:03:40 - mmengine - INFO - Epoch(train) [64][50/293] lr: 5.000000e-04 eta: 6:14:46 time: 0.606612 data_time: 0.132639 memory: 9504 loss_kpt: 0.000683 acc_pose: 0.838793 loss: 0.000683 2022/09/26 14:04:09 - mmengine - INFO - Epoch(train) [64][100/293] lr: 5.000000e-04 eta: 6:14:26 time: 0.575795 data_time: 0.096994 memory: 9504 loss_kpt: 0.000689 acc_pose: 0.728595 loss: 0.000689 2022/09/26 14:04:37 - mmengine - INFO - Epoch(train) [64][150/293] lr: 5.000000e-04 eta: 6:14:06 time: 0.573132 data_time: 0.077654 memory: 9504 loss_kpt: 0.000700 acc_pose: 0.788002 loss: 0.000700 2022/09/26 14:05:06 - mmengine - INFO - Epoch(train) [64][200/293] lr: 5.000000e-04 eta: 6:13:46 time: 0.582525 data_time: 0.082557 memory: 9504 loss_kpt: 0.000681 acc_pose: 0.818611 loss: 0.000681 2022/09/26 14:05:36 - mmengine - INFO - Epoch(train) [64][250/293] lr: 5.000000e-04 eta: 6:13:28 time: 0.586130 data_time: 0.102933 memory: 9504 loss_kpt: 0.000698 acc_pose: 0.821976 loss: 0.000698 2022/09/26 14:06:01 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:06:31 - mmengine - INFO - Epoch(train) [65][50/293] lr: 5.000000e-04 eta: 6:11:58 time: 0.611066 data_time: 0.124138 memory: 9504 loss_kpt: 0.000687 acc_pose: 0.816446 loss: 0.000687 2022/09/26 14:07:01 - mmengine - INFO - Epoch(train) [65][100/293] lr: 5.000000e-04 eta: 6:11:39 time: 0.592290 data_time: 0.089718 memory: 9504 loss_kpt: 0.000671 acc_pose: 0.812457 loss: 0.000671 2022/09/26 14:07:30 - mmengine - INFO - Epoch(train) [65][150/293] lr: 5.000000e-04 eta: 6:11:20 time: 0.580620 data_time: 0.087294 memory: 9504 loss_kpt: 0.000677 acc_pose: 0.715313 loss: 0.000677 2022/09/26 14:08:01 - mmengine - INFO - Epoch(train) [65][200/293] lr: 5.000000e-04 eta: 6:11:04 time: 0.613428 data_time: 0.207751 memory: 9504 loss_kpt: 0.000672 acc_pose: 0.836432 loss: 0.000672 2022/09/26 14:08:28 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:08:29 - mmengine - INFO - Epoch(train) [65][250/293] lr: 5.000000e-04 eta: 6:10:43 time: 0.571150 data_time: 0.096326 memory: 9504 loss_kpt: 0.000673 acc_pose: 0.813998 loss: 0.000673 2022/09/26 14:08:54 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:09:25 - mmengine - INFO - Epoch(train) [66][50/293] lr: 5.000000e-04 eta: 6:09:15 time: 0.615238 data_time: 0.111761 memory: 9504 loss_kpt: 0.000679 acc_pose: 0.781838 loss: 0.000679 2022/09/26 14:09:55 - mmengine - INFO - Epoch(train) [66][100/293] lr: 5.000000e-04 eta: 6:08:56 time: 0.588911 data_time: 0.084653 memory: 9504 loss_kpt: 0.000689 acc_pose: 0.799432 loss: 0.000689 2022/09/26 14:10:23 - mmengine - INFO - Epoch(train) [66][150/293] lr: 5.000000e-04 eta: 6:08:36 time: 0.577575 data_time: 0.156931 memory: 9504 loss_kpt: 0.000696 acc_pose: 0.807708 loss: 0.000696 2022/09/26 14:10:52 - mmengine - INFO - Epoch(train) [66][200/293] lr: 5.000000e-04 eta: 6:08:16 time: 0.578396 data_time: 0.097080 memory: 9504 loss_kpt: 0.000686 acc_pose: 0.870222 loss: 0.000686 2022/09/26 14:11:21 - mmengine - INFO - Epoch(train) [66][250/293] lr: 5.000000e-04 eta: 6:07:56 time: 0.574092 data_time: 0.240244 memory: 9504 loss_kpt: 0.000677 acc_pose: 0.819707 loss: 0.000677 2022/09/26 14:11:46 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:12:17 - mmengine - INFO - Epoch(train) [67][50/293] lr: 5.000000e-04 eta: 6:06:28 time: 0.609987 data_time: 0.101647 memory: 9504 loss_kpt: 0.000682 acc_pose: 0.871262 loss: 0.000682 2022/09/26 14:12:47 - mmengine - INFO - Epoch(train) [67][100/293] lr: 5.000000e-04 eta: 6:06:10 time: 0.599526 data_time: 0.096875 memory: 9504 loss_kpt: 0.000677 acc_pose: 0.811262 loss: 0.000677 2022/09/26 14:13:16 - mmengine - INFO - Epoch(train) [67][150/293] lr: 5.000000e-04 eta: 6:05:51 time: 0.587414 data_time: 0.189158 memory: 9504 loss_kpt: 0.000697 acc_pose: 0.812328 loss: 0.000697 2022/09/26 14:13:45 - mmengine - INFO - Epoch(train) [67][200/293] lr: 5.000000e-04 eta: 6:05:31 time: 0.582843 data_time: 0.145074 memory: 9504 loss_kpt: 0.000697 acc_pose: 0.795645 loss: 0.000697 2022/09/26 14:14:15 - mmengine - INFO - Epoch(train) [67][250/293] lr: 5.000000e-04 eta: 6:05:13 time: 0.595434 data_time: 0.108025 memory: 9504 loss_kpt: 0.000682 acc_pose: 0.854789 loss: 0.000682 2022/09/26 14:14:41 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:15:11 - mmengine - INFO - Epoch(train) [68][50/293] lr: 5.000000e-04 eta: 6:03:45 time: 0.604382 data_time: 0.237215 memory: 9504 loss_kpt: 0.000685 acc_pose: 0.851474 loss: 0.000685 2022/09/26 14:15:41 - mmengine - INFO - Epoch(train) [68][100/293] lr: 5.000000e-04 eta: 6:03:28 time: 0.605795 data_time: 0.122539 memory: 9504 loss_kpt: 0.000671 acc_pose: 0.848101 loss: 0.000671 2022/09/26 14:16:10 - mmengine - INFO - Epoch(train) [68][150/293] lr: 5.000000e-04 eta: 6:03:09 time: 0.586574 data_time: 0.083642 memory: 9504 loss_kpt: 0.000680 acc_pose: 0.811939 loss: 0.000680 2022/09/26 14:16:40 - mmengine - INFO - Epoch(train) [68][200/293] lr: 5.000000e-04 eta: 6:02:49 time: 0.580864 data_time: 0.109399 memory: 9504 loss_kpt: 0.000678 acc_pose: 0.787760 loss: 0.000678 2022/09/26 14:17:09 - mmengine - INFO - Epoch(train) [68][250/293] lr: 5.000000e-04 eta: 6:02:30 time: 0.584146 data_time: 0.132778 memory: 9504 loss_kpt: 0.000682 acc_pose: 0.799393 loss: 0.000682 2022/09/26 14:17:34 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:18:04 - mmengine - INFO - Epoch(train) [69][50/293] lr: 5.000000e-04 eta: 6:01:03 time: 0.607789 data_time: 0.251080 memory: 9504 loss_kpt: 0.000683 acc_pose: 0.826172 loss: 0.000683 2022/09/26 14:18:20 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:18:33 - mmengine - INFO - Epoch(train) [69][100/293] lr: 5.000000e-04 eta: 6:00:43 time: 0.581590 data_time: 0.134910 memory: 9504 loss_kpt: 0.000693 acc_pose: 0.747133 loss: 0.000693 2022/09/26 14:19:04 - mmengine - INFO - Epoch(train) [69][150/293] lr: 5.000000e-04 eta: 6:00:26 time: 0.603265 data_time: 0.075835 memory: 9504 loss_kpt: 0.000672 acc_pose: 0.830665 loss: 0.000672 2022/09/26 14:19:32 - mmengine - INFO - Epoch(train) [69][200/293] lr: 5.000000e-04 eta: 6:00:05 time: 0.576017 data_time: 0.112227 memory: 9504 loss_kpt: 0.000679 acc_pose: 0.785235 loss: 0.000679 2022/09/26 14:20:01 - mmengine - INFO - Epoch(train) [69][250/293] lr: 5.000000e-04 eta: 5:59:44 time: 0.567414 data_time: 0.159208 memory: 9504 loss_kpt: 0.000672 acc_pose: 0.839269 loss: 0.000672 2022/09/26 14:20:26 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:20:55 - mmengine - INFO - Epoch(train) [70][50/293] lr: 5.000000e-04 eta: 5:58:15 time: 0.575861 data_time: 0.144891 memory: 9504 loss_kpt: 0.000683 acc_pose: 0.803030 loss: 0.000683 2022/09/26 14:21:26 - mmengine - INFO - Epoch(train) [70][100/293] lr: 5.000000e-04 eta: 5:57:59 time: 0.618093 data_time: 0.086376 memory: 9504 loss_kpt: 0.000687 acc_pose: 0.778249 loss: 0.000687 2022/09/26 14:21:56 - mmengine - INFO - Epoch(train) [70][150/293] lr: 5.000000e-04 eta: 5:57:41 time: 0.606481 data_time: 0.108559 memory: 9504 loss_kpt: 0.000679 acc_pose: 0.790745 loss: 0.000679 2022/09/26 14:22:26 - mmengine - INFO - Epoch(train) [70][200/293] lr: 5.000000e-04 eta: 5:57:22 time: 0.592030 data_time: 0.084791 memory: 9504 loss_kpt: 0.000704 acc_pose: 0.816225 loss: 0.000704 2022/09/26 14:22:54 - mmengine - INFO - Epoch(train) [70][250/293] lr: 5.000000e-04 eta: 5:57:00 time: 0.554729 data_time: 0.079737 memory: 9504 loss_kpt: 0.000677 acc_pose: 0.805262 loss: 0.000677 2022/09/26 14:23:18 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:23:18 - mmengine - INFO - Saving checkpoint at 70 epochs 2022/09/26 14:23:43 - mmengine - INFO - Epoch(val) [70][50/407] eta: 0:01:55 time: 0.324101 data_time: 0.135053 memory: 9504 2022/09/26 14:23:59 - mmengine - INFO - Epoch(val) [70][100/407] eta: 0:01:35 time: 0.310484 data_time: 0.137951 memory: 1378 2022/09/26 14:24:15 - mmengine - INFO - Epoch(val) [70][150/407] eta: 0:01:23 time: 0.325039 data_time: 0.132111 memory: 1378 2022/09/26 14:24:31 - mmengine - INFO - Epoch(val) [70][200/407] eta: 0:01:07 time: 0.326713 data_time: 0.132557 memory: 1378 2022/09/26 14:24:49 - mmengine - INFO - Epoch(val) [70][250/407] eta: 0:00:54 time: 0.345492 data_time: 0.181495 memory: 1378 2022/09/26 14:25:03 - mmengine - INFO - Epoch(val) [70][300/407] eta: 0:00:31 time: 0.298306 data_time: 0.139825 memory: 1378 2022/09/26 14:25:19 - mmengine - INFO - Epoch(val) [70][350/407] eta: 0:00:17 time: 0.310746 data_time: 0.159921 memory: 1378 2022/09/26 14:25:29 - mmengine - INFO - Epoch(val) [70][400/407] eta: 0:00:01 time: 0.191676 data_time: 0.080760 memory: 1378 2022/09/26 14:26:01 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 14:26:15 - mmengine - INFO - Epoch(val) [70][407/407] coco/AP: 0.695915 coco/AP .5: 0.890063 coco/AP .75: 0.774618 coco/AP (M): 0.660215 coco/AP (L): 0.763003 coco/AR: 0.754880 coco/AR .5: 0.930573 coco/AR .75: 0.824622 coco/AR (M): 0.711718 coco/AR (L): 0.816797 2022/09/26 14:26:15 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_60.pth is removed 2022/09/26 14:26:17 - mmengine - INFO - The best checkpoint with 0.6959 coco/AP at 70 epoch is saved to best_coco/AP_epoch_70.pth. 2022/09/26 14:26:46 - mmengine - INFO - Epoch(train) [71][50/293] lr: 5.000000e-04 eta: 5:55:31 time: 0.569622 data_time: 0.203836 memory: 9504 loss_kpt: 0.000690 acc_pose: 0.813952 loss: 0.000690 2022/09/26 14:27:15 - mmengine - INFO - Epoch(train) [71][100/293] lr: 5.000000e-04 eta: 5:55:10 time: 0.571136 data_time: 0.137540 memory: 9504 loss_kpt: 0.000676 acc_pose: 0.806927 loss: 0.000676 2022/09/26 14:27:43 - mmengine - INFO - Epoch(train) [71][150/293] lr: 5.000000e-04 eta: 5:54:49 time: 0.573168 data_time: 0.085209 memory: 9504 loss_kpt: 0.000680 acc_pose: 0.819087 loss: 0.000680 2022/09/26 14:28:12 - mmengine - INFO - Epoch(train) [71][200/293] lr: 5.000000e-04 eta: 5:54:30 time: 0.584839 data_time: 0.086478 memory: 9504 loss_kpt: 0.000676 acc_pose: 0.827461 loss: 0.000676 2022/09/26 14:28:41 - mmengine - INFO - Epoch(train) [71][250/293] lr: 5.000000e-04 eta: 5:54:09 time: 0.580400 data_time: 0.079832 memory: 9504 loss_kpt: 0.000681 acc_pose: 0.852964 loss: 0.000681 2022/09/26 14:29:07 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:29:37 - mmengine - INFO - Epoch(train) [72][50/293] lr: 5.000000e-04 eta: 5:52:45 time: 0.601687 data_time: 0.150853 memory: 9504 loss_kpt: 0.000672 acc_pose: 0.829534 loss: 0.000672 2022/09/26 14:30:06 - mmengine - INFO - Epoch(train) [72][100/293] lr: 5.000000e-04 eta: 5:52:25 time: 0.580790 data_time: 0.070252 memory: 9504 loss_kpt: 0.000673 acc_pose: 0.832430 loss: 0.000673 2022/09/26 14:30:35 - mmengine - INFO - Epoch(train) [72][150/293] lr: 5.000000e-04 eta: 5:52:05 time: 0.582901 data_time: 0.130008 memory: 9504 loss_kpt: 0.000674 acc_pose: 0.776340 loss: 0.000674 2022/09/26 14:31:02 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:31:04 - mmengine - INFO - Epoch(train) [72][200/293] lr: 5.000000e-04 eta: 5:51:45 time: 0.579518 data_time: 0.083741 memory: 9504 loss_kpt: 0.000682 acc_pose: 0.801646 loss: 0.000682 2022/09/26 14:31:36 - mmengine - INFO - Epoch(train) [72][250/293] lr: 5.000000e-04 eta: 5:51:30 time: 0.638021 data_time: 0.133655 memory: 9504 loss_kpt: 0.000683 acc_pose: 0.761102 loss: 0.000683 2022/09/26 14:32:00 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:32:31 - mmengine - INFO - Epoch(train) [73][50/293] lr: 5.000000e-04 eta: 5:50:07 time: 0.609826 data_time: 0.106426 memory: 9504 loss_kpt: 0.000669 acc_pose: 0.790461 loss: 0.000669 2022/09/26 14:33:00 - mmengine - INFO - Epoch(train) [73][100/293] lr: 5.000000e-04 eta: 5:49:47 time: 0.585236 data_time: 0.185706 memory: 9504 loss_kpt: 0.000662 acc_pose: 0.798580 loss: 0.000662 2022/09/26 14:33:30 - mmengine - INFO - Epoch(train) [73][150/293] lr: 5.000000e-04 eta: 5:49:28 time: 0.588214 data_time: 0.106655 memory: 9504 loss_kpt: 0.000661 acc_pose: 0.799685 loss: 0.000661 2022/09/26 14:33:58 - mmengine - INFO - Epoch(train) [73][200/293] lr: 5.000000e-04 eta: 5:49:06 time: 0.562539 data_time: 0.139771 memory: 9504 loss_kpt: 0.000687 acc_pose: 0.784272 loss: 0.000687 2022/09/26 14:34:26 - mmengine - INFO - Epoch(train) [73][250/293] lr: 5.000000e-04 eta: 5:48:44 time: 0.572484 data_time: 0.100927 memory: 9504 loss_kpt: 0.000677 acc_pose: 0.825619 loss: 0.000677 2022/09/26 14:34:52 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:35:23 - mmengine - INFO - Epoch(train) [74][50/293] lr: 5.000000e-04 eta: 5:47:22 time: 0.607666 data_time: 0.099622 memory: 9504 loss_kpt: 0.000673 acc_pose: 0.811827 loss: 0.000673 2022/09/26 14:35:52 - mmengine - INFO - Epoch(train) [74][100/293] lr: 5.000000e-04 eta: 5:47:03 time: 0.586404 data_time: 0.137881 memory: 9504 loss_kpt: 0.000666 acc_pose: 0.782749 loss: 0.000666 2022/09/26 14:36:22 - mmengine - INFO - Epoch(train) [74][150/293] lr: 5.000000e-04 eta: 5:46:44 time: 0.600846 data_time: 0.144479 memory: 9504 loss_kpt: 0.000670 acc_pose: 0.851959 loss: 0.000670 2022/09/26 14:36:52 - mmengine - INFO - Epoch(train) [74][200/293] lr: 5.000000e-04 eta: 5:46:25 time: 0.593281 data_time: 0.114106 memory: 9504 loss_kpt: 0.000672 acc_pose: 0.791353 loss: 0.000672 2022/09/26 14:37:20 - mmengine - INFO - Epoch(train) [74][250/293] lr: 5.000000e-04 eta: 5:46:03 time: 0.562881 data_time: 0.129247 memory: 9504 loss_kpt: 0.000679 acc_pose: 0.789671 loss: 0.000679 2022/09/26 14:37:45 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:38:15 - mmengine - INFO - Epoch(train) [75][50/293] lr: 5.000000e-04 eta: 5:44:42 time: 0.611185 data_time: 0.099006 memory: 9504 loss_kpt: 0.000673 acc_pose: 0.786460 loss: 0.000673 2022/09/26 14:38:44 - mmengine - INFO - Epoch(train) [75][100/293] lr: 5.000000e-04 eta: 5:44:21 time: 0.582035 data_time: 0.111648 memory: 9504 loss_kpt: 0.000678 acc_pose: 0.797909 loss: 0.000678 2022/09/26 14:39:14 - mmengine - INFO - Epoch(train) [75][150/293] lr: 5.000000e-04 eta: 5:44:01 time: 0.584453 data_time: 0.107637 memory: 9504 loss_kpt: 0.000671 acc_pose: 0.798633 loss: 0.000671 2022/09/26 14:39:43 - mmengine - INFO - Epoch(train) [75][200/293] lr: 5.000000e-04 eta: 5:43:42 time: 0.591080 data_time: 0.099348 memory: 9504 loss_kpt: 0.000677 acc_pose: 0.802846 loss: 0.000677 2022/09/26 14:40:13 - mmengine - INFO - Epoch(train) [75][250/293] lr: 5.000000e-04 eta: 5:43:21 time: 0.584586 data_time: 0.088785 memory: 9504 loss_kpt: 0.000674 acc_pose: 0.792016 loss: 0.000674 2022/09/26 14:40:38 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:40:53 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:41:08 - mmengine - INFO - Epoch(train) [76][50/293] lr: 5.000000e-04 eta: 5:42:00 time: 0.595029 data_time: 0.194122 memory: 9504 loss_kpt: 0.000690 acc_pose: 0.837617 loss: 0.000690 2022/09/26 14:41:36 - mmengine - INFO - Epoch(train) [76][100/293] lr: 5.000000e-04 eta: 5:41:38 time: 0.570661 data_time: 0.169107 memory: 9504 loss_kpt: 0.000666 acc_pose: 0.802963 loss: 0.000666 2022/09/26 14:42:07 - mmengine - INFO - Epoch(train) [76][150/293] lr: 5.000000e-04 eta: 5:41:22 time: 0.626076 data_time: 0.128921 memory: 9504 loss_kpt: 0.000671 acc_pose: 0.795317 loss: 0.000671 2022/09/26 14:42:37 - mmengine - INFO - Epoch(train) [76][200/293] lr: 5.000000e-04 eta: 5:41:02 time: 0.591379 data_time: 0.093858 memory: 9504 loss_kpt: 0.000663 acc_pose: 0.829480 loss: 0.000663 2022/09/26 14:43:07 - mmengine - INFO - Epoch(train) [76][250/293] lr: 5.000000e-04 eta: 5:40:42 time: 0.591323 data_time: 0.081050 memory: 9504 loss_kpt: 0.000680 acc_pose: 0.845598 loss: 0.000680 2022/09/26 14:43:31 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:44:01 - mmengine - INFO - Epoch(train) [77][50/293] lr: 5.000000e-04 eta: 5:39:22 time: 0.597101 data_time: 0.121355 memory: 9504 loss_kpt: 0.000684 acc_pose: 0.779421 loss: 0.000684 2022/09/26 14:44:29 - mmengine - INFO - Epoch(train) [77][100/293] lr: 5.000000e-04 eta: 5:39:01 time: 0.577488 data_time: 0.076000 memory: 9504 loss_kpt: 0.000686 acc_pose: 0.800030 loss: 0.000686 2022/09/26 14:44:59 - mmengine - INFO - Epoch(train) [77][150/293] lr: 5.000000e-04 eta: 5:38:41 time: 0.589215 data_time: 0.193596 memory: 9504 loss_kpt: 0.000673 acc_pose: 0.762325 loss: 0.000673 2022/09/26 14:45:28 - mmengine - INFO - Epoch(train) [77][200/293] lr: 5.000000e-04 eta: 5:38:20 time: 0.583919 data_time: 0.103025 memory: 9504 loss_kpt: 0.000672 acc_pose: 0.764906 loss: 0.000672 2022/09/26 14:45:58 - mmengine - INFO - Epoch(train) [77][250/293] lr: 5.000000e-04 eta: 5:38:01 time: 0.592540 data_time: 0.089472 memory: 9504 loss_kpt: 0.000678 acc_pose: 0.769465 loss: 0.000678 2022/09/26 14:46:22 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:46:52 - mmengine - INFO - Epoch(train) [78][50/293] lr: 5.000000e-04 eta: 5:36:40 time: 0.594083 data_time: 0.136263 memory: 9504 loss_kpt: 0.000660 acc_pose: 0.832677 loss: 0.000660 2022/09/26 14:47:20 - mmengine - INFO - Epoch(train) [78][100/293] lr: 5.000000e-04 eta: 5:36:19 time: 0.571095 data_time: 0.080481 memory: 9504 loss_kpt: 0.000660 acc_pose: 0.813954 loss: 0.000660 2022/09/26 14:47:50 - mmengine - INFO - Epoch(train) [78][150/293] lr: 5.000000e-04 eta: 5:35:59 time: 0.593502 data_time: 0.235337 memory: 9504 loss_kpt: 0.000665 acc_pose: 0.837690 loss: 0.000665 2022/09/26 14:48:18 - mmengine - INFO - Epoch(train) [78][200/293] lr: 5.000000e-04 eta: 5:35:37 time: 0.568311 data_time: 0.227747 memory: 9504 loss_kpt: 0.000673 acc_pose: 0.836868 loss: 0.000673 2022/09/26 14:48:46 - mmengine - INFO - Epoch(train) [78][250/293] lr: 5.000000e-04 eta: 5:35:15 time: 0.563444 data_time: 0.158338 memory: 9504 loss_kpt: 0.000653 acc_pose: 0.837306 loss: 0.000653 2022/09/26 14:49:12 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:49:42 - mmengine - INFO - Epoch(train) [79][50/293] lr: 5.000000e-04 eta: 5:33:56 time: 0.604307 data_time: 0.105754 memory: 9504 loss_kpt: 0.000681 acc_pose: 0.779823 loss: 0.000681 2022/09/26 14:50:11 - mmengine - INFO - Epoch(train) [79][100/293] lr: 5.000000e-04 eta: 5:33:35 time: 0.575439 data_time: 0.080531 memory: 9504 loss_kpt: 0.000670 acc_pose: 0.798455 loss: 0.000670 2022/09/26 14:50:39 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:50:41 - mmengine - INFO - Epoch(train) [79][150/293] lr: 5.000000e-04 eta: 5:33:15 time: 0.592615 data_time: 0.181700 memory: 9504 loss_kpt: 0.000672 acc_pose: 0.763910 loss: 0.000672 2022/09/26 14:51:11 - mmengine - INFO - Epoch(train) [79][200/293] lr: 5.000000e-04 eta: 5:32:56 time: 0.598243 data_time: 0.172055 memory: 9504 loss_kpt: 0.000680 acc_pose: 0.771245 loss: 0.000680 2022/09/26 14:51:39 - mmengine - INFO - Epoch(train) [79][250/293] lr: 5.000000e-04 eta: 5:32:35 time: 0.572941 data_time: 0.072131 memory: 9504 loss_kpt: 0.000663 acc_pose: 0.784115 loss: 0.000663 2022/09/26 14:52:04 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:52:35 - mmengine - INFO - Epoch(train) [80][50/293] lr: 5.000000e-04 eta: 5:31:18 time: 0.620974 data_time: 0.095021 memory: 9504 loss_kpt: 0.000670 acc_pose: 0.811317 loss: 0.000670 2022/09/26 14:53:04 - mmengine - INFO - Epoch(train) [80][100/293] lr: 5.000000e-04 eta: 5:30:57 time: 0.579760 data_time: 0.094291 memory: 9504 loss_kpt: 0.000658 acc_pose: 0.870052 loss: 0.000658 2022/09/26 14:53:33 - mmengine - INFO - Epoch(train) [80][150/293] lr: 5.000000e-04 eta: 5:30:35 time: 0.573770 data_time: 0.171184 memory: 9504 loss_kpt: 0.000651 acc_pose: 0.804457 loss: 0.000651 2022/09/26 14:54:02 - mmengine - INFO - Epoch(train) [80][200/293] lr: 5.000000e-04 eta: 5:30:16 time: 0.597409 data_time: 0.126731 memory: 9504 loss_kpt: 0.000666 acc_pose: 0.751769 loss: 0.000666 2022/09/26 14:54:32 - mmengine - INFO - Epoch(train) [80][250/293] lr: 5.000000e-04 eta: 5:29:55 time: 0.582084 data_time: 0.075816 memory: 9504 loss_kpt: 0.000667 acc_pose: 0.843744 loss: 0.000667 2022/09/26 14:54:57 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 14:54:57 - mmengine - INFO - Saving checkpoint at 80 epochs 2022/09/26 14:55:20 - mmengine - INFO - Epoch(val) [80][50/407] eta: 0:01:55 time: 0.324526 data_time: 0.153259 memory: 9504 2022/09/26 14:55:36 - mmengine - INFO - Epoch(val) [80][100/407] eta: 0:01:36 time: 0.313134 data_time: 0.138297 memory: 1378 2022/09/26 14:55:52 - mmengine - INFO - Epoch(val) [80][150/407] eta: 0:01:22 time: 0.319453 data_time: 0.126985 memory: 1378 2022/09/26 14:56:07 - mmengine - INFO - Epoch(val) [80][200/407] eta: 0:01:02 time: 0.300324 data_time: 0.129844 memory: 1378 2022/09/26 14:56:23 - mmengine - INFO - Epoch(val) [80][250/407] eta: 0:00:49 time: 0.313594 data_time: 0.157491 memory: 1378 2022/09/26 14:56:38 - mmengine - INFO - Epoch(val) [80][300/407] eta: 0:00:33 time: 0.309837 data_time: 0.137327 memory: 1378 2022/09/26 14:56:56 - mmengine - INFO - Epoch(val) [80][350/407] eta: 0:00:20 time: 0.356554 data_time: 0.178531 memory: 1378 2022/09/26 14:57:07 - mmengine - INFO - Epoch(val) [80][400/407] eta: 0:00:01 time: 0.223899 data_time: 0.095510 memory: 1378 2022/09/26 14:57:40 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 14:57:53 - mmengine - INFO - Epoch(val) [80][407/407] coco/AP: 0.697875 coco/AP .5: 0.887894 coco/AP .75: 0.779056 coco/AP (M): 0.664519 coco/AP (L): 0.762304 coco/AR: 0.756864 coco/AR .5: 0.929628 coco/AR .75: 0.829503 coco/AR (M): 0.715624 coco/AR (L): 0.816388 2022/09/26 14:57:53 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_70.pth is removed 2022/09/26 14:57:55 - mmengine - INFO - The best checkpoint with 0.6979 coco/AP at 80 epoch is saved to best_coco/AP_epoch_80.pth. 2022/09/26 14:58:24 - mmengine - INFO - Epoch(train) [81][50/293] lr: 5.000000e-04 eta: 5:28:35 time: 0.569872 data_time: 0.166998 memory: 9504 loss_kpt: 0.000667 acc_pose: 0.792188 loss: 0.000667 2022/09/26 14:58:54 - mmengine - INFO - Epoch(train) [81][100/293] lr: 5.000000e-04 eta: 5:28:15 time: 0.597582 data_time: 0.084864 memory: 9504 loss_kpt: 0.000669 acc_pose: 0.792145 loss: 0.000669 2022/09/26 14:59:23 - mmengine - INFO - Epoch(train) [81][150/293] lr: 5.000000e-04 eta: 5:27:55 time: 0.584468 data_time: 0.116516 memory: 9504 loss_kpt: 0.000673 acc_pose: 0.804168 loss: 0.000673 2022/09/26 14:59:52 - mmengine - INFO - Epoch(train) [81][200/293] lr: 5.000000e-04 eta: 5:27:33 time: 0.577278 data_time: 0.097690 memory: 9504 loss_kpt: 0.000650 acc_pose: 0.798736 loss: 0.000650 2022/09/26 15:00:22 - mmengine - INFO - Epoch(train) [81][250/293] lr: 5.000000e-04 eta: 5:27:14 time: 0.601411 data_time: 0.085841 memory: 9504 loss_kpt: 0.000657 acc_pose: 0.759929 loss: 0.000657 2022/09/26 15:00:46 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:01:17 - mmengine - INFO - Epoch(train) [82][50/293] lr: 5.000000e-04 eta: 5:25:58 time: 0.610117 data_time: 0.195331 memory: 9504 loss_kpt: 0.000659 acc_pose: 0.784044 loss: 0.000659 2022/09/26 15:01:45 - mmengine - INFO - Epoch(train) [82][100/293] lr: 5.000000e-04 eta: 5:25:36 time: 0.565448 data_time: 0.076097 memory: 9504 loss_kpt: 0.000656 acc_pose: 0.811013 loss: 0.000656 2022/09/26 15:02:15 - mmengine - INFO - Epoch(train) [82][150/293] lr: 5.000000e-04 eta: 5:25:16 time: 0.600583 data_time: 0.116015 memory: 9504 loss_kpt: 0.000663 acc_pose: 0.777235 loss: 0.000663 2022/09/26 15:02:44 - mmengine - INFO - Epoch(train) [82][200/293] lr: 5.000000e-04 eta: 5:24:54 time: 0.570669 data_time: 0.204257 memory: 9504 loss_kpt: 0.000667 acc_pose: 0.801152 loss: 0.000667 2022/09/26 15:03:13 - mmengine - INFO - Epoch(train) [82][250/293] lr: 5.000000e-04 eta: 5:24:34 time: 0.589925 data_time: 0.119257 memory: 9504 loss_kpt: 0.000659 acc_pose: 0.825168 loss: 0.000659 2022/09/26 15:03:23 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:03:38 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:04:07 - mmengine - INFO - Epoch(train) [83][50/293] lr: 5.000000e-04 eta: 5:23:16 time: 0.585146 data_time: 0.090854 memory: 9504 loss_kpt: 0.000651 acc_pose: 0.782386 loss: 0.000651 2022/09/26 15:04:36 - mmengine - INFO - Epoch(train) [83][100/293] lr: 5.000000e-04 eta: 5:22:54 time: 0.569874 data_time: 0.128892 memory: 9504 loss_kpt: 0.000667 acc_pose: 0.817973 loss: 0.000667 2022/09/26 15:05:05 - mmengine - INFO - Epoch(train) [83][150/293] lr: 5.000000e-04 eta: 5:22:34 time: 0.590778 data_time: 0.139446 memory: 9504 loss_kpt: 0.000665 acc_pose: 0.779558 loss: 0.000665 2022/09/26 15:05:35 - mmengine - INFO - Epoch(train) [83][200/293] lr: 5.000000e-04 eta: 5:22:13 time: 0.584383 data_time: 0.164993 memory: 9504 loss_kpt: 0.000665 acc_pose: 0.832685 loss: 0.000665 2022/09/26 15:06:03 - mmengine - INFO - Epoch(train) [83][250/293] lr: 5.000000e-04 eta: 5:21:52 time: 0.576086 data_time: 0.124563 memory: 9504 loss_kpt: 0.000660 acc_pose: 0.840165 loss: 0.000660 2022/09/26 15:06:28 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:06:58 - mmengine - INFO - Epoch(train) [84][50/293] lr: 5.000000e-04 eta: 5:20:36 time: 0.596929 data_time: 0.117834 memory: 9504 loss_kpt: 0.000670 acc_pose: 0.856817 loss: 0.000670 2022/09/26 15:07:27 - mmengine - INFO - Epoch(train) [84][100/293] lr: 5.000000e-04 eta: 5:20:14 time: 0.569440 data_time: 0.136633 memory: 9504 loss_kpt: 0.000650 acc_pose: 0.777766 loss: 0.000650 2022/09/26 15:07:55 - mmengine - INFO - Epoch(train) [84][150/293] lr: 5.000000e-04 eta: 5:19:51 time: 0.567278 data_time: 0.191079 memory: 9504 loss_kpt: 0.000653 acc_pose: 0.806691 loss: 0.000653 2022/09/26 15:08:23 - mmengine - INFO - Epoch(train) [84][200/293] lr: 5.000000e-04 eta: 5:19:29 time: 0.564959 data_time: 0.100153 memory: 9504 loss_kpt: 0.000650 acc_pose: 0.816538 loss: 0.000650 2022/09/26 15:08:53 - mmengine - INFO - Epoch(train) [84][250/293] lr: 5.000000e-04 eta: 5:19:09 time: 0.590027 data_time: 0.094786 memory: 9504 loss_kpt: 0.000660 acc_pose: 0.846607 loss: 0.000660 2022/09/26 15:09:18 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:09:48 - mmengine - INFO - Epoch(train) [85][50/293] lr: 5.000000e-04 eta: 5:17:54 time: 0.607295 data_time: 0.115192 memory: 9504 loss_kpt: 0.000655 acc_pose: 0.837408 loss: 0.000655 2022/09/26 15:10:18 - mmengine - INFO - Epoch(train) [85][100/293] lr: 5.000000e-04 eta: 5:17:34 time: 0.604368 data_time: 0.171393 memory: 9504 loss_kpt: 0.000666 acc_pose: 0.823591 loss: 0.000666 2022/09/26 15:10:47 - mmengine - INFO - Epoch(train) [85][150/293] lr: 5.000000e-04 eta: 5:17:13 time: 0.573986 data_time: 0.082443 memory: 9504 loss_kpt: 0.000663 acc_pose: 0.821537 loss: 0.000663 2022/09/26 15:11:16 - mmengine - INFO - Epoch(train) [85][200/293] lr: 5.000000e-04 eta: 5:16:52 time: 0.583661 data_time: 0.185577 memory: 9504 loss_kpt: 0.000657 acc_pose: 0.763117 loss: 0.000657 2022/09/26 15:11:45 - mmengine - INFO - Epoch(train) [85][250/293] lr: 5.000000e-04 eta: 5:16:30 time: 0.568645 data_time: 0.102736 memory: 9504 loss_kpt: 0.000672 acc_pose: 0.832781 loss: 0.000672 2022/09/26 15:12:09 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:12:39 - mmengine - INFO - Epoch(train) [86][50/293] lr: 5.000000e-04 eta: 5:15:15 time: 0.606014 data_time: 0.096178 memory: 9504 loss_kpt: 0.000658 acc_pose: 0.830564 loss: 0.000658 2022/09/26 15:13:05 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:13:08 - mmengine - INFO - Epoch(train) [86][100/293] lr: 5.000000e-04 eta: 5:14:53 time: 0.568928 data_time: 0.067004 memory: 9504 loss_kpt: 0.000655 acc_pose: 0.812395 loss: 0.000655 2022/09/26 15:13:37 - mmengine - INFO - Epoch(train) [86][150/293] lr: 5.000000e-04 eta: 5:14:32 time: 0.587542 data_time: 0.089795 memory: 9504 loss_kpt: 0.000664 acc_pose: 0.792492 loss: 0.000664 2022/09/26 15:14:06 - mmengine - INFO - Epoch(train) [86][200/293] lr: 5.000000e-04 eta: 5:14:10 time: 0.571652 data_time: 0.119639 memory: 9504 loss_kpt: 0.000656 acc_pose: 0.809817 loss: 0.000656 2022/09/26 15:14:36 - mmengine - INFO - Epoch(train) [86][250/293] lr: 5.000000e-04 eta: 5:13:51 time: 0.603830 data_time: 0.095155 memory: 9504 loss_kpt: 0.000663 acc_pose: 0.832424 loss: 0.000663 2022/09/26 15:15:00 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:15:31 - mmengine - INFO - Epoch(train) [87][50/293] lr: 5.000000e-04 eta: 5:12:37 time: 0.609303 data_time: 0.095624 memory: 9504 loss_kpt: 0.000663 acc_pose: 0.826049 loss: 0.000663 2022/09/26 15:15:59 - mmengine - INFO - Epoch(train) [87][100/293] lr: 5.000000e-04 eta: 5:12:15 time: 0.562621 data_time: 0.076422 memory: 9504 loss_kpt: 0.000652 acc_pose: 0.826159 loss: 0.000652 2022/09/26 15:16:27 - mmengine - INFO - Epoch(train) [87][150/293] lr: 5.000000e-04 eta: 5:11:53 time: 0.576260 data_time: 0.101711 memory: 9504 loss_kpt: 0.000661 acc_pose: 0.828073 loss: 0.000661 2022/09/26 15:16:57 - mmengine - INFO - Epoch(train) [87][200/293] lr: 5.000000e-04 eta: 5:11:32 time: 0.590736 data_time: 0.115270 memory: 9504 loss_kpt: 0.000663 acc_pose: 0.815779 loss: 0.000663 2022/09/26 15:17:26 - mmengine - INFO - Epoch(train) [87][250/293] lr: 5.000000e-04 eta: 5:11:11 time: 0.578087 data_time: 0.079024 memory: 9504 loss_kpt: 0.000668 acc_pose: 0.850727 loss: 0.000668 2022/09/26 15:17:50 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:18:21 - mmengine - INFO - Epoch(train) [88][50/293] lr: 5.000000e-04 eta: 5:09:58 time: 0.610205 data_time: 0.191148 memory: 9504 loss_kpt: 0.000661 acc_pose: 0.789611 loss: 0.000661 2022/09/26 15:18:51 - mmengine - INFO - Epoch(train) [88][100/293] lr: 5.000000e-04 eta: 5:09:38 time: 0.606765 data_time: 0.153611 memory: 9504 loss_kpt: 0.000665 acc_pose: 0.799828 loss: 0.000665 2022/09/26 15:19:21 - mmengine - INFO - Epoch(train) [88][150/293] lr: 5.000000e-04 eta: 5:09:18 time: 0.587279 data_time: 0.161637 memory: 9504 loss_kpt: 0.000649 acc_pose: 0.831788 loss: 0.000649 2022/09/26 15:19:49 - mmengine - INFO - Epoch(train) [88][200/293] lr: 5.000000e-04 eta: 5:08:55 time: 0.570752 data_time: 0.091673 memory: 9504 loss_kpt: 0.000668 acc_pose: 0.823490 loss: 0.000668 2022/09/26 15:20:18 - mmengine - INFO - Epoch(train) [88][250/293] lr: 5.000000e-04 eta: 5:08:33 time: 0.574124 data_time: 0.104960 memory: 9504 loss_kpt: 0.000664 acc_pose: 0.843350 loss: 0.000664 2022/09/26 15:20:42 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:21:13 - mmengine - INFO - Epoch(train) [89][50/293] lr: 5.000000e-04 eta: 5:07:22 time: 0.616859 data_time: 0.136253 memory: 9504 loss_kpt: 0.000652 acc_pose: 0.806914 loss: 0.000652 2022/09/26 15:21:42 - mmengine - INFO - Epoch(train) [89][100/293] lr: 5.000000e-04 eta: 5:07:00 time: 0.584114 data_time: 0.078202 memory: 9504 loss_kpt: 0.000659 acc_pose: 0.837756 loss: 0.000659 2022/09/26 15:22:11 - mmengine - INFO - Epoch(train) [89][150/293] lr: 5.000000e-04 eta: 5:06:39 time: 0.582420 data_time: 0.083199 memory: 9504 loss_kpt: 0.000667 acc_pose: 0.773922 loss: 0.000667 2022/09/26 15:22:42 - mmengine - INFO - Epoch(train) [89][200/293] lr: 5.000000e-04 eta: 5:06:19 time: 0.602476 data_time: 0.092958 memory: 9504 loss_kpt: 0.000655 acc_pose: 0.805031 loss: 0.000655 2022/09/26 15:22:51 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:23:11 - mmengine - INFO - Epoch(train) [89][250/293] lr: 5.000000e-04 eta: 5:05:58 time: 0.582670 data_time: 0.067890 memory: 9504 loss_kpt: 0.000663 acc_pose: 0.812442 loss: 0.000663 2022/09/26 15:23:36 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:24:06 - mmengine - INFO - Epoch(train) [90][50/293] lr: 5.000000e-04 eta: 5:04:45 time: 0.597268 data_time: 0.131963 memory: 9504 loss_kpt: 0.000672 acc_pose: 0.834829 loss: 0.000672 2022/09/26 15:24:33 - mmengine - INFO - Epoch(train) [90][100/293] lr: 5.000000e-04 eta: 5:04:21 time: 0.541154 data_time: 0.199995 memory: 9504 loss_kpt: 0.000660 acc_pose: 0.797063 loss: 0.000660 2022/09/26 15:25:02 - mmengine - INFO - Epoch(train) [90][150/293] lr: 5.000000e-04 eta: 5:03:59 time: 0.577705 data_time: 0.101370 memory: 9504 loss_kpt: 0.000663 acc_pose: 0.801527 loss: 0.000663 2022/09/26 15:25:32 - mmengine - INFO - Epoch(train) [90][200/293] lr: 5.000000e-04 eta: 5:03:39 time: 0.598529 data_time: 0.090936 memory: 9504 loss_kpt: 0.000647 acc_pose: 0.786593 loss: 0.000647 2022/09/26 15:26:01 - mmengine - INFO - Epoch(train) [90][250/293] lr: 5.000000e-04 eta: 5:03:18 time: 0.592447 data_time: 0.175333 memory: 9504 loss_kpt: 0.000670 acc_pose: 0.816413 loss: 0.000670 2022/09/26 15:26:26 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:26:26 - mmengine - INFO - Saving checkpoint at 90 epochs 2022/09/26 15:26:49 - mmengine - INFO - Epoch(val) [90][50/407] eta: 0:01:55 time: 0.323264 data_time: 0.167867 memory: 9504 2022/09/26 15:27:05 - mmengine - INFO - Epoch(val) [90][100/407] eta: 0:01:35 time: 0.310195 data_time: 0.129004 memory: 1378 2022/09/26 15:27:20 - mmengine - INFO - Epoch(val) [90][150/407] eta: 0:01:17 time: 0.300136 data_time: 0.130954 memory: 1378 2022/09/26 15:27:36 - mmengine - INFO - Epoch(val) [90][200/407] eta: 0:01:05 time: 0.315608 data_time: 0.156027 memory: 1378 2022/09/26 15:27:51 - mmengine - INFO - Epoch(val) [90][250/407] eta: 0:00:47 time: 0.300439 data_time: 0.136436 memory: 1378 2022/09/26 15:28:07 - mmengine - INFO - Epoch(val) [90][300/407] eta: 0:00:33 time: 0.315259 data_time: 0.148747 memory: 1378 2022/09/26 15:28:22 - mmengine - INFO - Epoch(val) [90][350/407] eta: 0:00:17 time: 0.310662 data_time: 0.145633 memory: 1378 2022/09/26 15:28:35 - mmengine - INFO - Epoch(val) [90][400/407] eta: 0:00:01 time: 0.250138 data_time: 0.100103 memory: 1378 2022/09/26 15:29:07 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 15:29:20 - mmengine - INFO - Epoch(val) [90][407/407] coco/AP: 0.700514 coco/AP .5: 0.892244 coco/AP .75: 0.775439 coco/AP (M): 0.664852 coco/AP (L): 0.765757 coco/AR: 0.757950 coco/AR .5: 0.930888 coco/AR .75: 0.823835 coco/AR (M): 0.715351 coco/AR (L): 0.818803 2022/09/26 15:29:20 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_80.pth is removed 2022/09/26 15:29:23 - mmengine - INFO - The best checkpoint with 0.7005 coco/AP at 90 epoch is saved to best_coco/AP_epoch_90.pth. 2022/09/26 15:29:53 - mmengine - INFO - Epoch(train) [91][50/293] lr: 5.000000e-04 eta: 5:02:06 time: 0.596968 data_time: 0.209629 memory: 9504 loss_kpt: 0.000661 acc_pose: 0.746724 loss: 0.000661 2022/09/26 15:30:22 - mmengine - INFO - Epoch(train) [91][100/293] lr: 5.000000e-04 eta: 5:01:44 time: 0.581085 data_time: 0.135076 memory: 9504 loss_kpt: 0.000650 acc_pose: 0.816581 loss: 0.000650 2022/09/26 15:30:50 - mmengine - INFO - Epoch(train) [91][150/293] lr: 5.000000e-04 eta: 5:01:20 time: 0.550621 data_time: 0.190293 memory: 9504 loss_kpt: 0.000671 acc_pose: 0.787095 loss: 0.000671 2022/09/26 15:31:19 - mmengine - INFO - Epoch(train) [91][200/293] lr: 5.000000e-04 eta: 5:01:00 time: 0.595833 data_time: 0.152260 memory: 9504 loss_kpt: 0.000657 acc_pose: 0.846194 loss: 0.000657 2022/09/26 15:31:49 - mmengine - INFO - Epoch(train) [91][250/293] lr: 5.000000e-04 eta: 5:00:39 time: 0.591146 data_time: 0.079015 memory: 9504 loss_kpt: 0.000651 acc_pose: 0.748657 loss: 0.000651 2022/09/26 15:32:14 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:32:44 - mmengine - INFO - Epoch(train) [92][50/293] lr: 5.000000e-04 eta: 4:59:27 time: 0.599167 data_time: 0.107327 memory: 9504 loss_kpt: 0.000658 acc_pose: 0.792876 loss: 0.000658 2022/09/26 15:33:14 - mmengine - INFO - Epoch(train) [92][100/293] lr: 5.000000e-04 eta: 4:59:06 time: 0.584769 data_time: 0.154773 memory: 9504 loss_kpt: 0.000638 acc_pose: 0.806878 loss: 0.000638 2022/09/26 15:33:44 - mmengine - INFO - Epoch(train) [92][150/293] lr: 5.000000e-04 eta: 4:58:46 time: 0.606474 data_time: 0.154767 memory: 9504 loss_kpt: 0.000654 acc_pose: 0.855857 loss: 0.000654 2022/09/26 15:34:13 - mmengine - INFO - Epoch(train) [92][200/293] lr: 5.000000e-04 eta: 4:58:25 time: 0.587754 data_time: 0.073115 memory: 9504 loss_kpt: 0.000651 acc_pose: 0.851564 loss: 0.000651 2022/09/26 15:34:44 - mmengine - INFO - Epoch(train) [92][250/293] lr: 5.000000e-04 eta: 4:58:05 time: 0.604245 data_time: 0.085488 memory: 9504 loss_kpt: 0.000652 acc_pose: 0.827465 loss: 0.000652 2022/09/26 15:35:09 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:35:37 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:35:41 - mmengine - INFO - Epoch(train) [93][50/293] lr: 5.000000e-04 eta: 4:56:56 time: 0.632247 data_time: 0.091906 memory: 9504 loss_kpt: 0.000650 acc_pose: 0.794508 loss: 0.000650 2022/09/26 15:36:09 - mmengine - INFO - Epoch(train) [93][100/293] lr: 5.000000e-04 eta: 4:56:33 time: 0.567310 data_time: 0.081502 memory: 9504 loss_kpt: 0.000653 acc_pose: 0.840972 loss: 0.000653 2022/09/26 15:36:38 - mmengine - INFO - Epoch(train) [93][150/293] lr: 5.000000e-04 eta: 4:56:10 time: 0.566773 data_time: 0.166486 memory: 9504 loss_kpt: 0.000651 acc_pose: 0.837667 loss: 0.000651 2022/09/26 15:37:08 - mmengine - INFO - Epoch(train) [93][200/293] lr: 5.000000e-04 eta: 4:55:50 time: 0.603333 data_time: 0.122470 memory: 9504 loss_kpt: 0.000651 acc_pose: 0.834455 loss: 0.000651 2022/09/26 15:37:38 - mmengine - INFO - Epoch(train) [93][250/293] lr: 5.000000e-04 eta: 4:55:30 time: 0.599872 data_time: 0.083311 memory: 9504 loss_kpt: 0.000660 acc_pose: 0.830339 loss: 0.000660 2022/09/26 15:38:02 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:38:31 - mmengine - INFO - Epoch(train) [94][50/293] lr: 5.000000e-04 eta: 4:54:18 time: 0.588117 data_time: 0.111888 memory: 9504 loss_kpt: 0.000652 acc_pose: 0.821099 loss: 0.000652 2022/09/26 15:39:00 - mmengine - INFO - Epoch(train) [94][100/293] lr: 5.000000e-04 eta: 4:53:57 time: 0.584668 data_time: 0.149486 memory: 9504 loss_kpt: 0.000648 acc_pose: 0.789648 loss: 0.000648 2022/09/26 15:39:29 - mmengine - INFO - Epoch(train) [94][150/293] lr: 5.000000e-04 eta: 4:53:34 time: 0.573936 data_time: 0.220189 memory: 9504 loss_kpt: 0.000639 acc_pose: 0.798350 loss: 0.000639 2022/09/26 15:39:59 - mmengine - INFO - Epoch(train) [94][200/293] lr: 5.000000e-04 eta: 4:53:13 time: 0.592882 data_time: 0.069504 memory: 9504 loss_kpt: 0.000655 acc_pose: 0.808864 loss: 0.000655 2022/09/26 15:40:28 - mmengine - INFO - Epoch(train) [94][250/293] lr: 5.000000e-04 eta: 4:52:52 time: 0.580738 data_time: 0.079020 memory: 9504 loss_kpt: 0.000641 acc_pose: 0.853893 loss: 0.000641 2022/09/26 15:40:53 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:41:23 - mmengine - INFO - Epoch(train) [95][50/293] lr: 5.000000e-04 eta: 4:51:41 time: 0.596955 data_time: 0.203757 memory: 9504 loss_kpt: 0.000659 acc_pose: 0.829779 loss: 0.000659 2022/09/26 15:41:51 - mmengine - INFO - Epoch(train) [95][100/293] lr: 5.000000e-04 eta: 4:51:19 time: 0.571092 data_time: 0.135597 memory: 9504 loss_kpt: 0.000650 acc_pose: 0.852403 loss: 0.000650 2022/09/26 15:42:20 - mmengine - INFO - Epoch(train) [95][150/293] lr: 5.000000e-04 eta: 4:50:57 time: 0.575355 data_time: 0.116376 memory: 9504 loss_kpt: 0.000651 acc_pose: 0.839654 loss: 0.000651 2022/09/26 15:42:49 - mmengine - INFO - Epoch(train) [95][200/293] lr: 5.000000e-04 eta: 4:50:35 time: 0.588453 data_time: 0.081401 memory: 9504 loss_kpt: 0.000644 acc_pose: 0.861033 loss: 0.000644 2022/09/26 15:43:18 - mmengine - INFO - Epoch(train) [95][250/293] lr: 5.000000e-04 eta: 4:50:13 time: 0.572228 data_time: 0.072997 memory: 9504 loss_kpt: 0.000657 acc_pose: 0.868553 loss: 0.000657 2022/09/26 15:43:41 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:44:11 - mmengine - INFO - Epoch(train) [96][50/293] lr: 5.000000e-04 eta: 4:49:03 time: 0.596365 data_time: 0.092709 memory: 9504 loss_kpt: 0.000657 acc_pose: 0.813326 loss: 0.000657 2022/09/26 15:44:41 - mmengine - INFO - Epoch(train) [96][100/293] lr: 5.000000e-04 eta: 4:48:42 time: 0.592406 data_time: 0.088239 memory: 9504 loss_kpt: 0.000661 acc_pose: 0.783478 loss: 0.000661 2022/09/26 15:45:10 - mmengine - INFO - Epoch(train) [96][150/293] lr: 5.000000e-04 eta: 4:48:19 time: 0.571296 data_time: 0.083984 memory: 9504 loss_kpt: 0.000651 acc_pose: 0.858297 loss: 0.000651 2022/09/26 15:45:18 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:45:38 - mmengine - INFO - Epoch(train) [96][200/293] lr: 5.000000e-04 eta: 4:47:57 time: 0.569489 data_time: 0.088577 memory: 9504 loss_kpt: 0.000636 acc_pose: 0.810050 loss: 0.000636 2022/09/26 15:46:06 - mmengine - INFO - Epoch(train) [96][250/293] lr: 5.000000e-04 eta: 4:47:34 time: 0.559829 data_time: 0.083780 memory: 9504 loss_kpt: 0.000660 acc_pose: 0.800092 loss: 0.000660 2022/09/26 15:46:31 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:47:00 - mmengine - INFO - Epoch(train) [97][50/293] lr: 5.000000e-04 eta: 4:46:23 time: 0.579909 data_time: 0.148055 memory: 9504 loss_kpt: 0.000649 acc_pose: 0.848018 loss: 0.000649 2022/09/26 15:47:29 - mmengine - INFO - Epoch(train) [97][100/293] lr: 5.000000e-04 eta: 4:46:01 time: 0.580905 data_time: 0.103442 memory: 9504 loss_kpt: 0.000647 acc_pose: 0.847879 loss: 0.000647 2022/09/26 15:47:59 - mmengine - INFO - Epoch(train) [97][150/293] lr: 5.000000e-04 eta: 4:45:40 time: 0.595580 data_time: 0.094933 memory: 9504 loss_kpt: 0.000651 acc_pose: 0.867507 loss: 0.000651 2022/09/26 15:48:28 - mmengine - INFO - Epoch(train) [97][200/293] lr: 5.000000e-04 eta: 4:45:18 time: 0.578408 data_time: 0.103529 memory: 9504 loss_kpt: 0.000657 acc_pose: 0.872479 loss: 0.000657 2022/09/26 15:48:57 - mmengine - INFO - Epoch(train) [97][250/293] lr: 5.000000e-04 eta: 4:44:56 time: 0.576498 data_time: 0.085187 memory: 9504 loss_kpt: 0.000643 acc_pose: 0.821327 loss: 0.000643 2022/09/26 15:49:21 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:49:52 - mmengine - INFO - Epoch(train) [98][50/293] lr: 5.000000e-04 eta: 4:43:48 time: 0.618733 data_time: 0.093931 memory: 9504 loss_kpt: 0.000636 acc_pose: 0.802275 loss: 0.000636 2022/09/26 15:50:23 - mmengine - INFO - Epoch(train) [98][100/293] lr: 5.000000e-04 eta: 4:43:28 time: 0.611281 data_time: 0.078952 memory: 9504 loss_kpt: 0.000664 acc_pose: 0.827886 loss: 0.000664 2022/09/26 15:50:53 - mmengine - INFO - Epoch(train) [98][150/293] lr: 5.000000e-04 eta: 4:43:07 time: 0.599615 data_time: 0.077363 memory: 9504 loss_kpt: 0.000656 acc_pose: 0.827797 loss: 0.000656 2022/09/26 15:51:22 - mmengine - INFO - Epoch(train) [98][200/293] lr: 5.000000e-04 eta: 4:42:45 time: 0.573968 data_time: 0.075591 memory: 9504 loss_kpt: 0.000635 acc_pose: 0.808991 loss: 0.000635 2022/09/26 15:51:49 - mmengine - INFO - Epoch(train) [98][250/293] lr: 5.000000e-04 eta: 4:42:21 time: 0.547320 data_time: 0.204325 memory: 9504 loss_kpt: 0.000641 acc_pose: 0.854685 loss: 0.000641 2022/09/26 15:52:14 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:52:43 - mmengine - INFO - Epoch(train) [99][50/293] lr: 5.000000e-04 eta: 4:41:11 time: 0.576307 data_time: 0.097033 memory: 9504 loss_kpt: 0.000637 acc_pose: 0.812700 loss: 0.000637 2022/09/26 15:53:12 - mmengine - INFO - Epoch(train) [99][100/293] lr: 5.000000e-04 eta: 4:40:49 time: 0.575104 data_time: 0.076904 memory: 9504 loss_kpt: 0.000643 acc_pose: 0.842061 loss: 0.000643 2022/09/26 15:53:40 - mmengine - INFO - Epoch(train) [99][150/293] lr: 5.000000e-04 eta: 4:40:26 time: 0.564193 data_time: 0.080578 memory: 9504 loss_kpt: 0.000654 acc_pose: 0.795037 loss: 0.000654 2022/09/26 15:54:10 - mmengine - INFO - Epoch(train) [99][200/293] lr: 5.000000e-04 eta: 4:40:05 time: 0.608896 data_time: 0.077863 memory: 9504 loss_kpt: 0.000657 acc_pose: 0.814201 loss: 0.000657 2022/09/26 15:54:38 - mmengine - INFO - Epoch(train) [99][250/293] lr: 5.000000e-04 eta: 4:39:42 time: 0.558573 data_time: 0.163092 memory: 9504 loss_kpt: 0.000641 acc_pose: 0.814624 loss: 0.000641 2022/09/26 15:54:59 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:55:02 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:55:32 - mmengine - INFO - Epoch(train) [100][50/293] lr: 5.000000e-04 eta: 4:38:33 time: 0.586778 data_time: 0.108271 memory: 9504 loss_kpt: 0.000636 acc_pose: 0.845110 loss: 0.000636 2022/09/26 15:56:00 - mmengine - INFO - Epoch(train) [100][100/293] lr: 5.000000e-04 eta: 4:38:11 time: 0.571182 data_time: 0.086330 memory: 9504 loss_kpt: 0.000649 acc_pose: 0.855723 loss: 0.000649 2022/09/26 15:56:30 - mmengine - INFO - Epoch(train) [100][150/293] lr: 5.000000e-04 eta: 4:37:49 time: 0.586181 data_time: 0.089056 memory: 9504 loss_kpt: 0.000647 acc_pose: 0.799906 loss: 0.000647 2022/09/26 15:56:59 - mmengine - INFO - Epoch(train) [100][200/293] lr: 5.000000e-04 eta: 4:37:26 time: 0.575165 data_time: 0.167028 memory: 9504 loss_kpt: 0.000638 acc_pose: 0.785464 loss: 0.000638 2022/09/26 15:57:27 - mmengine - INFO - Epoch(train) [100][250/293] lr: 5.000000e-04 eta: 4:37:04 time: 0.576582 data_time: 0.119378 memory: 9504 loss_kpt: 0.000639 acc_pose: 0.833206 loss: 0.000639 2022/09/26 15:57:51 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 15:57:51 - mmengine - INFO - Saving checkpoint at 100 epochs 2022/09/26 15:58:17 - mmengine - INFO - Epoch(val) [100][50/407] eta: 0:01:59 time: 0.335774 data_time: 0.154296 memory: 9504 2022/09/26 15:58:32 - mmengine - INFO - Epoch(val) [100][100/407] eta: 0:01:35 time: 0.309715 data_time: 0.142903 memory: 1378 2022/09/26 15:58:48 - mmengine - INFO - Epoch(val) [100][150/407] eta: 0:01:22 time: 0.321897 data_time: 0.155507 memory: 1378 2022/09/26 15:59:03 - mmengine - INFO - Epoch(val) [100][200/407] eta: 0:01:01 time: 0.299170 data_time: 0.112175 memory: 1378 2022/09/26 15:59:19 - mmengine - INFO - Epoch(val) [100][250/407] eta: 0:00:48 time: 0.311905 data_time: 0.131424 memory: 1378 2022/09/26 15:59:34 - mmengine - INFO - Epoch(val) [100][300/407] eta: 0:00:33 time: 0.313268 data_time: 0.136189 memory: 1378 2022/09/26 15:59:51 - mmengine - INFO - Epoch(val) [100][350/407] eta: 0:00:19 time: 0.333452 data_time: 0.148139 memory: 1378 2022/09/26 16:00:01 - mmengine - INFO - Epoch(val) [100][400/407] eta: 0:00:01 time: 0.197492 data_time: 0.105891 memory: 1378 2022/09/26 16:00:33 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 16:00:46 - mmengine - INFO - Epoch(val) [100][407/407] coco/AP: 0.705583 coco/AP .5: 0.894152 coco/AP .75: 0.785985 coco/AP (M): 0.671877 coco/AP (L): 0.770032 coco/AR: 0.762846 coco/AR .5: 0.932777 coco/AR .75: 0.835013 coco/AR (M): 0.721388 coco/AR (L): 0.822928 2022/09/26 16:00:46 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_90.pth is removed 2022/09/26 16:00:49 - mmengine - INFO - The best checkpoint with 0.7056 coco/AP at 100 epoch is saved to best_coco/AP_epoch_100.pth. 2022/09/26 16:01:17 - mmengine - INFO - Epoch(train) [101][50/293] lr: 5.000000e-04 eta: 4:35:54 time: 0.557408 data_time: 0.211426 memory: 9504 loss_kpt: 0.000646 acc_pose: 0.824005 loss: 0.000646 2022/09/26 16:01:45 - mmengine - INFO - Epoch(train) [101][100/293] lr: 5.000000e-04 eta: 4:35:31 time: 0.566423 data_time: 0.120706 memory: 9504 loss_kpt: 0.000645 acc_pose: 0.775898 loss: 0.000645 2022/09/26 16:02:13 - mmengine - INFO - Epoch(train) [101][150/293] lr: 5.000000e-04 eta: 4:35:08 time: 0.557243 data_time: 0.083868 memory: 9504 loss_kpt: 0.000638 acc_pose: 0.832439 loss: 0.000638 2022/09/26 16:02:41 - mmengine - INFO - Epoch(train) [101][200/293] lr: 5.000000e-04 eta: 4:34:45 time: 0.564875 data_time: 0.082409 memory: 9504 loss_kpt: 0.000663 acc_pose: 0.817229 loss: 0.000663 2022/09/26 16:03:09 - mmengine - INFO - Epoch(train) [101][250/293] lr: 5.000000e-04 eta: 4:34:21 time: 0.554758 data_time: 0.083325 memory: 9504 loss_kpt: 0.000648 acc_pose: 0.826872 loss: 0.000648 2022/09/26 16:03:33 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:04:02 - mmengine - INFO - Epoch(train) [102][50/293] lr: 5.000000e-04 eta: 4:33:13 time: 0.578631 data_time: 0.093380 memory: 9504 loss_kpt: 0.000652 acc_pose: 0.828895 loss: 0.000652 2022/09/26 16:04:30 - mmengine - INFO - Epoch(train) [102][100/293] lr: 5.000000e-04 eta: 4:32:51 time: 0.575909 data_time: 0.080515 memory: 9504 loss_kpt: 0.000643 acc_pose: 0.793338 loss: 0.000643 2022/09/26 16:05:56 - mmengine - INFO - Epoch(train) [102][150/293] lr: 5.000000e-04 eta: 4:33:29 time: 1.715832 data_time: 0.285502 memory: 9504 loss_kpt: 0.000634 acc_pose: 0.816451 loss: 0.000634 2022/09/26 16:08:05 - mmengine - INFO - Epoch(train) [102][200/293] lr: 5.000000e-04 eta: 4:34:54 time: 2.583874 data_time: 0.582146 memory: 9504 loss_kpt: 0.000636 acc_pose: 0.817438 loss: 0.000636 2022/09/26 16:08:46 - mmengine - INFO - Epoch(train) [102][250/293] lr: 5.000000e-04 eta: 4:34:43 time: 0.802336 data_time: 0.163327 memory: 9504 loss_kpt: 0.000655 acc_pose: 0.824777 loss: 0.000655 2022/09/26 16:09:24 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:09:58 - mmengine - INFO - Epoch(train) [103][50/293] lr: 5.000000e-04 eta: 4:33:39 time: 0.678429 data_time: 0.114364 memory: 9504 loss_kpt: 0.000639 acc_pose: 0.859939 loss: 0.000639 2022/09/26 16:10:32 - mmengine - INFO - Epoch(train) [103][100/293] lr: 5.000000e-04 eta: 4:33:21 time: 0.672917 data_time: 0.107337 memory: 9504 loss_kpt: 0.000640 acc_pose: 0.840146 loss: 0.000640 2022/09/26 16:10:39 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:11:00 - mmengine - INFO - Epoch(train) [103][150/293] lr: 5.000000e-04 eta: 4:32:58 time: 0.569682 data_time: 0.101715 memory: 9504 loss_kpt: 0.000637 acc_pose: 0.868735 loss: 0.000637 2022/09/26 16:11:29 - mmengine - INFO - Epoch(train) [103][200/293] lr: 5.000000e-04 eta: 4:32:35 time: 0.586233 data_time: 0.126618 memory: 9504 loss_kpt: 0.000652 acc_pose: 0.796302 loss: 0.000652 2022/09/26 16:11:59 - mmengine - INFO - Epoch(train) [103][250/293] lr: 5.000000e-04 eta: 4:32:12 time: 0.585249 data_time: 0.126003 memory: 9504 loss_kpt: 0.000650 acc_pose: 0.832256 loss: 0.000650 2022/09/26 16:12:27 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:12:57 - mmengine - INFO - Epoch(train) [104][50/293] lr: 5.000000e-04 eta: 4:31:06 time: 0.614579 data_time: 0.144252 memory: 9504 loss_kpt: 0.000651 acc_pose: 0.824916 loss: 0.000651 2022/09/26 16:13:25 - mmengine - INFO - Epoch(train) [104][100/293] lr: 5.000000e-04 eta: 4:30:42 time: 0.556573 data_time: 0.130044 memory: 9504 loss_kpt: 0.000639 acc_pose: 0.863462 loss: 0.000639 2022/09/26 16:13:54 - mmengine - INFO - Epoch(train) [104][150/293] lr: 5.000000e-04 eta: 4:30:19 time: 0.575756 data_time: 0.142797 memory: 9504 loss_kpt: 0.000650 acc_pose: 0.803902 loss: 0.000650 2022/09/26 16:14:24 - mmengine - INFO - Epoch(train) [104][200/293] lr: 5.000000e-04 eta: 4:29:57 time: 0.603191 data_time: 0.103355 memory: 9504 loss_kpt: 0.000636 acc_pose: 0.867131 loss: 0.000636 2022/09/26 16:14:53 - mmengine - INFO - Epoch(train) [104][250/293] lr: 5.000000e-04 eta: 4:29:33 time: 0.566006 data_time: 0.116826 memory: 9504 loss_kpt: 0.000653 acc_pose: 0.778764 loss: 0.000653 2022/09/26 16:15:16 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:15:47 - mmengine - INFO - Epoch(train) [105][50/293] lr: 5.000000e-04 eta: 4:28:27 time: 0.606501 data_time: 0.096718 memory: 9504 loss_kpt: 0.000640 acc_pose: 0.795299 loss: 0.000640 2022/09/26 16:16:15 - mmengine - INFO - Epoch(train) [105][100/293] lr: 5.000000e-04 eta: 4:28:03 time: 0.564890 data_time: 0.082268 memory: 9504 loss_kpt: 0.000641 acc_pose: 0.876158 loss: 0.000641 2022/09/26 16:16:43 - mmengine - INFO - Epoch(train) [105][150/293] lr: 5.000000e-04 eta: 4:27:39 time: 0.562719 data_time: 0.129473 memory: 9504 loss_kpt: 0.000627 acc_pose: 0.831193 loss: 0.000627 2022/09/26 16:17:11 - mmengine - INFO - Epoch(train) [105][200/293] lr: 5.000000e-04 eta: 4:27:15 time: 0.552499 data_time: 0.127463 memory: 9504 loss_kpt: 0.000651 acc_pose: 0.761256 loss: 0.000651 2022/09/26 16:17:40 - mmengine - INFO - Epoch(train) [105][250/293] lr: 5.000000e-04 eta: 4:26:52 time: 0.574058 data_time: 0.133657 memory: 9504 loss_kpt: 0.000646 acc_pose: 0.847791 loss: 0.000646 2022/09/26 16:18:03 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:18:33 - mmengine - INFO - Epoch(train) [106][50/293] lr: 5.000000e-04 eta: 4:25:45 time: 0.600128 data_time: 0.153723 memory: 9504 loss_kpt: 0.000643 acc_pose: 0.865367 loss: 0.000643 2022/09/26 16:19:03 - mmengine - INFO - Epoch(train) [106][100/293] lr: 5.000000e-04 eta: 4:25:22 time: 0.583863 data_time: 0.156124 memory: 9504 loss_kpt: 0.000638 acc_pose: 0.834902 loss: 0.000638 2022/09/26 16:19:31 - mmengine - INFO - Epoch(train) [106][150/293] lr: 5.000000e-04 eta: 4:24:58 time: 0.563367 data_time: 0.076521 memory: 9504 loss_kpt: 0.000638 acc_pose: 0.851642 loss: 0.000638 2022/09/26 16:19:59 - mmengine - INFO - Epoch(train) [106][200/293] lr: 5.000000e-04 eta: 4:24:35 time: 0.570802 data_time: 0.086147 memory: 9504 loss_kpt: 0.000653 acc_pose: 0.810767 loss: 0.000653 2022/09/26 16:20:20 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:20:28 - mmengine - INFO - Epoch(train) [106][250/293] lr: 5.000000e-04 eta: 4:24:12 time: 0.577149 data_time: 0.174222 memory: 9504 loss_kpt: 0.000637 acc_pose: 0.842737 loss: 0.000637 2022/09/26 16:20:54 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:21:23 - mmengine - INFO - Epoch(train) [107][50/293] lr: 5.000000e-04 eta: 4:23:05 time: 0.590098 data_time: 0.086134 memory: 9504 loss_kpt: 0.000646 acc_pose: 0.845358 loss: 0.000646 2022/09/26 16:21:52 - mmengine - INFO - Epoch(train) [107][100/293] lr: 5.000000e-04 eta: 4:22:42 time: 0.571158 data_time: 0.130965 memory: 9504 loss_kpt: 0.000646 acc_pose: 0.796695 loss: 0.000646 2022/09/26 16:22:21 - mmengine - INFO - Epoch(train) [107][150/293] lr: 5.000000e-04 eta: 4:22:19 time: 0.580393 data_time: 0.109776 memory: 9504 loss_kpt: 0.000643 acc_pose: 0.820386 loss: 0.000643 2022/09/26 16:22:49 - mmengine - INFO - Epoch(train) [107][200/293] lr: 5.000000e-04 eta: 4:21:55 time: 0.567487 data_time: 0.100603 memory: 9504 loss_kpt: 0.000638 acc_pose: 0.841021 loss: 0.000638 2022/09/26 16:23:19 - mmengine - INFO - Epoch(train) [107][250/293] lr: 5.000000e-04 eta: 4:21:33 time: 0.593764 data_time: 0.115706 memory: 9504 loss_kpt: 0.000638 acc_pose: 0.825777 loss: 0.000638 2022/09/26 16:23:43 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:24:13 - mmengine - INFO - Epoch(train) [108][50/293] lr: 5.000000e-04 eta: 4:20:26 time: 0.585366 data_time: 0.121671 memory: 9504 loss_kpt: 0.000638 acc_pose: 0.798197 loss: 0.000638 2022/09/26 16:24:41 - mmengine - INFO - Epoch(train) [108][100/293] lr: 5.000000e-04 eta: 4:20:03 time: 0.568515 data_time: 0.107939 memory: 9504 loss_kpt: 0.000624 acc_pose: 0.795491 loss: 0.000624 2022/09/26 16:25:09 - mmengine - INFO - Epoch(train) [108][150/293] lr: 5.000000e-04 eta: 4:19:39 time: 0.555280 data_time: 0.094978 memory: 9504 loss_kpt: 0.000636 acc_pose: 0.789708 loss: 0.000636 2022/09/26 16:25:38 - mmengine - INFO - Epoch(train) [108][200/293] lr: 5.000000e-04 eta: 4:19:16 time: 0.583872 data_time: 0.126961 memory: 9504 loss_kpt: 0.000646 acc_pose: 0.809154 loss: 0.000646 2022/09/26 16:26:07 - mmengine - INFO - Epoch(train) [108][250/293] lr: 5.000000e-04 eta: 4:18:53 time: 0.581297 data_time: 0.079421 memory: 9504 loss_kpt: 0.000649 acc_pose: 0.792383 loss: 0.000649 2022/09/26 16:26:32 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:27:01 - mmengine - INFO - Epoch(train) [109][50/293] lr: 5.000000e-04 eta: 4:17:47 time: 0.594640 data_time: 0.170162 memory: 9504 loss_kpt: 0.000636 acc_pose: 0.780797 loss: 0.000636 2022/09/26 16:27:30 - mmengine - INFO - Epoch(train) [109][100/293] lr: 5.000000e-04 eta: 4:17:24 time: 0.573899 data_time: 0.089966 memory: 9504 loss_kpt: 0.000642 acc_pose: 0.868505 loss: 0.000642 2022/09/26 16:27:58 - mmengine - INFO - Epoch(train) [109][150/293] lr: 5.000000e-04 eta: 4:17:00 time: 0.566784 data_time: 0.081221 memory: 9504 loss_kpt: 0.000634 acc_pose: 0.858659 loss: 0.000634 2022/09/26 16:28:27 - mmengine - INFO - Epoch(train) [109][200/293] lr: 5.000000e-04 eta: 4:16:36 time: 0.570041 data_time: 0.112039 memory: 9504 loss_kpt: 0.000652 acc_pose: 0.856090 loss: 0.000652 2022/09/26 16:28:57 - mmengine - INFO - Epoch(train) [109][250/293] lr: 5.000000e-04 eta: 4:16:14 time: 0.596437 data_time: 0.164746 memory: 9504 loss_kpt: 0.000644 acc_pose: 0.841834 loss: 0.000644 2022/09/26 16:29:22 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:29:53 - mmengine - INFO - Epoch(train) [110][50/293] lr: 5.000000e-04 eta: 4:15:10 time: 0.612627 data_time: 0.102670 memory: 9504 loss_kpt: 0.000646 acc_pose: 0.880585 loss: 0.000646 2022/09/26 16:30:00 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:30:21 - mmengine - INFO - Epoch(train) [110][100/293] lr: 5.000000e-04 eta: 4:14:46 time: 0.566183 data_time: 0.088213 memory: 9504 loss_kpt: 0.000634 acc_pose: 0.814720 loss: 0.000634 2022/09/26 16:30:50 - mmengine - INFO - Epoch(train) [110][150/293] lr: 5.000000e-04 eta: 4:14:23 time: 0.573885 data_time: 0.097531 memory: 9504 loss_kpt: 0.000647 acc_pose: 0.824014 loss: 0.000647 2022/09/26 16:31:18 - mmengine - INFO - Epoch(train) [110][200/293] lr: 5.000000e-04 eta: 4:13:59 time: 0.563008 data_time: 0.101974 memory: 9504 loss_kpt: 0.000637 acc_pose: 0.839937 loss: 0.000637 2022/09/26 16:31:47 - mmengine - INFO - Epoch(train) [110][250/293] lr: 5.000000e-04 eta: 4:13:35 time: 0.575099 data_time: 0.120400 memory: 9504 loss_kpt: 0.000645 acc_pose: 0.865232 loss: 0.000645 2022/09/26 16:32:12 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:32:12 - mmengine - INFO - Saving checkpoint at 110 epochs 2022/09/26 16:32:36 - mmengine - INFO - Epoch(val) [110][50/407] eta: 0:01:59 time: 0.333962 data_time: 0.179627 memory: 9504 2022/09/26 16:32:52 - mmengine - INFO - Epoch(val) [110][100/407] eta: 0:01:33 time: 0.305784 data_time: 0.137134 memory: 1378 2022/09/26 16:33:08 - mmengine - INFO - Epoch(val) [110][150/407] eta: 0:01:24 time: 0.329309 data_time: 0.158245 memory: 1378 2022/09/26 16:33:23 - mmengine - INFO - Epoch(val) [110][200/407] eta: 0:01:02 time: 0.300393 data_time: 0.124547 memory: 1378 2022/09/26 16:33:38 - mmengine - INFO - Epoch(val) [110][250/407] eta: 0:00:45 time: 0.291963 data_time: 0.134385 memory: 1378 2022/09/26 16:33:54 - mmengine - INFO - Epoch(val) [110][300/407] eta: 0:00:34 time: 0.318624 data_time: 0.159004 memory: 1378 2022/09/26 16:34:11 - mmengine - INFO - Epoch(val) [110][350/407] eta: 0:00:20 time: 0.358000 data_time: 0.150275 memory: 1378 2022/09/26 16:34:24 - mmengine - INFO - Epoch(val) [110][400/407] eta: 0:00:01 time: 0.247750 data_time: 0.105124 memory: 1378 2022/09/26 16:34:56 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 16:35:09 - mmengine - INFO - Epoch(val) [110][407/407] coco/AP: 0.706090 coco/AP .5: 0.892122 coco/AP .75: 0.787013 coco/AP (M): 0.670246 coco/AP (L): 0.772172 coco/AR: 0.764861 coco/AR .5: 0.931832 coco/AR .75: 0.837689 coco/AR (M): 0.722234 coco/AR (L): 0.826013 2022/09/26 16:35:09 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_100.pth is removed 2022/09/26 16:35:13 - mmengine - INFO - The best checkpoint with 0.7061 coco/AP at 110 epoch is saved to best_coco/AP_epoch_110.pth. 2022/09/26 16:35:41 - mmengine - INFO - Epoch(train) [111][50/293] lr: 5.000000e-04 eta: 4:12:29 time: 0.564552 data_time: 0.175593 memory: 9504 loss_kpt: 0.000642 acc_pose: 0.845003 loss: 0.000642 2022/09/26 16:36:09 - mmengine - INFO - Epoch(train) [111][100/293] lr: 5.000000e-04 eta: 4:12:05 time: 0.568715 data_time: 0.075117 memory: 9504 loss_kpt: 0.000642 acc_pose: 0.860674 loss: 0.000642 2022/09/26 16:36:38 - mmengine - INFO - Epoch(train) [111][150/293] lr: 5.000000e-04 eta: 4:11:42 time: 0.566739 data_time: 0.081222 memory: 9504 loss_kpt: 0.000633 acc_pose: 0.833062 loss: 0.000633 2022/09/26 16:37:05 - mmengine - INFO - Epoch(train) [111][200/293] lr: 5.000000e-04 eta: 4:11:17 time: 0.548055 data_time: 0.068602 memory: 9504 loss_kpt: 0.000644 acc_pose: 0.831754 loss: 0.000644 2022/09/26 16:37:32 - mmengine - INFO - Epoch(train) [111][250/293] lr: 5.000000e-04 eta: 4:10:53 time: 0.547968 data_time: 0.098089 memory: 9504 loss_kpt: 0.000630 acc_pose: 0.854581 loss: 0.000630 2022/09/26 16:37:56 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:38:25 - mmengine - INFO - Epoch(train) [112][50/293] lr: 5.000000e-04 eta: 4:09:48 time: 0.587745 data_time: 0.101540 memory: 9504 loss_kpt: 0.000649 acc_pose: 0.832960 loss: 0.000649 2022/09/26 16:38:53 - mmengine - INFO - Epoch(train) [112][100/293] lr: 5.000000e-04 eta: 4:09:23 time: 0.552248 data_time: 0.107970 memory: 9504 loss_kpt: 0.000638 acc_pose: 0.853221 loss: 0.000638 2022/09/26 16:39:21 - mmengine - INFO - Epoch(train) [112][150/293] lr: 5.000000e-04 eta: 4:08:59 time: 0.552225 data_time: 0.119404 memory: 9504 loss_kpt: 0.000630 acc_pose: 0.834688 loss: 0.000630 2022/09/26 16:39:49 - mmengine - INFO - Epoch(train) [112][200/293] lr: 5.000000e-04 eta: 4:08:35 time: 0.564511 data_time: 0.079745 memory: 9504 loss_kpt: 0.000633 acc_pose: 0.833941 loss: 0.000633 2022/09/26 16:40:17 - mmengine - INFO - Epoch(train) [112][250/293] lr: 5.000000e-04 eta: 4:08:11 time: 0.568289 data_time: 0.077415 memory: 9504 loss_kpt: 0.000651 acc_pose: 0.807790 loss: 0.000651 2022/09/26 16:40:42 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:41:11 - mmengine - INFO - Epoch(train) [113][50/293] lr: 5.000000e-04 eta: 4:07:07 time: 0.589396 data_time: 0.175734 memory: 9504 loss_kpt: 0.000637 acc_pose: 0.837308 loss: 0.000637 2022/09/26 16:41:38 - mmengine - INFO - Epoch(train) [113][100/293] lr: 5.000000e-04 eta: 4:06:42 time: 0.537204 data_time: 0.077696 memory: 9504 loss_kpt: 0.000635 acc_pose: 0.850436 loss: 0.000635 2022/09/26 16:42:07 - mmengine - INFO - Epoch(train) [113][150/293] lr: 5.000000e-04 eta: 4:06:19 time: 0.577410 data_time: 0.094445 memory: 9504 loss_kpt: 0.000628 acc_pose: 0.837120 loss: 0.000628 2022/09/26 16:42:26 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:42:35 - mmengine - INFO - Epoch(train) [113][200/293] lr: 5.000000e-04 eta: 4:05:54 time: 0.553369 data_time: 0.084635 memory: 9504 loss_kpt: 0.000649 acc_pose: 0.869202 loss: 0.000649 2022/09/26 16:43:02 - mmengine - INFO - Epoch(train) [113][250/293] lr: 5.000000e-04 eta: 4:05:29 time: 0.537088 data_time: 0.089743 memory: 9504 loss_kpt: 0.000626 acc_pose: 0.812661 loss: 0.000626 2022/09/26 16:43:27 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:43:56 - mmengine - INFO - Epoch(train) [114][50/293] lr: 5.000000e-04 eta: 4:04:25 time: 0.578092 data_time: 0.089626 memory: 9504 loss_kpt: 0.000634 acc_pose: 0.839697 loss: 0.000634 2022/09/26 16:44:24 - mmengine - INFO - Epoch(train) [114][100/293] lr: 5.000000e-04 eta: 4:04:01 time: 0.561787 data_time: 0.094682 memory: 9504 loss_kpt: 0.000623 acc_pose: 0.836687 loss: 0.000623 2022/09/26 16:44:53 - mmengine - INFO - Epoch(train) [114][150/293] lr: 5.000000e-04 eta: 4:03:38 time: 0.581726 data_time: 0.133331 memory: 9504 loss_kpt: 0.000624 acc_pose: 0.789975 loss: 0.000624 2022/09/26 16:45:20 - mmengine - INFO - Epoch(train) [114][200/293] lr: 5.000000e-04 eta: 4:03:13 time: 0.553045 data_time: 0.126901 memory: 9504 loss_kpt: 0.000636 acc_pose: 0.806864 loss: 0.000636 2022/09/26 16:45:49 - mmengine - INFO - Epoch(train) [114][250/293] lr: 5.000000e-04 eta: 4:02:50 time: 0.576582 data_time: 0.131823 memory: 9504 loss_kpt: 0.000640 acc_pose: 0.821408 loss: 0.000640 2022/09/26 16:46:14 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:46:44 - mmengine - INFO - Epoch(train) [115][50/293] lr: 5.000000e-04 eta: 4:01:47 time: 0.598776 data_time: 0.106273 memory: 9504 loss_kpt: 0.000619 acc_pose: 0.835000 loss: 0.000619 2022/09/26 16:47:12 - mmengine - INFO - Epoch(train) [115][100/293] lr: 5.000000e-04 eta: 4:01:23 time: 0.558104 data_time: 0.076671 memory: 9504 loss_kpt: 0.000635 acc_pose: 0.846781 loss: 0.000635 2022/09/26 16:47:42 - mmengine - INFO - Epoch(train) [115][150/293] lr: 5.000000e-04 eta: 4:01:00 time: 0.599556 data_time: 0.082852 memory: 9504 loss_kpt: 0.000629 acc_pose: 0.827706 loss: 0.000629 2022/09/26 16:48:10 - mmengine - INFO - Epoch(train) [115][200/293] lr: 5.000000e-04 eta: 4:00:37 time: 0.574642 data_time: 0.069880 memory: 9504 loss_kpt: 0.000619 acc_pose: 0.825847 loss: 0.000619 2022/09/26 16:48:39 - mmengine - INFO - Epoch(train) [115][250/293] lr: 5.000000e-04 eta: 4:00:13 time: 0.566896 data_time: 0.078871 memory: 9504 loss_kpt: 0.000628 acc_pose: 0.843804 loss: 0.000628 2022/09/26 16:49:03 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:49:32 - mmengine - INFO - Epoch(train) [116][50/293] lr: 5.000000e-04 eta: 3:59:09 time: 0.579062 data_time: 0.229369 memory: 9504 loss_kpt: 0.000639 acc_pose: 0.814746 loss: 0.000639 2022/09/26 16:50:01 - mmengine - INFO - Epoch(train) [116][100/293] lr: 5.000000e-04 eta: 3:58:46 time: 0.576726 data_time: 0.207639 memory: 9504 loss_kpt: 0.000630 acc_pose: 0.811401 loss: 0.000630 2022/09/26 16:50:29 - mmengine - INFO - Epoch(train) [116][150/293] lr: 5.000000e-04 eta: 3:58:21 time: 0.549249 data_time: 0.227202 memory: 9504 loss_kpt: 0.000629 acc_pose: 0.836991 loss: 0.000629 2022/09/26 16:50:56 - mmengine - INFO - Epoch(train) [116][200/293] lr: 5.000000e-04 eta: 3:57:57 time: 0.546688 data_time: 0.177416 memory: 9504 loss_kpt: 0.000628 acc_pose: 0.833292 loss: 0.000628 2022/09/26 16:51:26 - mmengine - INFO - Epoch(train) [116][250/293] lr: 5.000000e-04 eta: 3:57:35 time: 0.606086 data_time: 0.102786 memory: 9504 loss_kpt: 0.000622 acc_pose: 0.807233 loss: 0.000622 2022/09/26 16:51:51 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:51:59 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:52:20 - mmengine - INFO - Epoch(train) [117][50/293] lr: 5.000000e-04 eta: 3:56:32 time: 0.590501 data_time: 0.110777 memory: 9504 loss_kpt: 0.000632 acc_pose: 0.851489 loss: 0.000632 2022/09/26 16:52:48 - mmengine - INFO - Epoch(train) [117][100/293] lr: 5.000000e-04 eta: 3:56:07 time: 0.560681 data_time: 0.170629 memory: 9504 loss_kpt: 0.000623 acc_pose: 0.764138 loss: 0.000623 2022/09/26 16:53:16 - mmengine - INFO - Epoch(train) [117][150/293] lr: 5.000000e-04 eta: 3:55:43 time: 0.554855 data_time: 0.090854 memory: 9504 loss_kpt: 0.000626 acc_pose: 0.818226 loss: 0.000626 2022/09/26 16:53:45 - mmengine - INFO - Epoch(train) [117][200/293] lr: 5.000000e-04 eta: 3:55:19 time: 0.567438 data_time: 0.098081 memory: 9504 loss_kpt: 0.000629 acc_pose: 0.833343 loss: 0.000629 2022/09/26 16:54:13 - mmengine - INFO - Epoch(train) [117][250/293] lr: 5.000000e-04 eta: 3:54:56 time: 0.566599 data_time: 0.148766 memory: 9504 loss_kpt: 0.000640 acc_pose: 0.880892 loss: 0.000640 2022/09/26 16:54:37 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:55:08 - mmengine - INFO - Epoch(train) [118][50/293] lr: 5.000000e-04 eta: 3:53:54 time: 0.608470 data_time: 0.114847 memory: 9504 loss_kpt: 0.000633 acc_pose: 0.828495 loss: 0.000633 2022/09/26 16:55:36 - mmengine - INFO - Epoch(train) [118][100/293] lr: 5.000000e-04 eta: 3:53:30 time: 0.561291 data_time: 0.071847 memory: 9504 loss_kpt: 0.000639 acc_pose: 0.780092 loss: 0.000639 2022/09/26 16:56:03 - mmengine - INFO - Epoch(train) [118][150/293] lr: 5.000000e-04 eta: 3:53:05 time: 0.551767 data_time: 0.086789 memory: 9504 loss_kpt: 0.000634 acc_pose: 0.833322 loss: 0.000634 2022/09/26 16:56:32 - mmengine - INFO - Epoch(train) [118][200/293] lr: 5.000000e-04 eta: 3:52:42 time: 0.570828 data_time: 0.104256 memory: 9504 loss_kpt: 0.000645 acc_pose: 0.808543 loss: 0.000645 2022/09/26 16:57:00 - mmengine - INFO - Epoch(train) [118][250/293] lr: 5.000000e-04 eta: 3:52:18 time: 0.567730 data_time: 0.072564 memory: 9504 loss_kpt: 0.000638 acc_pose: 0.849574 loss: 0.000638 2022/09/26 16:57:26 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 16:57:56 - mmengine - INFO - Epoch(train) [119][50/293] lr: 5.000000e-04 eta: 3:51:15 time: 0.589192 data_time: 0.099062 memory: 9504 loss_kpt: 0.000627 acc_pose: 0.826516 loss: 0.000627 2022/09/26 16:58:26 - mmengine - INFO - Epoch(train) [119][100/293] lr: 5.000000e-04 eta: 3:50:53 time: 0.600481 data_time: 0.100110 memory: 9504 loss_kpt: 0.000635 acc_pose: 0.835352 loss: 0.000635 2022/09/26 16:58:53 - mmengine - INFO - Epoch(train) [119][150/293] lr: 5.000000e-04 eta: 3:50:29 time: 0.553094 data_time: 0.082963 memory: 9504 loss_kpt: 0.000638 acc_pose: 0.798393 loss: 0.000638 2022/09/26 16:59:22 - mmengine - INFO - Epoch(train) [119][200/293] lr: 5.000000e-04 eta: 3:50:05 time: 0.565972 data_time: 0.107743 memory: 9504 loss_kpt: 0.000642 acc_pose: 0.816699 loss: 0.000642 2022/09/26 16:59:51 - mmengine - INFO - Epoch(train) [119][250/293] lr: 5.000000e-04 eta: 3:49:41 time: 0.583925 data_time: 0.131612 memory: 9504 loss_kpt: 0.000628 acc_pose: 0.818205 loss: 0.000628 2022/09/26 17:00:16 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:00:46 - mmengine - INFO - Epoch(train) [120][50/293] lr: 5.000000e-04 eta: 3:48:40 time: 0.609267 data_time: 0.109232 memory: 9504 loss_kpt: 0.000626 acc_pose: 0.806898 loss: 0.000626 2022/09/26 17:01:15 - mmengine - INFO - Epoch(train) [120][100/293] lr: 5.000000e-04 eta: 3:48:17 time: 0.584798 data_time: 0.115365 memory: 9504 loss_kpt: 0.000632 acc_pose: 0.880457 loss: 0.000632 2022/09/26 17:01:34 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:01:44 - mmengine - INFO - Epoch(train) [120][150/293] lr: 5.000000e-04 eta: 3:47:53 time: 0.575743 data_time: 0.088571 memory: 9504 loss_kpt: 0.000636 acc_pose: 0.828072 loss: 0.000636 2022/09/26 17:02:12 - mmengine - INFO - Epoch(train) [120][200/293] lr: 5.000000e-04 eta: 3:47:29 time: 0.553438 data_time: 0.085729 memory: 9504 loss_kpt: 0.000620 acc_pose: 0.859835 loss: 0.000620 2022/09/26 17:02:40 - mmengine - INFO - Epoch(train) [120][250/293] lr: 5.000000e-04 eta: 3:47:05 time: 0.555911 data_time: 0.080782 memory: 9504 loss_kpt: 0.000633 acc_pose: 0.885645 loss: 0.000633 2022/09/26 17:03:04 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:03:04 - mmengine - INFO - Saving checkpoint at 120 epochs 2022/09/26 17:03:29 - mmengine - INFO - Epoch(val) [120][50/407] eta: 0:01:57 time: 0.328750 data_time: 0.140659 memory: 9504 2022/09/26 17:03:45 - mmengine - INFO - Epoch(val) [120][100/407] eta: 0:01:41 time: 0.329843 data_time: 0.135912 memory: 1378 2022/09/26 17:04:01 - mmengine - INFO - Epoch(val) [120][150/407] eta: 0:01:21 time: 0.315398 data_time: 0.138323 memory: 1378 2022/09/26 17:04:17 - mmengine - INFO - Epoch(val) [120][200/407] eta: 0:01:05 time: 0.316309 data_time: 0.134223 memory: 1378 2022/09/26 17:04:32 - mmengine - INFO - Epoch(val) [120][250/407] eta: 0:00:48 time: 0.311054 data_time: 0.132908 memory: 1378 2022/09/26 17:04:49 - mmengine - INFO - Epoch(val) [120][300/407] eta: 0:00:35 time: 0.334203 data_time: 0.141441 memory: 1378 2022/09/26 17:05:05 - mmengine - INFO - Epoch(val) [120][350/407] eta: 0:00:17 time: 0.312488 data_time: 0.131923 memory: 1378 2022/09/26 17:05:14 - mmengine - INFO - Epoch(val) [120][400/407] eta: 0:00:01 time: 0.187098 data_time: 0.086785 memory: 1378 2022/09/26 17:05:47 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 17:06:00 - mmengine - INFO - Epoch(val) [120][407/407] coco/AP: 0.707632 coco/AP .5: 0.892090 coco/AP .75: 0.786908 coco/AP (M): 0.673383 coco/AP (L): 0.772613 coco/AR: 0.765617 coco/AR .5: 0.932305 coco/AR .75: 0.836902 coco/AR (M): 0.724693 coco/AR (L): 0.825158 2022/09/26 17:06:00 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_110.pth is removed 2022/09/26 17:06:04 - mmengine - INFO - The best checkpoint with 0.7076 coco/AP at 120 epoch is saved to best_coco/AP_epoch_120.pth. 2022/09/26 17:06:32 - mmengine - INFO - Epoch(train) [121][50/293] lr: 5.000000e-04 eta: 3:46:02 time: 0.570894 data_time: 0.152260 memory: 9504 loss_kpt: 0.000632 acc_pose: 0.802806 loss: 0.000632 2022/09/26 17:07:00 - mmengine - INFO - Epoch(train) [121][100/293] lr: 5.000000e-04 eta: 3:45:38 time: 0.553218 data_time: 0.092513 memory: 9504 loss_kpt: 0.000618 acc_pose: 0.829535 loss: 0.000618 2022/09/26 17:07:29 - mmengine - INFO - Epoch(train) [121][150/293] lr: 5.000000e-04 eta: 3:45:14 time: 0.576966 data_time: 0.110198 memory: 9504 loss_kpt: 0.000630 acc_pose: 0.852934 loss: 0.000630 2022/09/26 17:07:58 - mmengine - INFO - Epoch(train) [121][200/293] lr: 5.000000e-04 eta: 3:44:51 time: 0.581088 data_time: 0.087315 memory: 9504 loss_kpt: 0.000634 acc_pose: 0.857957 loss: 0.000634 2022/09/26 17:08:27 - mmengine - INFO - Epoch(train) [121][250/293] lr: 5.000000e-04 eta: 3:44:28 time: 0.579162 data_time: 0.091919 memory: 9504 loss_kpt: 0.000638 acc_pose: 0.820305 loss: 0.000638 2022/09/26 17:08:50 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:09:20 - mmengine - INFO - Epoch(train) [122][50/293] lr: 5.000000e-04 eta: 3:43:27 time: 0.599199 data_time: 0.097651 memory: 9504 loss_kpt: 0.000627 acc_pose: 0.835577 loss: 0.000627 2022/09/26 17:09:50 - mmengine - INFO - Epoch(train) [122][100/293] lr: 5.000000e-04 eta: 3:43:04 time: 0.593672 data_time: 0.088394 memory: 9504 loss_kpt: 0.000631 acc_pose: 0.807627 loss: 0.000631 2022/09/26 17:10:19 - mmengine - INFO - Epoch(train) [122][150/293] lr: 5.000000e-04 eta: 3:42:40 time: 0.567224 data_time: 0.078437 memory: 9504 loss_kpt: 0.000628 acc_pose: 0.868786 loss: 0.000628 2022/09/26 17:10:46 - mmengine - INFO - Epoch(train) [122][200/293] lr: 5.000000e-04 eta: 3:42:15 time: 0.542215 data_time: 0.082553 memory: 9504 loss_kpt: 0.000637 acc_pose: 0.764871 loss: 0.000637 2022/09/26 17:11:15 - mmengine - INFO - Epoch(train) [122][250/293] lr: 5.000000e-04 eta: 3:41:52 time: 0.582600 data_time: 0.098766 memory: 9504 loss_kpt: 0.000626 acc_pose: 0.841329 loss: 0.000626 2022/09/26 17:11:39 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:12:09 - mmengine - INFO - Epoch(train) [123][50/293] lr: 5.000000e-04 eta: 3:40:51 time: 0.604776 data_time: 0.166432 memory: 9504 loss_kpt: 0.000628 acc_pose: 0.851844 loss: 0.000628 2022/09/26 17:12:38 - mmengine - INFO - Epoch(train) [123][100/293] lr: 5.000000e-04 eta: 3:40:28 time: 0.588775 data_time: 0.151843 memory: 9504 loss_kpt: 0.000617 acc_pose: 0.825929 loss: 0.000617 2022/09/26 17:13:08 - mmengine - INFO - Epoch(train) [123][150/293] lr: 5.000000e-04 eta: 3:40:05 time: 0.595978 data_time: 0.200845 memory: 9504 loss_kpt: 0.000620 acc_pose: 0.789363 loss: 0.000620 2022/09/26 17:13:36 - mmengine - INFO - Epoch(train) [123][200/293] lr: 5.000000e-04 eta: 3:39:41 time: 0.565346 data_time: 0.100741 memory: 9504 loss_kpt: 0.000628 acc_pose: 0.818789 loss: 0.000628 2022/09/26 17:14:05 - mmengine - INFO - Epoch(train) [123][250/293] lr: 5.000000e-04 eta: 3:39:17 time: 0.567448 data_time: 0.084181 memory: 9504 loss_kpt: 0.000633 acc_pose: 0.787905 loss: 0.000633 2022/09/26 17:14:07 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:14:29 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:14:59 - mmengine - INFO - Epoch(train) [124][50/293] lr: 5.000000e-04 eta: 3:38:17 time: 0.601937 data_time: 0.127185 memory: 9504 loss_kpt: 0.000632 acc_pose: 0.839844 loss: 0.000632 2022/09/26 17:15:28 - mmengine - INFO - Epoch(train) [124][100/293] lr: 5.000000e-04 eta: 3:37:53 time: 0.569539 data_time: 0.140778 memory: 9504 loss_kpt: 0.000622 acc_pose: 0.862447 loss: 0.000622 2022/09/26 17:15:55 - mmengine - INFO - Epoch(train) [124][150/293] lr: 5.000000e-04 eta: 3:37:28 time: 0.549787 data_time: 0.077888 memory: 9504 loss_kpt: 0.000618 acc_pose: 0.863201 loss: 0.000618 2022/09/26 17:16:23 - mmengine - INFO - Epoch(train) [124][200/293] lr: 5.000000e-04 eta: 3:37:04 time: 0.554577 data_time: 0.071499 memory: 9504 loss_kpt: 0.000631 acc_pose: 0.835894 loss: 0.000631 2022/09/26 17:16:52 - mmengine - INFO - Epoch(train) [124][250/293] lr: 5.000000e-04 eta: 3:36:41 time: 0.587825 data_time: 0.076804 memory: 9504 loss_kpt: 0.000629 acc_pose: 0.851382 loss: 0.000629 2022/09/26 17:17:17 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:17:46 - mmengine - INFO - Epoch(train) [125][50/293] lr: 5.000000e-04 eta: 3:35:40 time: 0.581227 data_time: 0.171948 memory: 9504 loss_kpt: 0.000619 acc_pose: 0.841789 loss: 0.000619 2022/09/26 17:18:14 - mmengine - INFO - Epoch(train) [125][100/293] lr: 5.000000e-04 eta: 3:35:16 time: 0.571449 data_time: 0.090916 memory: 9504 loss_kpt: 0.000618 acc_pose: 0.833668 loss: 0.000618 2022/09/26 17:18:42 - mmengine - INFO - Epoch(train) [125][150/293] lr: 5.000000e-04 eta: 3:34:52 time: 0.561890 data_time: 0.095872 memory: 9504 loss_kpt: 0.000623 acc_pose: 0.808098 loss: 0.000623 2022/09/26 17:19:11 - mmengine - INFO - Epoch(train) [125][200/293] lr: 5.000000e-04 eta: 3:34:28 time: 0.578544 data_time: 0.080349 memory: 9504 loss_kpt: 0.000631 acc_pose: 0.819434 loss: 0.000631 2022/09/26 17:19:39 - mmengine - INFO - Epoch(train) [125][250/293] lr: 5.000000e-04 eta: 3:34:04 time: 0.556981 data_time: 0.085834 memory: 9504 loss_kpt: 0.000627 acc_pose: 0.862511 loss: 0.000627 2022/09/26 17:20:03 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:20:34 - mmengine - INFO - Epoch(train) [126][50/293] lr: 5.000000e-04 eta: 3:33:04 time: 0.606478 data_time: 0.100625 memory: 9504 loss_kpt: 0.000624 acc_pose: 0.813996 loss: 0.000624 2022/09/26 17:21:02 - mmengine - INFO - Epoch(train) [126][100/293] lr: 5.000000e-04 eta: 3:32:40 time: 0.565301 data_time: 0.089496 memory: 9504 loss_kpt: 0.000613 acc_pose: 0.832963 loss: 0.000613 2022/09/26 17:21:31 - mmengine - INFO - Epoch(train) [126][150/293] lr: 5.000000e-04 eta: 3:32:17 time: 0.589460 data_time: 0.125604 memory: 9504 loss_kpt: 0.000613 acc_pose: 0.853011 loss: 0.000613 2022/09/26 17:22:00 - mmengine - INFO - Epoch(train) [126][200/293] lr: 5.000000e-04 eta: 3:31:53 time: 0.574726 data_time: 0.096151 memory: 9504 loss_kpt: 0.000629 acc_pose: 0.853503 loss: 0.000629 2022/09/26 17:22:29 - mmengine - INFO - Epoch(train) [126][250/293] lr: 5.000000e-04 eta: 3:31:30 time: 0.569756 data_time: 0.120523 memory: 9504 loss_kpt: 0.000632 acc_pose: 0.825143 loss: 0.000632 2022/09/26 17:22:53 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:23:23 - mmengine - INFO - Epoch(train) [127][50/293] lr: 5.000000e-04 eta: 3:30:30 time: 0.600448 data_time: 0.106409 memory: 9504 loss_kpt: 0.000623 acc_pose: 0.843501 loss: 0.000623 2022/09/26 17:23:41 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:23:52 - mmengine - INFO - Epoch(train) [127][100/293] lr: 5.000000e-04 eta: 3:30:06 time: 0.576915 data_time: 0.213505 memory: 9504 loss_kpt: 0.000625 acc_pose: 0.859562 loss: 0.000625 2022/09/26 17:24:20 - mmengine - INFO - Epoch(train) [127][150/293] lr: 5.000000e-04 eta: 3:29:42 time: 0.566816 data_time: 0.156225 memory: 9504 loss_kpt: 0.000622 acc_pose: 0.825486 loss: 0.000622 2022/09/26 17:24:49 - mmengine - INFO - Epoch(train) [127][200/293] lr: 5.000000e-04 eta: 3:29:19 time: 0.588524 data_time: 0.129362 memory: 9504 loss_kpt: 0.000623 acc_pose: 0.838647 loss: 0.000623 2022/09/26 17:25:19 - mmengine - INFO - Epoch(train) [127][250/293] lr: 5.000000e-04 eta: 3:28:55 time: 0.582784 data_time: 0.102344 memory: 9504 loss_kpt: 0.000624 acc_pose: 0.869976 loss: 0.000624 2022/09/26 17:25:42 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:26:11 - mmengine - INFO - Epoch(train) [128][50/293] lr: 5.000000e-04 eta: 3:27:55 time: 0.584702 data_time: 0.123090 memory: 9504 loss_kpt: 0.000629 acc_pose: 0.819642 loss: 0.000629 2022/09/26 17:26:39 - mmengine - INFO - Epoch(train) [128][100/293] lr: 5.000000e-04 eta: 3:27:31 time: 0.551545 data_time: 0.090898 memory: 9504 loss_kpt: 0.000622 acc_pose: 0.823438 loss: 0.000622 2022/09/26 17:27:07 - mmengine - INFO - Epoch(train) [128][150/293] lr: 5.000000e-04 eta: 3:27:07 time: 0.568068 data_time: 0.091752 memory: 9504 loss_kpt: 0.000627 acc_pose: 0.853144 loss: 0.000627 2022/09/26 17:27:36 - mmengine - INFO - Epoch(train) [128][200/293] lr: 5.000000e-04 eta: 3:26:43 time: 0.572893 data_time: 0.109913 memory: 9504 loss_kpt: 0.000619 acc_pose: 0.802278 loss: 0.000619 2022/09/26 17:28:04 - mmengine - INFO - Epoch(train) [128][250/293] lr: 5.000000e-04 eta: 3:26:19 time: 0.570746 data_time: 0.208241 memory: 9504 loss_kpt: 0.000611 acc_pose: 0.833168 loss: 0.000611 2022/09/26 17:28:29 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:29:00 - mmengine - INFO - Epoch(train) [129][50/293] lr: 5.000000e-04 eta: 3:25:20 time: 0.607225 data_time: 0.090190 memory: 9504 loss_kpt: 0.000633 acc_pose: 0.829108 loss: 0.000633 2022/09/26 17:29:27 - mmengine - INFO - Epoch(train) [129][100/293] lr: 5.000000e-04 eta: 3:24:56 time: 0.552062 data_time: 0.074832 memory: 9504 loss_kpt: 0.000633 acc_pose: 0.817834 loss: 0.000633 2022/09/26 17:29:55 - mmengine - INFO - Epoch(train) [129][150/293] lr: 5.000000e-04 eta: 3:24:31 time: 0.555635 data_time: 0.076338 memory: 9504 loss_kpt: 0.000610 acc_pose: 0.815901 loss: 0.000610 2022/09/26 17:30:24 - mmengine - INFO - Epoch(train) [129][200/293] lr: 5.000000e-04 eta: 3:24:08 time: 0.571207 data_time: 0.082540 memory: 9504 loss_kpt: 0.000621 acc_pose: 0.832645 loss: 0.000621 2022/09/26 17:30:52 - mmengine - INFO - Epoch(train) [129][250/293] lr: 5.000000e-04 eta: 3:23:44 time: 0.569425 data_time: 0.092333 memory: 9504 loss_kpt: 0.000629 acc_pose: 0.812059 loss: 0.000629 2022/09/26 17:31:16 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:31:45 - mmengine - INFO - Epoch(train) [130][50/293] lr: 5.000000e-04 eta: 3:22:44 time: 0.588142 data_time: 0.107739 memory: 9504 loss_kpt: 0.000615 acc_pose: 0.864296 loss: 0.000615 2022/09/26 17:32:14 - mmengine - INFO - Epoch(train) [130][100/293] lr: 5.000000e-04 eta: 3:22:20 time: 0.575164 data_time: 0.084248 memory: 9504 loss_kpt: 0.000630 acc_pose: 0.842297 loss: 0.000630 2022/09/26 17:32:42 - mmengine - INFO - Epoch(train) [130][150/293] lr: 5.000000e-04 eta: 3:21:56 time: 0.564510 data_time: 0.115316 memory: 9504 loss_kpt: 0.000610 acc_pose: 0.837541 loss: 0.000610 2022/09/26 17:33:11 - mmengine - INFO - Epoch(train) [130][200/293] lr: 5.000000e-04 eta: 3:21:32 time: 0.563013 data_time: 0.150821 memory: 9504 loss_kpt: 0.000617 acc_pose: 0.808555 loss: 0.000617 2022/09/26 17:33:12 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:33:39 - mmengine - INFO - Epoch(train) [130][250/293] lr: 5.000000e-04 eta: 3:21:08 time: 0.566586 data_time: 0.064889 memory: 9504 loss_kpt: 0.000630 acc_pose: 0.835551 loss: 0.000630 2022/09/26 17:34:03 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:34:03 - mmengine - INFO - Saving checkpoint at 130 epochs 2022/09/26 17:34:28 - mmengine - INFO - Epoch(val) [130][50/407] eta: 0:01:50 time: 0.310134 data_time: 0.144762 memory: 9504 2022/09/26 17:34:47 - mmengine - INFO - Epoch(val) [130][100/407] eta: 0:01:54 time: 0.373748 data_time: 0.185995 memory: 1378 2022/09/26 17:35:03 - mmengine - INFO - Epoch(val) [130][150/407] eta: 0:01:20 time: 0.312663 data_time: 0.131271 memory: 1378 2022/09/26 17:35:18 - mmengine - INFO - Epoch(val) [130][200/407] eta: 0:01:02 time: 0.299810 data_time: 0.120336 memory: 1378 2022/09/26 17:35:33 - mmengine - INFO - Epoch(val) [130][250/407] eta: 0:00:48 time: 0.306474 data_time: 0.145677 memory: 1378 2022/09/26 17:35:48 - mmengine - INFO - Epoch(val) [130][300/407] eta: 0:00:32 time: 0.305161 data_time: 0.139886 memory: 1378 2022/09/26 17:36:03 - mmengine - INFO - Epoch(val) [130][350/407] eta: 0:00:17 time: 0.299324 data_time: 0.128409 memory: 1378 2022/09/26 17:36:13 - mmengine - INFO - Epoch(val) [130][400/407] eta: 0:00:01 time: 0.187296 data_time: 0.088607 memory: 1378 2022/09/26 17:36:44 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 17:36:57 - mmengine - INFO - Epoch(val) [130][407/407] coco/AP: 0.712503 coco/AP .5: 0.895421 coco/AP .75: 0.790946 coco/AP (M): 0.678363 coco/AP (L): 0.777497 coco/AR: 0.769994 coco/AR .5: 0.935768 coco/AR .75: 0.838948 coco/AR (M): 0.728872 coco/AR (L): 0.828948 2022/09/26 17:36:58 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_120.pth is removed 2022/09/26 17:37:01 - mmengine - INFO - The best checkpoint with 0.7125 coco/AP at 130 epoch is saved to best_coco/AP_epoch_130.pth. 2022/09/26 17:37:28 - mmengine - INFO - Epoch(train) [131][50/293] lr: 5.000000e-04 eta: 3:20:08 time: 0.556180 data_time: 0.246757 memory: 9504 loss_kpt: 0.000619 acc_pose: 0.809176 loss: 0.000619 2022/09/26 17:37:57 - mmengine - INFO - Epoch(train) [131][100/293] lr: 5.000000e-04 eta: 3:19:44 time: 0.572082 data_time: 0.101442 memory: 9504 loss_kpt: 0.000609 acc_pose: 0.824990 loss: 0.000609 2022/09/26 17:38:26 - mmengine - INFO - Epoch(train) [131][150/293] lr: 5.000000e-04 eta: 3:19:20 time: 0.577291 data_time: 0.102838 memory: 9504 loss_kpt: 0.000611 acc_pose: 0.829369 loss: 0.000611 2022/09/26 17:38:54 - mmengine - INFO - Epoch(train) [131][200/293] lr: 5.000000e-04 eta: 3:18:56 time: 0.565936 data_time: 0.089122 memory: 9504 loss_kpt: 0.000633 acc_pose: 0.835054 loss: 0.000633 2022/09/26 17:39:22 - mmengine - INFO - Epoch(train) [131][250/293] lr: 5.000000e-04 eta: 3:18:32 time: 0.560327 data_time: 0.091690 memory: 9504 loss_kpt: 0.000626 acc_pose: 0.841564 loss: 0.000626 2022/09/26 17:39:48 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:40:16 - mmengine - INFO - Epoch(train) [132][50/293] lr: 5.000000e-04 eta: 3:17:33 time: 0.575225 data_time: 0.086839 memory: 9504 loss_kpt: 0.000620 acc_pose: 0.836385 loss: 0.000620 2022/09/26 17:40:44 - mmengine - INFO - Epoch(train) [132][100/293] lr: 5.000000e-04 eta: 3:17:09 time: 0.556213 data_time: 0.134390 memory: 9504 loss_kpt: 0.000618 acc_pose: 0.852261 loss: 0.000618 2022/09/26 17:41:13 - mmengine - INFO - Epoch(train) [132][150/293] lr: 5.000000e-04 eta: 3:16:45 time: 0.575139 data_time: 0.196791 memory: 9504 loss_kpt: 0.000607 acc_pose: 0.844265 loss: 0.000607 2022/09/26 17:41:41 - mmengine - INFO - Epoch(train) [132][200/293] lr: 5.000000e-04 eta: 3:16:21 time: 0.565813 data_time: 0.093097 memory: 9504 loss_kpt: 0.000613 acc_pose: 0.850805 loss: 0.000613 2022/09/26 17:42:11 - mmengine - INFO - Epoch(train) [132][250/293] lr: 5.000000e-04 eta: 3:15:57 time: 0.589917 data_time: 0.095311 memory: 9504 loss_kpt: 0.000628 acc_pose: 0.808572 loss: 0.000628 2022/09/26 17:42:35 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:43:04 - mmengine - INFO - Epoch(train) [133][50/293] lr: 5.000000e-04 eta: 3:14:58 time: 0.582749 data_time: 0.091491 memory: 9504 loss_kpt: 0.000616 acc_pose: 0.858479 loss: 0.000616 2022/09/26 17:43:33 - mmengine - INFO - Epoch(train) [133][100/293] lr: 5.000000e-04 eta: 3:14:35 time: 0.575221 data_time: 0.093997 memory: 9504 loss_kpt: 0.000609 acc_pose: 0.835395 loss: 0.000609 2022/09/26 17:44:01 - mmengine - INFO - Epoch(train) [133][150/293] lr: 5.000000e-04 eta: 3:14:11 time: 0.574353 data_time: 0.079372 memory: 9504 loss_kpt: 0.000621 acc_pose: 0.832153 loss: 0.000621 2022/09/26 17:44:29 - mmengine - INFO - Epoch(train) [133][200/293] lr: 5.000000e-04 eta: 3:13:46 time: 0.558631 data_time: 0.080682 memory: 9504 loss_kpt: 0.000622 acc_pose: 0.828178 loss: 0.000622 2022/09/26 17:44:57 - mmengine - INFO - Epoch(train) [133][250/293] lr: 5.000000e-04 eta: 3:13:22 time: 0.549564 data_time: 0.152602 memory: 9504 loss_kpt: 0.000628 acc_pose: 0.821215 loss: 0.000628 2022/09/26 17:45:20 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:45:39 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:45:49 - mmengine - INFO - Epoch(train) [134][50/293] lr: 5.000000e-04 eta: 3:12:23 time: 0.568341 data_time: 0.159867 memory: 9504 loss_kpt: 0.000617 acc_pose: 0.833261 loss: 0.000617 2022/09/26 17:46:15 - mmengine - INFO - Epoch(train) [134][100/293] lr: 5.000000e-04 eta: 3:11:58 time: 0.527452 data_time: 0.093639 memory: 9504 loss_kpt: 0.000608 acc_pose: 0.802840 loss: 0.000608 2022/09/26 17:46:44 - mmengine - INFO - Epoch(train) [134][150/293] lr: 5.000000e-04 eta: 3:11:34 time: 0.580346 data_time: 0.144134 memory: 9504 loss_kpt: 0.000634 acc_pose: 0.836005 loss: 0.000634 2022/09/26 17:47:12 - mmengine - INFO - Epoch(train) [134][200/293] lr: 5.000000e-04 eta: 3:11:10 time: 0.566247 data_time: 0.194952 memory: 9504 loss_kpt: 0.000624 acc_pose: 0.859385 loss: 0.000624 2022/09/26 17:47:41 - mmengine - INFO - Epoch(train) [134][250/293] lr: 5.000000e-04 eta: 3:10:46 time: 0.563303 data_time: 0.107486 memory: 9504 loss_kpt: 0.000613 acc_pose: 0.787345 loss: 0.000613 2022/09/26 17:48:04 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:48:35 - mmengine - INFO - Epoch(train) [135][50/293] lr: 5.000000e-04 eta: 3:09:49 time: 0.618911 data_time: 0.113020 memory: 9504 loss_kpt: 0.000610 acc_pose: 0.869826 loss: 0.000610 2022/09/26 17:49:05 - mmengine - INFO - Epoch(train) [135][100/293] lr: 5.000000e-04 eta: 3:09:25 time: 0.598833 data_time: 0.102497 memory: 9504 loss_kpt: 0.000621 acc_pose: 0.785152 loss: 0.000621 2022/09/26 17:49:33 - mmengine - INFO - Epoch(train) [135][150/293] lr: 5.000000e-04 eta: 3:09:01 time: 0.555008 data_time: 0.081215 memory: 9504 loss_kpt: 0.000611 acc_pose: 0.818374 loss: 0.000611 2022/09/26 17:50:02 - mmengine - INFO - Epoch(train) [135][200/293] lr: 5.000000e-04 eta: 3:08:37 time: 0.576181 data_time: 0.077375 memory: 9504 loss_kpt: 0.000624 acc_pose: 0.868928 loss: 0.000624 2022/09/26 17:50:29 - mmengine - INFO - Epoch(train) [135][250/293] lr: 5.000000e-04 eta: 3:08:12 time: 0.552911 data_time: 0.101824 memory: 9504 loss_kpt: 0.000618 acc_pose: 0.874835 loss: 0.000618 2022/09/26 17:50:54 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:51:24 - mmengine - INFO - Epoch(train) [136][50/293] lr: 5.000000e-04 eta: 3:07:15 time: 0.607183 data_time: 0.105764 memory: 9504 loss_kpt: 0.000615 acc_pose: 0.841818 loss: 0.000615 2022/09/26 17:51:52 - mmengine - INFO - Epoch(train) [136][100/293] lr: 5.000000e-04 eta: 3:06:51 time: 0.550401 data_time: 0.135794 memory: 9504 loss_kpt: 0.000615 acc_pose: 0.797043 loss: 0.000615 2022/09/26 17:52:21 - mmengine - INFO - Epoch(train) [136][150/293] lr: 5.000000e-04 eta: 3:06:27 time: 0.576103 data_time: 0.213581 memory: 9504 loss_kpt: 0.000615 acc_pose: 0.849053 loss: 0.000615 2022/09/26 17:52:49 - mmengine - INFO - Epoch(train) [136][200/293] lr: 5.000000e-04 eta: 3:06:02 time: 0.559554 data_time: 0.082653 memory: 9504 loss_kpt: 0.000627 acc_pose: 0.834615 loss: 0.000627 2022/09/26 17:53:17 - mmengine - INFO - Epoch(train) [136][250/293] lr: 5.000000e-04 eta: 3:05:38 time: 0.571713 data_time: 0.092017 memory: 9504 loss_kpt: 0.000605 acc_pose: 0.797944 loss: 0.000605 2022/09/26 17:53:41 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:54:11 - mmengine - INFO - Epoch(train) [137][50/293] lr: 5.000000e-04 eta: 3:04:41 time: 0.583999 data_time: 0.203357 memory: 9504 loss_kpt: 0.000631 acc_pose: 0.802843 loss: 0.000631 2022/09/26 17:54:39 - mmengine - INFO - Epoch(train) [137][100/293] lr: 5.000000e-04 eta: 3:04:16 time: 0.561869 data_time: 0.125955 memory: 9504 loss_kpt: 0.000606 acc_pose: 0.860446 loss: 0.000606 2022/09/26 17:55:08 - mmengine - INFO - Epoch(train) [137][150/293] lr: 5.000000e-04 eta: 3:03:53 time: 0.583446 data_time: 0.207319 memory: 9504 loss_kpt: 0.000613 acc_pose: 0.856477 loss: 0.000613 2022/09/26 17:55:09 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:55:37 - mmengine - INFO - Epoch(train) [137][200/293] lr: 5.000000e-04 eta: 3:03:29 time: 0.582857 data_time: 0.082022 memory: 9504 loss_kpt: 0.000612 acc_pose: 0.849722 loss: 0.000612 2022/09/26 17:56:06 - mmengine - INFO - Epoch(train) [137][250/293] lr: 5.000000e-04 eta: 3:03:05 time: 0.586061 data_time: 0.112674 memory: 9504 loss_kpt: 0.000621 acc_pose: 0.870087 loss: 0.000621 2022/09/26 17:56:30 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:57:00 - mmengine - INFO - Epoch(train) [138][50/293] lr: 5.000000e-04 eta: 3:02:08 time: 0.590601 data_time: 0.205946 memory: 9504 loss_kpt: 0.000611 acc_pose: 0.850071 loss: 0.000611 2022/09/26 17:57:28 - mmengine - INFO - Epoch(train) [138][100/293] lr: 5.000000e-04 eta: 3:01:44 time: 0.564201 data_time: 0.171336 memory: 9504 loss_kpt: 0.000619 acc_pose: 0.849871 loss: 0.000619 2022/09/26 17:57:57 - mmengine - INFO - Epoch(train) [138][150/293] lr: 5.000000e-04 eta: 3:01:20 time: 0.573445 data_time: 0.094573 memory: 9504 loss_kpt: 0.000612 acc_pose: 0.852443 loss: 0.000612 2022/09/26 17:58:24 - mmengine - INFO - Epoch(train) [138][200/293] lr: 5.000000e-04 eta: 3:00:55 time: 0.542308 data_time: 0.133189 memory: 9504 loss_kpt: 0.000618 acc_pose: 0.816265 loss: 0.000618 2022/09/26 17:58:53 - mmengine - INFO - Epoch(train) [138][250/293] lr: 5.000000e-04 eta: 3:00:31 time: 0.583730 data_time: 0.118294 memory: 9504 loss_kpt: 0.000619 acc_pose: 0.832307 loss: 0.000619 2022/09/26 17:59:17 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 17:59:47 - mmengine - INFO - Epoch(train) [139][50/293] lr: 5.000000e-04 eta: 2:59:34 time: 0.596134 data_time: 0.110309 memory: 9504 loss_kpt: 0.000614 acc_pose: 0.873790 loss: 0.000614 2022/09/26 18:00:18 - mmengine - INFO - Epoch(train) [139][100/293] lr: 5.000000e-04 eta: 2:59:11 time: 0.614534 data_time: 0.166498 memory: 9504 loss_kpt: 0.000616 acc_pose: 0.820690 loss: 0.000616 2022/09/26 18:00:47 - mmengine - INFO - Epoch(train) [139][150/293] lr: 5.000000e-04 eta: 2:58:47 time: 0.576107 data_time: 0.146584 memory: 9504 loss_kpt: 0.000611 acc_pose: 0.815857 loss: 0.000611 2022/09/26 18:01:14 - mmengine - INFO - Epoch(train) [139][200/293] lr: 5.000000e-04 eta: 2:58:23 time: 0.548913 data_time: 0.066516 memory: 9504 loss_kpt: 0.000614 acc_pose: 0.842454 loss: 0.000614 2022/09/26 18:01:43 - mmengine - INFO - Epoch(train) [139][250/293] lr: 5.000000e-04 eta: 2:57:59 time: 0.582623 data_time: 0.079625 memory: 9504 loss_kpt: 0.000627 acc_pose: 0.808943 loss: 0.000627 2022/09/26 18:02:08 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:02:39 - mmengine - INFO - Epoch(train) [140][50/293] lr: 5.000000e-04 eta: 2:57:03 time: 0.618656 data_time: 0.099990 memory: 9504 loss_kpt: 0.000614 acc_pose: 0.834520 loss: 0.000614 2022/09/26 18:03:08 - mmengine - INFO - Epoch(train) [140][100/293] lr: 5.000000e-04 eta: 2:56:38 time: 0.562850 data_time: 0.076344 memory: 9504 loss_kpt: 0.000608 acc_pose: 0.828113 loss: 0.000608 2022/09/26 18:03:37 - mmengine - INFO - Epoch(train) [140][150/293] lr: 5.000000e-04 eta: 2:56:15 time: 0.590091 data_time: 0.170549 memory: 9504 loss_kpt: 0.000613 acc_pose: 0.849220 loss: 0.000613 2022/09/26 18:04:05 - mmengine - INFO - Epoch(train) [140][200/293] lr: 5.000000e-04 eta: 2:55:50 time: 0.556606 data_time: 0.191962 memory: 9504 loss_kpt: 0.000607 acc_pose: 0.843010 loss: 0.000607 2022/09/26 18:04:34 - mmengine - INFO - Epoch(train) [140][250/293] lr: 5.000000e-04 eta: 2:55:26 time: 0.574959 data_time: 0.198820 memory: 9504 loss_kpt: 0.000608 acc_pose: 0.824706 loss: 0.000608 2022/09/26 18:04:47 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:04:59 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:04:59 - mmengine - INFO - Saving checkpoint at 140 epochs 2022/09/26 18:05:25 - mmengine - INFO - Epoch(val) [140][50/407] eta: 0:01:55 time: 0.324516 data_time: 0.153051 memory: 9504 2022/09/26 18:05:40 - mmengine - INFO - Epoch(val) [140][100/407] eta: 0:01:37 time: 0.317291 data_time: 0.151787 memory: 1378 2022/09/26 18:05:56 - mmengine - INFO - Epoch(val) [140][150/407] eta: 0:01:20 time: 0.312664 data_time: 0.131835 memory: 1378 2022/09/26 18:06:12 - mmengine - INFO - Epoch(val) [140][200/407] eta: 0:01:05 time: 0.318203 data_time: 0.142291 memory: 1378 2022/09/26 18:06:27 - mmengine - INFO - Epoch(val) [140][250/407] eta: 0:00:47 time: 0.301211 data_time: 0.124327 memory: 1378 2022/09/26 18:06:42 - mmengine - INFO - Epoch(val) [140][300/407] eta: 0:00:32 time: 0.304445 data_time: 0.131245 memory: 1378 2022/09/26 18:06:59 - mmengine - INFO - Epoch(val) [140][350/407] eta: 0:00:18 time: 0.325566 data_time: 0.160487 memory: 1378 2022/09/26 18:07:08 - mmengine - INFO - Epoch(val) [140][400/407] eta: 0:00:01 time: 0.186548 data_time: 0.087463 memory: 1378 2022/09/26 18:07:41 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 18:07:54 - mmengine - INFO - Epoch(val) [140][407/407] coco/AP: 0.707010 coco/AP .5: 0.892151 coco/AP .75: 0.784609 coco/AP (M): 0.671560 coco/AP (L): 0.774125 coco/AR: 0.763460 coco/AR .5: 0.932935 coco/AR .75: 0.832336 coco/AR (M): 0.720896 coco/AR (L): 0.824749 2022/09/26 18:08:23 - mmengine - INFO - Epoch(train) [141][50/293] lr: 5.000000e-04 eta: 2:54:29 time: 0.573937 data_time: 0.098563 memory: 9504 loss_kpt: 0.000598 acc_pose: 0.846789 loss: 0.000598 2022/09/26 18:08:52 - mmengine - INFO - Epoch(train) [141][100/293] lr: 5.000000e-04 eta: 2:54:05 time: 0.583253 data_time: 0.087027 memory: 9504 loss_kpt: 0.000608 acc_pose: 0.841980 loss: 0.000608 2022/09/26 18:09:20 - mmengine - INFO - Epoch(train) [141][150/293] lr: 5.000000e-04 eta: 2:53:41 time: 0.561739 data_time: 0.088182 memory: 9504 loss_kpt: 0.000609 acc_pose: 0.818923 loss: 0.000609 2022/09/26 18:09:48 - mmengine - INFO - Epoch(train) [141][200/293] lr: 5.000000e-04 eta: 2:53:17 time: 0.564668 data_time: 0.115092 memory: 9504 loss_kpt: 0.000608 acc_pose: 0.805841 loss: 0.000608 2022/09/26 18:10:17 - mmengine - INFO - Epoch(train) [141][250/293] lr: 5.000000e-04 eta: 2:52:53 time: 0.570187 data_time: 0.138536 memory: 9504 loss_kpt: 0.000616 acc_pose: 0.847053 loss: 0.000616 2022/09/26 18:10:41 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:11:11 - mmengine - INFO - Epoch(train) [142][50/293] lr: 5.000000e-04 eta: 2:51:56 time: 0.602950 data_time: 0.175258 memory: 9504 loss_kpt: 0.000600 acc_pose: 0.862898 loss: 0.000600 2022/09/26 18:11:40 - mmengine - INFO - Epoch(train) [142][100/293] lr: 5.000000e-04 eta: 2:51:33 time: 0.582742 data_time: 0.079405 memory: 9504 loss_kpt: 0.000613 acc_pose: 0.863703 loss: 0.000613 2022/09/26 18:12:08 - mmengine - INFO - Epoch(train) [142][150/293] lr: 5.000000e-04 eta: 2:51:08 time: 0.555806 data_time: 0.112770 memory: 9504 loss_kpt: 0.000597 acc_pose: 0.849781 loss: 0.000597 2022/09/26 18:12:37 - mmengine - INFO - Epoch(train) [142][200/293] lr: 5.000000e-04 eta: 2:50:44 time: 0.578841 data_time: 0.070849 memory: 9504 loss_kpt: 0.000607 acc_pose: 0.821497 loss: 0.000607 2022/09/26 18:13:04 - mmengine - INFO - Epoch(train) [142][250/293] lr: 5.000000e-04 eta: 2:50:19 time: 0.543200 data_time: 0.078038 memory: 9504 loss_kpt: 0.000623 acc_pose: 0.830980 loss: 0.000623 2022/09/26 18:13:28 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:13:57 - mmengine - INFO - Epoch(train) [143][50/293] lr: 5.000000e-04 eta: 2:49:23 time: 0.584513 data_time: 0.138260 memory: 9504 loss_kpt: 0.000601 acc_pose: 0.843145 loss: 0.000601 2022/09/26 18:14:27 - mmengine - INFO - Epoch(train) [143][100/293] lr: 5.000000e-04 eta: 2:49:00 time: 0.603947 data_time: 0.140425 memory: 9504 loss_kpt: 0.000614 acc_pose: 0.830222 loss: 0.000614 2022/09/26 18:14:56 - mmengine - INFO - Epoch(train) [143][150/293] lr: 5.000000e-04 eta: 2:48:35 time: 0.571665 data_time: 0.083911 memory: 9504 loss_kpt: 0.000600 acc_pose: 0.875159 loss: 0.000600 2022/09/26 18:15:25 - mmengine - INFO - Epoch(train) [143][200/293] lr: 5.000000e-04 eta: 2:48:11 time: 0.571885 data_time: 0.078883 memory: 9504 loss_kpt: 0.000611 acc_pose: 0.856877 loss: 0.000611 2022/09/26 18:15:53 - mmengine - INFO - Epoch(train) [143][250/293] lr: 5.000000e-04 eta: 2:47:47 time: 0.563957 data_time: 0.077160 memory: 9504 loss_kpt: 0.000613 acc_pose: 0.857924 loss: 0.000613 2022/09/26 18:16:18 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:16:47 - mmengine - INFO - Epoch(train) [144][50/293] lr: 5.000000e-04 eta: 2:46:51 time: 0.576050 data_time: 0.102731 memory: 9504 loss_kpt: 0.000616 acc_pose: 0.858652 loss: 0.000616 2022/09/26 18:17:16 - mmengine - INFO - Epoch(train) [144][100/293] lr: 5.000000e-04 eta: 2:46:27 time: 0.585934 data_time: 0.161189 memory: 9504 loss_kpt: 0.000606 acc_pose: 0.836843 loss: 0.000606 2022/09/26 18:17:17 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:17:45 - mmengine - INFO - Epoch(train) [144][150/293] lr: 5.000000e-04 eta: 2:46:03 time: 0.585835 data_time: 0.098822 memory: 9504 loss_kpt: 0.000625 acc_pose: 0.814516 loss: 0.000625 2022/09/26 18:18:15 - mmengine - INFO - Epoch(train) [144][200/293] lr: 5.000000e-04 eta: 2:45:39 time: 0.587351 data_time: 0.083689 memory: 9504 loss_kpt: 0.000606 acc_pose: 0.778920 loss: 0.000606 2022/09/26 18:18:43 - mmengine - INFO - Epoch(train) [144][250/293] lr: 5.000000e-04 eta: 2:45:15 time: 0.563600 data_time: 0.086021 memory: 9504 loss_kpt: 0.000609 acc_pose: 0.832890 loss: 0.000609 2022/09/26 18:19:06 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:19:35 - mmengine - INFO - Epoch(train) [145][50/293] lr: 5.000000e-04 eta: 2:44:19 time: 0.590652 data_time: 0.096086 memory: 9504 loss_kpt: 0.000591 acc_pose: 0.825718 loss: 0.000591 2022/09/26 18:20:03 - mmengine - INFO - Epoch(train) [145][100/293] lr: 5.000000e-04 eta: 2:43:54 time: 0.562495 data_time: 0.064546 memory: 9504 loss_kpt: 0.000619 acc_pose: 0.828504 loss: 0.000619 2022/09/26 18:20:32 - mmengine - INFO - Epoch(train) [145][150/293] lr: 5.000000e-04 eta: 2:43:30 time: 0.570362 data_time: 0.100520 memory: 9504 loss_kpt: 0.000625 acc_pose: 0.825771 loss: 0.000625 2022/09/26 18:21:01 - mmengine - INFO - Epoch(train) [145][200/293] lr: 5.000000e-04 eta: 2:43:06 time: 0.572869 data_time: 0.085064 memory: 9504 loss_kpt: 0.000620 acc_pose: 0.853564 loss: 0.000620 2022/09/26 18:21:29 - mmengine - INFO - Epoch(train) [145][250/293] lr: 5.000000e-04 eta: 2:42:42 time: 0.570971 data_time: 0.091264 memory: 9504 loss_kpt: 0.000613 acc_pose: 0.825585 loss: 0.000613 2022/09/26 18:21:54 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:22:24 - mmengine - INFO - Epoch(train) [146][50/293] lr: 5.000000e-04 eta: 2:41:46 time: 0.593589 data_time: 0.112509 memory: 9504 loss_kpt: 0.000608 acc_pose: 0.812036 loss: 0.000608 2022/09/26 18:22:52 - mmengine - INFO - Epoch(train) [146][100/293] lr: 5.000000e-04 eta: 2:41:22 time: 0.573076 data_time: 0.074507 memory: 9504 loss_kpt: 0.000598 acc_pose: 0.843549 loss: 0.000598 2022/09/26 18:23:20 - mmengine - INFO - Epoch(train) [146][150/293] lr: 5.000000e-04 eta: 2:40:58 time: 0.555791 data_time: 0.079785 memory: 9504 loss_kpt: 0.000615 acc_pose: 0.856329 loss: 0.000615 2022/09/26 18:23:49 - mmengine - INFO - Epoch(train) [146][200/293] lr: 5.000000e-04 eta: 2:40:33 time: 0.573802 data_time: 0.093133 memory: 9504 loss_kpt: 0.000610 acc_pose: 0.825759 loss: 0.000610 2022/09/26 18:24:17 - mmengine - INFO - Epoch(train) [146][250/293] lr: 5.000000e-04 eta: 2:40:09 time: 0.560568 data_time: 0.070709 memory: 9504 loss_kpt: 0.000603 acc_pose: 0.838816 loss: 0.000603 2022/09/26 18:24:41 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:25:10 - mmengine - INFO - Epoch(train) [147][50/293] lr: 5.000000e-04 eta: 2:39:13 time: 0.577166 data_time: 0.118008 memory: 9504 loss_kpt: 0.000603 acc_pose: 0.796927 loss: 0.000603 2022/09/26 18:25:37 - mmengine - INFO - Epoch(train) [147][100/293] lr: 5.000000e-04 eta: 2:38:49 time: 0.551244 data_time: 0.124995 memory: 9504 loss_kpt: 0.000610 acc_pose: 0.855871 loss: 0.000610 2022/09/26 18:26:05 - mmengine - INFO - Epoch(train) [147][150/293] lr: 5.000000e-04 eta: 2:38:24 time: 0.556823 data_time: 0.131053 memory: 9504 loss_kpt: 0.000624 acc_pose: 0.849367 loss: 0.000624 2022/09/26 18:26:34 - mmengine - INFO - Epoch(train) [147][200/293] lr: 5.000000e-04 eta: 2:38:00 time: 0.568022 data_time: 0.084327 memory: 9504 loss_kpt: 0.000603 acc_pose: 0.860907 loss: 0.000603 2022/09/26 18:26:47 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:27:03 - mmengine - INFO - Epoch(train) [147][250/293] lr: 5.000000e-04 eta: 2:37:36 time: 0.585656 data_time: 0.107723 memory: 9504 loss_kpt: 0.000617 acc_pose: 0.822937 loss: 0.000617 2022/09/26 18:27:27 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:27:57 - mmengine - INFO - Epoch(train) [148][50/293] lr: 5.000000e-04 eta: 2:36:41 time: 0.595843 data_time: 0.114362 memory: 9504 loss_kpt: 0.000607 acc_pose: 0.839700 loss: 0.000607 2022/09/26 18:28:24 - mmengine - INFO - Epoch(train) [148][100/293] lr: 5.000000e-04 eta: 2:36:16 time: 0.553159 data_time: 0.101114 memory: 9504 loss_kpt: 0.000598 acc_pose: 0.844745 loss: 0.000598 2022/09/26 18:28:53 - mmengine - INFO - Epoch(train) [148][150/293] lr: 5.000000e-04 eta: 2:35:52 time: 0.581162 data_time: 0.121250 memory: 9504 loss_kpt: 0.000611 acc_pose: 0.845974 loss: 0.000611 2022/09/26 18:29:22 - mmengine - INFO - Epoch(train) [148][200/293] lr: 5.000000e-04 eta: 2:35:28 time: 0.579167 data_time: 0.123024 memory: 9504 loss_kpt: 0.000595 acc_pose: 0.845475 loss: 0.000595 2022/09/26 18:29:51 - mmengine - INFO - Epoch(train) [148][250/293] lr: 5.000000e-04 eta: 2:35:04 time: 0.578184 data_time: 0.143415 memory: 9504 loss_kpt: 0.000613 acc_pose: 0.858553 loss: 0.000613 2022/09/26 18:30:16 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:30:46 - mmengine - INFO - Epoch(train) [149][50/293] lr: 5.000000e-04 eta: 2:34:09 time: 0.604899 data_time: 0.163733 memory: 9504 loss_kpt: 0.000615 acc_pose: 0.827527 loss: 0.000615 2022/09/26 18:31:14 - mmengine - INFO - Epoch(train) [149][100/293] lr: 5.000000e-04 eta: 2:33:45 time: 0.566818 data_time: 0.143730 memory: 9504 loss_kpt: 0.000612 acc_pose: 0.841378 loss: 0.000612 2022/09/26 18:31:42 - mmengine - INFO - Epoch(train) [149][150/293] lr: 5.000000e-04 eta: 2:33:20 time: 0.555500 data_time: 0.152018 memory: 9504 loss_kpt: 0.000607 acc_pose: 0.851342 loss: 0.000607 2022/09/26 18:32:11 - mmengine - INFO - Epoch(train) [149][200/293] lr: 5.000000e-04 eta: 2:32:56 time: 0.577463 data_time: 0.093125 memory: 9504 loss_kpt: 0.000609 acc_pose: 0.836466 loss: 0.000609 2022/09/26 18:32:40 - mmengine - INFO - Epoch(train) [149][250/293] lr: 5.000000e-04 eta: 2:32:32 time: 0.577380 data_time: 0.116966 memory: 9504 loss_kpt: 0.000612 acc_pose: 0.830092 loss: 0.000612 2022/09/26 18:33:05 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:33:34 - mmengine - INFO - Epoch(train) [150][50/293] lr: 5.000000e-04 eta: 2:31:37 time: 0.592625 data_time: 0.101952 memory: 9504 loss_kpt: 0.000600 acc_pose: 0.835126 loss: 0.000600 2022/09/26 18:34:03 - mmengine - INFO - Epoch(train) [150][100/293] lr: 5.000000e-04 eta: 2:31:13 time: 0.569521 data_time: 0.096832 memory: 9504 loss_kpt: 0.000603 acc_pose: 0.847842 loss: 0.000603 2022/09/26 18:34:31 - mmengine - INFO - Epoch(train) [150][150/293] lr: 5.000000e-04 eta: 2:30:48 time: 0.568099 data_time: 0.095215 memory: 9504 loss_kpt: 0.000613 acc_pose: 0.879631 loss: 0.000613 2022/09/26 18:34:59 - mmengine - INFO - Epoch(train) [150][200/293] lr: 5.000000e-04 eta: 2:30:24 time: 0.559368 data_time: 0.089887 memory: 9504 loss_kpt: 0.000610 acc_pose: 0.849154 loss: 0.000610 2022/09/26 18:35:28 - mmengine - INFO - Epoch(train) [150][250/293] lr: 5.000000e-04 eta: 2:30:00 time: 0.581198 data_time: 0.176251 memory: 9504 loss_kpt: 0.000610 acc_pose: 0.828700 loss: 0.000610 2022/09/26 18:35:53 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:35:53 - mmengine - INFO - Saving checkpoint at 150 epochs 2022/09/26 18:36:19 - mmengine - INFO - Epoch(val) [150][50/407] eta: 0:02:02 time: 0.342144 data_time: 0.162308 memory: 9504 2022/09/26 18:36:36 - mmengine - INFO - Epoch(val) [150][100/407] eta: 0:01:40 time: 0.326143 data_time: 0.149428 memory: 1378 2022/09/26 18:36:51 - mmengine - INFO - Epoch(val) [150][150/407] eta: 0:01:20 time: 0.312010 data_time: 0.163017 memory: 1378 2022/09/26 18:37:06 - mmengine - INFO - Epoch(val) [150][200/407] eta: 0:01:02 time: 0.303707 data_time: 0.159478 memory: 1378 2022/09/26 18:37:23 - mmengine - INFO - Epoch(val) [150][250/407] eta: 0:00:52 time: 0.331962 data_time: 0.169042 memory: 1378 2022/09/26 18:37:39 - mmengine - INFO - Epoch(val) [150][300/407] eta: 0:00:33 time: 0.315229 data_time: 0.139710 memory: 1378 2022/09/26 18:37:55 - mmengine - INFO - Epoch(val) [150][350/407] eta: 0:00:18 time: 0.329873 data_time: 0.164626 memory: 1378 2022/09/26 18:38:03 - mmengine - INFO - Epoch(val) [150][400/407] eta: 0:00:01 time: 0.150215 data_time: 0.075465 memory: 1378 2022/09/26 18:38:35 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 18:38:48 - mmengine - INFO - Epoch(val) [150][407/407] coco/AP: 0.711308 coco/AP .5: 0.893098 coco/AP .75: 0.788718 coco/AP (M): 0.676491 coco/AP (L): 0.778469 coco/AR: 0.768168 coco/AR .5: 0.932147 coco/AR .75: 0.838791 coco/AR (M): 0.725376 coco/AR (L): 0.829617 2022/09/26 18:39:17 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:39:17 - mmengine - INFO - Epoch(train) [151][50/293] lr: 5.000000e-04 eta: 2:29:05 time: 0.578592 data_time: 0.101202 memory: 9504 loss_kpt: 0.000614 acc_pose: 0.894615 loss: 0.000614 2022/09/26 18:39:46 - mmengine - INFO - Epoch(train) [151][100/293] lr: 5.000000e-04 eta: 2:28:41 time: 0.584765 data_time: 0.082341 memory: 9504 loss_kpt: 0.000612 acc_pose: 0.906640 loss: 0.000612 2022/09/26 18:40:15 - mmengine - INFO - Epoch(train) [151][150/293] lr: 5.000000e-04 eta: 2:28:17 time: 0.582186 data_time: 0.106418 memory: 9504 loss_kpt: 0.000600 acc_pose: 0.857205 loss: 0.000600 2022/09/26 18:40:44 - mmengine - INFO - Epoch(train) [151][200/293] lr: 5.000000e-04 eta: 2:27:52 time: 0.575554 data_time: 0.071990 memory: 9504 loss_kpt: 0.000610 acc_pose: 0.804102 loss: 0.000610 2022/09/26 18:41:12 - mmengine - INFO - Epoch(train) [151][250/293] lr: 5.000000e-04 eta: 2:27:28 time: 0.559626 data_time: 0.076434 memory: 9504 loss_kpt: 0.000615 acc_pose: 0.848286 loss: 0.000615 2022/09/26 18:41:37 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:42:07 - mmengine - INFO - Epoch(train) [152][50/293] lr: 5.000000e-04 eta: 2:26:33 time: 0.599187 data_time: 0.179959 memory: 9504 loss_kpt: 0.000599 acc_pose: 0.774991 loss: 0.000599 2022/09/26 18:42:35 - mmengine - INFO - Epoch(train) [152][100/293] lr: 5.000000e-04 eta: 2:26:09 time: 0.553915 data_time: 0.101614 memory: 9504 loss_kpt: 0.000594 acc_pose: 0.866720 loss: 0.000594 2022/09/26 18:43:03 - mmengine - INFO - Epoch(train) [152][150/293] lr: 5.000000e-04 eta: 2:25:44 time: 0.559414 data_time: 0.072343 memory: 9504 loss_kpt: 0.000614 acc_pose: 0.855817 loss: 0.000614 2022/09/26 18:43:32 - mmengine - INFO - Epoch(train) [152][200/293] lr: 5.000000e-04 eta: 2:25:20 time: 0.575213 data_time: 0.086404 memory: 9504 loss_kpt: 0.000612 acc_pose: 0.818349 loss: 0.000612 2022/09/26 18:44:01 - mmengine - INFO - Epoch(train) [152][250/293] lr: 5.000000e-04 eta: 2:24:56 time: 0.581833 data_time: 0.096295 memory: 9504 loss_kpt: 0.000617 acc_pose: 0.847671 loss: 0.000617 2022/09/26 18:44:26 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:44:56 - mmengine - INFO - Epoch(train) [153][50/293] lr: 5.000000e-04 eta: 2:24:02 time: 0.602060 data_time: 0.117061 memory: 9504 loss_kpt: 0.000595 acc_pose: 0.846320 loss: 0.000595 2022/09/26 18:45:25 - mmengine - INFO - Epoch(train) [153][100/293] lr: 5.000000e-04 eta: 2:23:38 time: 0.581741 data_time: 0.079055 memory: 9504 loss_kpt: 0.000604 acc_pose: 0.869237 loss: 0.000604 2022/09/26 18:45:54 - mmengine - INFO - Epoch(train) [153][150/293] lr: 5.000000e-04 eta: 2:23:13 time: 0.578500 data_time: 0.093209 memory: 9504 loss_kpt: 0.000599 acc_pose: 0.857234 loss: 0.000599 2022/09/26 18:46:24 - mmengine - INFO - Epoch(train) [153][200/293] lr: 5.000000e-04 eta: 2:22:49 time: 0.595876 data_time: 0.117166 memory: 9504 loss_kpt: 0.000615 acc_pose: 0.870881 loss: 0.000615 2022/09/26 18:46:52 - mmengine - INFO - Epoch(train) [153][250/293] lr: 5.000000e-04 eta: 2:22:25 time: 0.562393 data_time: 0.143801 memory: 9504 loss_kpt: 0.000616 acc_pose: 0.843948 loss: 0.000616 2022/09/26 18:47:18 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:47:48 - mmengine - INFO - Epoch(train) [154][50/293] lr: 5.000000e-04 eta: 2:21:31 time: 0.613355 data_time: 0.102820 memory: 9504 loss_kpt: 0.000598 acc_pose: 0.862049 loss: 0.000598 2022/09/26 18:48:15 - mmengine - INFO - Epoch(train) [154][100/293] lr: 5.000000e-04 eta: 2:21:06 time: 0.544202 data_time: 0.073886 memory: 9504 loss_kpt: 0.000593 acc_pose: 0.867758 loss: 0.000593 2022/09/26 18:48:44 - mmengine - INFO - Epoch(train) [154][150/293] lr: 5.000000e-04 eta: 2:20:42 time: 0.579849 data_time: 0.123255 memory: 9504 loss_kpt: 0.000604 acc_pose: 0.872623 loss: 0.000604 2022/09/26 18:48:57 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:49:13 - mmengine - INFO - Epoch(train) [154][200/293] lr: 5.000000e-04 eta: 2:20:18 time: 0.565234 data_time: 0.074817 memory: 9504 loss_kpt: 0.000608 acc_pose: 0.869934 loss: 0.000608 2022/09/26 18:49:41 - mmengine - INFO - Epoch(train) [154][250/293] lr: 5.000000e-04 eta: 2:19:53 time: 0.559957 data_time: 0.077687 memory: 9504 loss_kpt: 0.000599 acc_pose: 0.865100 loss: 0.000599 2022/09/26 18:50:04 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:50:33 - mmengine - INFO - Epoch(train) [155][50/293] lr: 5.000000e-04 eta: 2:18:59 time: 0.580517 data_time: 0.133661 memory: 9504 loss_kpt: 0.000602 acc_pose: 0.894066 loss: 0.000602 2022/09/26 18:51:02 - mmengine - INFO - Epoch(train) [155][100/293] lr: 5.000000e-04 eta: 2:18:35 time: 0.577124 data_time: 0.070011 memory: 9504 loss_kpt: 0.000597 acc_pose: 0.842789 loss: 0.000597 2022/09/26 18:51:32 - mmengine - INFO - Epoch(train) [155][150/293] lr: 5.000000e-04 eta: 2:18:11 time: 0.601966 data_time: 0.122492 memory: 9504 loss_kpt: 0.000594 acc_pose: 0.807629 loss: 0.000594 2022/09/26 18:52:01 - mmengine - INFO - Epoch(train) [155][200/293] lr: 5.000000e-04 eta: 2:17:47 time: 0.588923 data_time: 0.085356 memory: 9504 loss_kpt: 0.000616 acc_pose: 0.849980 loss: 0.000616 2022/09/26 18:52:30 - mmengine - INFO - Epoch(train) [155][250/293] lr: 5.000000e-04 eta: 2:17:22 time: 0.563546 data_time: 0.144390 memory: 9504 loss_kpt: 0.000608 acc_pose: 0.827525 loss: 0.000608 2022/09/26 18:52:55 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:53:25 - mmengine - INFO - Epoch(train) [156][50/293] lr: 5.000000e-04 eta: 2:16:29 time: 0.605085 data_time: 0.086722 memory: 9504 loss_kpt: 0.000593 acc_pose: 0.840986 loss: 0.000593 2022/09/26 18:53:54 - mmengine - INFO - Epoch(train) [156][100/293] lr: 5.000000e-04 eta: 2:16:04 time: 0.575495 data_time: 0.123889 memory: 9504 loss_kpt: 0.000606 acc_pose: 0.882410 loss: 0.000606 2022/09/26 18:54:23 - mmengine - INFO - Epoch(train) [156][150/293] lr: 5.000000e-04 eta: 2:15:40 time: 0.574191 data_time: 0.205731 memory: 9504 loss_kpt: 0.000603 acc_pose: 0.822985 loss: 0.000603 2022/09/26 18:54:52 - mmengine - INFO - Epoch(train) [156][200/293] lr: 5.000000e-04 eta: 2:15:16 time: 0.579563 data_time: 0.070254 memory: 9504 loss_kpt: 0.000611 acc_pose: 0.768696 loss: 0.000611 2022/09/26 18:55:21 - mmengine - INFO - Epoch(train) [156][250/293] lr: 5.000000e-04 eta: 2:14:51 time: 0.576571 data_time: 0.158445 memory: 9504 loss_kpt: 0.000603 acc_pose: 0.855526 loss: 0.000603 2022/09/26 18:55:45 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:56:15 - mmengine - INFO - Epoch(train) [157][50/293] lr: 5.000000e-04 eta: 2:13:58 time: 0.590612 data_time: 0.115222 memory: 9504 loss_kpt: 0.000592 acc_pose: 0.845967 loss: 0.000592 2022/09/26 18:56:44 - mmengine - INFO - Epoch(train) [157][100/293] lr: 5.000000e-04 eta: 2:13:34 time: 0.582515 data_time: 0.085437 memory: 9504 loss_kpt: 0.000594 acc_pose: 0.812145 loss: 0.000594 2022/09/26 18:57:13 - mmengine - INFO - Epoch(train) [157][150/293] lr: 5.000000e-04 eta: 2:13:09 time: 0.582853 data_time: 0.074075 memory: 9504 loss_kpt: 0.000609 acc_pose: 0.831018 loss: 0.000609 2022/09/26 18:57:42 - mmengine - INFO - Epoch(train) [157][200/293] lr: 5.000000e-04 eta: 2:12:45 time: 0.576279 data_time: 0.090294 memory: 9504 loss_kpt: 0.000603 acc_pose: 0.849349 loss: 0.000603 2022/09/26 18:58:11 - mmengine - INFO - Epoch(train) [157][250/293] lr: 5.000000e-04 eta: 2:12:21 time: 0.586438 data_time: 0.134186 memory: 9504 loss_kpt: 0.000606 acc_pose: 0.842223 loss: 0.000606 2022/09/26 18:58:36 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:58:36 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 18:59:07 - mmengine - INFO - Epoch(train) [158][50/293] lr: 5.000000e-04 eta: 2:11:28 time: 0.605517 data_time: 0.108080 memory: 9504 loss_kpt: 0.000606 acc_pose: 0.822664 loss: 0.000606 2022/09/26 18:59:35 - mmengine - INFO - Epoch(train) [158][100/293] lr: 5.000000e-04 eta: 2:11:03 time: 0.560362 data_time: 0.071712 memory: 9504 loss_kpt: 0.000594 acc_pose: 0.823565 loss: 0.000594 2022/09/26 19:00:05 - mmengine - INFO - Epoch(train) [158][150/293] lr: 5.000000e-04 eta: 2:10:39 time: 0.600736 data_time: 0.065194 memory: 9504 loss_kpt: 0.000600 acc_pose: 0.839450 loss: 0.000600 2022/09/26 19:00:33 - mmengine - INFO - Epoch(train) [158][200/293] lr: 5.000000e-04 eta: 2:10:14 time: 0.573230 data_time: 0.090008 memory: 9504 loss_kpt: 0.000597 acc_pose: 0.805609 loss: 0.000597 2022/09/26 19:01:02 - mmengine - INFO - Epoch(train) [158][250/293] lr: 5.000000e-04 eta: 2:09:50 time: 0.562665 data_time: 0.101568 memory: 9504 loss_kpt: 0.000583 acc_pose: 0.864730 loss: 0.000583 2022/09/26 19:01:26 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:01:56 - mmengine - INFO - Epoch(train) [159][50/293] lr: 5.000000e-04 eta: 2:08:57 time: 0.603605 data_time: 0.091001 memory: 9504 loss_kpt: 0.000603 acc_pose: 0.879487 loss: 0.000603 2022/09/26 19:02:25 - mmengine - INFO - Epoch(train) [159][100/293] lr: 5.000000e-04 eta: 2:08:32 time: 0.578812 data_time: 0.122686 memory: 9504 loss_kpt: 0.000601 acc_pose: 0.864749 loss: 0.000601 2022/09/26 19:02:53 - mmengine - INFO - Epoch(train) [159][150/293] lr: 5.000000e-04 eta: 2:08:08 time: 0.555246 data_time: 0.081528 memory: 9504 loss_kpt: 0.000593 acc_pose: 0.885407 loss: 0.000593 2022/09/26 19:03:23 - mmengine - INFO - Epoch(train) [159][200/293] lr: 5.000000e-04 eta: 2:07:43 time: 0.586752 data_time: 0.192299 memory: 9504 loss_kpt: 0.000605 acc_pose: 0.844175 loss: 0.000605 2022/09/26 19:03:52 - mmengine - INFO - Epoch(train) [159][250/293] lr: 5.000000e-04 eta: 2:07:19 time: 0.586661 data_time: 0.127357 memory: 9504 loss_kpt: 0.000601 acc_pose: 0.879753 loss: 0.000601 2022/09/26 19:04:15 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:04:46 - mmengine - INFO - Epoch(train) [160][50/293] lr: 5.000000e-04 eta: 2:06:26 time: 0.608609 data_time: 0.129361 memory: 9504 loss_kpt: 0.000597 acc_pose: 0.886664 loss: 0.000597 2022/09/26 19:05:14 - mmengine - INFO - Epoch(train) [160][100/293] lr: 5.000000e-04 eta: 2:06:02 time: 0.563840 data_time: 0.075287 memory: 9504 loss_kpt: 0.000612 acc_pose: 0.843681 loss: 0.000612 2022/09/26 19:05:42 - mmengine - INFO - Epoch(train) [160][150/293] lr: 5.000000e-04 eta: 2:05:37 time: 0.567211 data_time: 0.069459 memory: 9504 loss_kpt: 0.000607 acc_pose: 0.885229 loss: 0.000607 2022/09/26 19:06:11 - mmengine - INFO - Epoch(train) [160][200/293] lr: 5.000000e-04 eta: 2:05:13 time: 0.580707 data_time: 0.108401 memory: 9504 loss_kpt: 0.000592 acc_pose: 0.809217 loss: 0.000592 2022/09/26 19:06:39 - mmengine - INFO - Epoch(train) [160][250/293] lr: 5.000000e-04 eta: 2:04:48 time: 0.560769 data_time: 0.144525 memory: 9504 loss_kpt: 0.000603 acc_pose: 0.849582 loss: 0.000603 2022/09/26 19:07:04 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:07:04 - mmengine - INFO - Saving checkpoint at 160 epochs 2022/09/26 19:07:28 - mmengine - INFO - Epoch(val) [160][50/407] eta: 0:01:54 time: 0.320166 data_time: 0.153375 memory: 9504 2022/09/26 19:07:44 - mmengine - INFO - Epoch(val) [160][100/407] eta: 0:01:37 time: 0.318477 data_time: 0.142973 memory: 1378 2022/09/26 19:08:00 - mmengine - INFO - Epoch(val) [160][150/407] eta: 0:01:21 time: 0.315373 data_time: 0.163077 memory: 1378 2022/09/26 19:08:16 - mmengine - INFO - Epoch(val) [160][200/407] eta: 0:01:06 time: 0.319109 data_time: 0.156609 memory: 1378 2022/09/26 19:08:31 - mmengine - INFO - Epoch(val) [160][250/407] eta: 0:00:48 time: 0.307683 data_time: 0.149076 memory: 1378 2022/09/26 19:08:47 - mmengine - INFO - Epoch(val) [160][300/407] eta: 0:00:33 time: 0.315402 data_time: 0.138822 memory: 1378 2022/09/26 19:09:03 - mmengine - INFO - Epoch(val) [160][350/407] eta: 0:00:17 time: 0.315107 data_time: 0.125577 memory: 1378 2022/09/26 19:09:14 - mmengine - INFO - Epoch(val) [160][400/407] eta: 0:00:01 time: 0.230316 data_time: 0.110976 memory: 1378 2022/09/26 19:09:47 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 19:10:00 - mmengine - INFO - Epoch(val) [160][407/407] coco/AP: 0.714701 coco/AP .5: 0.897098 coco/AP .75: 0.792498 coco/AP (M): 0.678526 coco/AP (L): 0.779665 coco/AR: 0.769458 coco/AR .5: 0.934037 coco/AR .75: 0.839893 coco/AR (M): 0.727151 coco/AR (L): 0.830769 2022/09/26 19:10:00 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_130.pth is removed 2022/09/26 19:10:04 - mmengine - INFO - The best checkpoint with 0.7147 coco/AP at 160 epoch is saved to best_coco/AP_epoch_160.pth. 2022/09/26 19:10:32 - mmengine - INFO - Epoch(train) [161][50/293] lr: 5.000000e-04 eta: 2:03:55 time: 0.575222 data_time: 0.162590 memory: 9504 loss_kpt: 0.000582 acc_pose: 0.780384 loss: 0.000582 2022/09/26 19:11:02 - mmengine - INFO - Epoch(train) [161][100/293] lr: 5.000000e-04 eta: 2:03:31 time: 0.595014 data_time: 0.104392 memory: 9504 loss_kpt: 0.000614 acc_pose: 0.853207 loss: 0.000614 2022/09/26 19:11:14 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:11:30 - mmengine - INFO - Epoch(train) [161][150/293] lr: 5.000000e-04 eta: 2:03:06 time: 0.564880 data_time: 0.077819 memory: 9504 loss_kpt: 0.000589 acc_pose: 0.836583 loss: 0.000589 2022/09/26 19:12:01 - mmengine - INFO - Epoch(train) [161][200/293] lr: 5.000000e-04 eta: 2:02:42 time: 0.607077 data_time: 0.119557 memory: 9504 loss_kpt: 0.000605 acc_pose: 0.815319 loss: 0.000605 2022/09/26 19:12:28 - mmengine - INFO - Epoch(train) [161][250/293] lr: 5.000000e-04 eta: 2:02:17 time: 0.552160 data_time: 0.088780 memory: 9504 loss_kpt: 0.000610 acc_pose: 0.899834 loss: 0.000610 2022/09/26 19:12:51 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:13:21 - mmengine - INFO - Epoch(train) [162][50/293] lr: 5.000000e-04 eta: 2:01:25 time: 0.588022 data_time: 0.175743 memory: 9504 loss_kpt: 0.000585 acc_pose: 0.852299 loss: 0.000585 2022/09/26 19:13:50 - mmengine - INFO - Epoch(train) [162][100/293] lr: 5.000000e-04 eta: 2:01:00 time: 0.589467 data_time: 0.143500 memory: 9504 loss_kpt: 0.000594 acc_pose: 0.903952 loss: 0.000594 2022/09/26 19:14:19 - mmengine - INFO - Epoch(train) [162][150/293] lr: 5.000000e-04 eta: 2:00:36 time: 0.578550 data_time: 0.104794 memory: 9504 loss_kpt: 0.000604 acc_pose: 0.843207 loss: 0.000604 2022/09/26 19:14:47 - mmengine - INFO - Epoch(train) [162][200/293] lr: 5.000000e-04 eta: 2:00:11 time: 0.552688 data_time: 0.083151 memory: 9504 loss_kpt: 0.000593 acc_pose: 0.853968 loss: 0.000593 2022/09/26 19:15:15 - mmengine - INFO - Epoch(train) [162][250/293] lr: 5.000000e-04 eta: 1:59:46 time: 0.557894 data_time: 0.073954 memory: 9504 loss_kpt: 0.000596 acc_pose: 0.871295 loss: 0.000596 2022/09/26 19:15:39 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:16:09 - mmengine - INFO - Epoch(train) [163][50/293] lr: 5.000000e-04 eta: 1:58:54 time: 0.591602 data_time: 0.102395 memory: 9504 loss_kpt: 0.000586 acc_pose: 0.859184 loss: 0.000586 2022/09/26 19:16:38 - mmengine - INFO - Epoch(train) [163][100/293] lr: 5.000000e-04 eta: 1:58:29 time: 0.589691 data_time: 0.098137 memory: 9504 loss_kpt: 0.000597 acc_pose: 0.890373 loss: 0.000597 2022/09/26 19:17:07 - mmengine - INFO - Epoch(train) [163][150/293] lr: 5.000000e-04 eta: 1:58:05 time: 0.575713 data_time: 0.084854 memory: 9504 loss_kpt: 0.000599 acc_pose: 0.798197 loss: 0.000599 2022/09/26 19:17:34 - mmengine - INFO - Epoch(train) [163][200/293] lr: 5.000000e-04 eta: 1:57:40 time: 0.539558 data_time: 0.147731 memory: 9504 loss_kpt: 0.000607 acc_pose: 0.799784 loss: 0.000607 2022/09/26 19:18:03 - mmengine - INFO - Epoch(train) [163][250/293] lr: 5.000000e-04 eta: 1:57:15 time: 0.573603 data_time: 0.112010 memory: 9504 loss_kpt: 0.000602 acc_pose: 0.857188 loss: 0.000602 2022/09/26 19:18:28 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:18:57 - mmengine - INFO - Epoch(train) [164][50/293] lr: 5.000000e-04 eta: 1:56:23 time: 0.578446 data_time: 0.197613 memory: 9504 loss_kpt: 0.000584 acc_pose: 0.834447 loss: 0.000584 2022/09/26 19:19:25 - mmengine - INFO - Epoch(train) [164][100/293] lr: 5.000000e-04 eta: 1:55:58 time: 0.561800 data_time: 0.116524 memory: 9504 loss_kpt: 0.000590 acc_pose: 0.847481 loss: 0.000590 2022/09/26 19:19:54 - mmengine - INFO - Epoch(train) [164][150/293] lr: 5.000000e-04 eta: 1:55:34 time: 0.591695 data_time: 0.112466 memory: 9504 loss_kpt: 0.000599 acc_pose: 0.844557 loss: 0.000599 2022/09/26 19:20:23 - mmengine - INFO - Epoch(train) [164][200/293] lr: 5.000000e-04 eta: 1:55:09 time: 0.567709 data_time: 0.079767 memory: 9504 loss_kpt: 0.000609 acc_pose: 0.830164 loss: 0.000609 2022/09/26 19:20:47 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:20:53 - mmengine - INFO - Epoch(train) [164][250/293] lr: 5.000000e-04 eta: 1:54:45 time: 0.596410 data_time: 0.086577 memory: 9504 loss_kpt: 0.000592 acc_pose: 0.884297 loss: 0.000592 2022/09/26 19:21:16 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:21:46 - mmengine - INFO - Epoch(train) [165][50/293] lr: 5.000000e-04 eta: 1:53:53 time: 0.595319 data_time: 0.110170 memory: 9504 loss_kpt: 0.000599 acc_pose: 0.851047 loss: 0.000599 2022/09/26 19:22:13 - mmengine - INFO - Epoch(train) [165][100/293] lr: 5.000000e-04 eta: 1:53:28 time: 0.549628 data_time: 0.143124 memory: 9504 loss_kpt: 0.000600 acc_pose: 0.809862 loss: 0.000600 2022/09/26 19:22:42 - mmengine - INFO - Epoch(train) [165][150/293] lr: 5.000000e-04 eta: 1:53:03 time: 0.580425 data_time: 0.079853 memory: 9504 loss_kpt: 0.000598 acc_pose: 0.808563 loss: 0.000598 2022/09/26 19:23:12 - mmengine - INFO - Epoch(train) [165][200/293] lr: 5.000000e-04 eta: 1:52:39 time: 0.584286 data_time: 0.095003 memory: 9504 loss_kpt: 0.000608 acc_pose: 0.849285 loss: 0.000608 2022/09/26 19:23:40 - mmengine - INFO - Epoch(train) [165][250/293] lr: 5.000000e-04 eta: 1:52:14 time: 0.568723 data_time: 0.171488 memory: 9504 loss_kpt: 0.000591 acc_pose: 0.872549 loss: 0.000591 2022/09/26 19:24:04 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:24:32 - mmengine - INFO - Epoch(train) [166][50/293] lr: 5.000000e-04 eta: 1:51:22 time: 0.565707 data_time: 0.193548 memory: 9504 loss_kpt: 0.000604 acc_pose: 0.859647 loss: 0.000604 2022/09/26 19:25:01 - mmengine - INFO - Epoch(train) [166][100/293] lr: 5.000000e-04 eta: 1:50:57 time: 0.564641 data_time: 0.131826 memory: 9504 loss_kpt: 0.000591 acc_pose: 0.863681 loss: 0.000591 2022/09/26 19:25:30 - mmengine - INFO - Epoch(train) [166][150/293] lr: 5.000000e-04 eta: 1:50:33 time: 0.592186 data_time: 0.185639 memory: 9504 loss_kpt: 0.000595 acc_pose: 0.877446 loss: 0.000595 2022/09/26 19:26:00 - mmengine - INFO - Epoch(train) [166][200/293] lr: 5.000000e-04 eta: 1:50:08 time: 0.585177 data_time: 0.146567 memory: 9504 loss_kpt: 0.000597 acc_pose: 0.840677 loss: 0.000597 2022/09/26 19:26:27 - mmengine - INFO - Epoch(train) [166][250/293] lr: 5.000000e-04 eta: 1:49:43 time: 0.549500 data_time: 0.087003 memory: 9504 loss_kpt: 0.000595 acc_pose: 0.886006 loss: 0.000595 2022/09/26 19:26:52 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:27:22 - mmengine - INFO - Epoch(train) [167][50/293] lr: 5.000000e-04 eta: 1:48:52 time: 0.601712 data_time: 0.118477 memory: 9504 loss_kpt: 0.000603 acc_pose: 0.885664 loss: 0.000603 2022/09/26 19:27:50 - mmengine - INFO - Epoch(train) [167][100/293] lr: 5.000000e-04 eta: 1:48:27 time: 0.565851 data_time: 0.083720 memory: 9504 loss_kpt: 0.000595 acc_pose: 0.848472 loss: 0.000595 2022/09/26 19:28:21 - mmengine - INFO - Epoch(train) [167][150/293] lr: 5.000000e-04 eta: 1:48:03 time: 0.607575 data_time: 0.139388 memory: 9504 loss_kpt: 0.000604 acc_pose: 0.859966 loss: 0.000604 2022/09/26 19:28:48 - mmengine - INFO - Epoch(train) [167][200/293] lr: 5.000000e-04 eta: 1:47:38 time: 0.542274 data_time: 0.089610 memory: 9504 loss_kpt: 0.000594 acc_pose: 0.878571 loss: 0.000594 2022/09/26 19:29:17 - mmengine - INFO - Epoch(train) [167][250/293] lr: 5.000000e-04 eta: 1:47:13 time: 0.578165 data_time: 0.075345 memory: 9504 loss_kpt: 0.000600 acc_pose: 0.872403 loss: 0.000600 2022/09/26 19:29:41 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:30:11 - mmengine - INFO - Epoch(train) [168][50/293] lr: 5.000000e-04 eta: 1:46:21 time: 0.591109 data_time: 0.176600 memory: 9504 loss_kpt: 0.000575 acc_pose: 0.877419 loss: 0.000575 2022/09/26 19:30:22 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:30:38 - mmengine - INFO - Epoch(train) [168][100/293] lr: 5.000000e-04 eta: 1:45:56 time: 0.555345 data_time: 0.096437 memory: 9504 loss_kpt: 0.000594 acc_pose: 0.862417 loss: 0.000594 2022/09/26 19:31:07 - mmengine - INFO - Epoch(train) [168][150/293] lr: 5.000000e-04 eta: 1:45:32 time: 0.574168 data_time: 0.138423 memory: 9504 loss_kpt: 0.000592 acc_pose: 0.868499 loss: 0.000592 2022/09/26 19:31:37 - mmengine - INFO - Epoch(train) [168][200/293] lr: 5.000000e-04 eta: 1:45:08 time: 0.601701 data_time: 0.130187 memory: 9504 loss_kpt: 0.000585 acc_pose: 0.845318 loss: 0.000585 2022/09/26 19:32:04 - mmengine - INFO - Epoch(train) [168][250/293] lr: 5.000000e-04 eta: 1:44:43 time: 0.545927 data_time: 0.069785 memory: 9504 loss_kpt: 0.000593 acc_pose: 0.881226 loss: 0.000593 2022/09/26 19:32:29 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:32:59 - mmengine - INFO - Epoch(train) [169][50/293] lr: 5.000000e-04 eta: 1:43:51 time: 0.581353 data_time: 0.134149 memory: 9504 loss_kpt: 0.000586 acc_pose: 0.907503 loss: 0.000586 2022/09/26 19:33:27 - mmengine - INFO - Epoch(train) [169][100/293] lr: 5.000000e-04 eta: 1:43:26 time: 0.574296 data_time: 0.089541 memory: 9504 loss_kpt: 0.000583 acc_pose: 0.871881 loss: 0.000583 2022/09/26 19:33:56 - mmengine - INFO - Epoch(train) [169][150/293] lr: 5.000000e-04 eta: 1:43:01 time: 0.564953 data_time: 0.102699 memory: 9504 loss_kpt: 0.000594 acc_pose: 0.849825 loss: 0.000594 2022/09/26 19:34:24 - mmengine - INFO - Epoch(train) [169][200/293] lr: 5.000000e-04 eta: 1:42:37 time: 0.570095 data_time: 0.135033 memory: 9504 loss_kpt: 0.000601 acc_pose: 0.842249 loss: 0.000601 2022/09/26 19:34:53 - mmengine - INFO - Epoch(train) [169][250/293] lr: 5.000000e-04 eta: 1:42:12 time: 0.582159 data_time: 0.148504 memory: 9504 loss_kpt: 0.000602 acc_pose: 0.847914 loss: 0.000602 2022/09/26 19:35:17 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:35:47 - mmengine - INFO - Epoch(train) [170][50/293] lr: 5.000000e-04 eta: 1:41:21 time: 0.583708 data_time: 0.106885 memory: 9504 loss_kpt: 0.000591 acc_pose: 0.815526 loss: 0.000591 2022/09/26 19:36:16 - mmengine - INFO - Epoch(train) [170][100/293] lr: 5.000000e-04 eta: 1:40:56 time: 0.576394 data_time: 0.104796 memory: 9504 loss_kpt: 0.000600 acc_pose: 0.862233 loss: 0.000600 2022/09/26 19:36:44 - mmengine - INFO - Epoch(train) [170][150/293] lr: 5.000000e-04 eta: 1:40:31 time: 0.567269 data_time: 0.087863 memory: 9504 loss_kpt: 0.000598 acc_pose: 0.860747 loss: 0.000598 2022/09/26 19:37:13 - mmengine - INFO - Epoch(train) [170][200/293] lr: 5.000000e-04 eta: 1:40:07 time: 0.578183 data_time: 0.097252 memory: 9504 loss_kpt: 0.000600 acc_pose: 0.830102 loss: 0.000600 2022/09/26 19:37:42 - mmengine - INFO - Epoch(train) [170][250/293] lr: 5.000000e-04 eta: 1:39:42 time: 0.575967 data_time: 0.114485 memory: 9504 loss_kpt: 0.000606 acc_pose: 0.878809 loss: 0.000606 2022/09/26 19:38:06 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:38:06 - mmengine - INFO - Saving checkpoint at 170 epochs 2022/09/26 19:38:31 - mmengine - INFO - Epoch(val) [170][50/407] eta: 0:01:58 time: 0.331649 data_time: 0.160024 memory: 9504 2022/09/26 19:38:46 - mmengine - INFO - Epoch(val) [170][100/407] eta: 0:01:33 time: 0.304569 data_time: 0.136885 memory: 1378 2022/09/26 19:39:02 - mmengine - INFO - Epoch(val) [170][150/407] eta: 0:01:24 time: 0.326920 data_time: 0.142083 memory: 1378 2022/09/26 19:39:19 - mmengine - INFO - Epoch(val) [170][200/407] eta: 0:01:06 time: 0.322229 data_time: 0.156993 memory: 1378 2022/09/26 19:39:34 - mmengine - INFO - Epoch(val) [170][250/407] eta: 0:00:49 time: 0.316140 data_time: 0.135002 memory: 1378 2022/09/26 19:39:50 - mmengine - INFO - Epoch(val) [170][300/407] eta: 0:00:34 time: 0.321265 data_time: 0.152461 memory: 1378 2022/09/26 19:40:08 - mmengine - INFO - Epoch(val) [170][350/407] eta: 0:00:19 time: 0.347169 data_time: 0.186282 memory: 1378 2022/09/26 19:40:16 - mmengine - INFO - Epoch(val) [170][400/407] eta: 0:00:01 time: 0.173761 data_time: 0.075837 memory: 1378 2022/09/26 19:40:49 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 19:41:03 - mmengine - INFO - Epoch(val) [170][407/407] coco/AP: 0.714361 coco/AP .5: 0.896291 coco/AP .75: 0.792156 coco/AP (M): 0.677110 coco/AP (L): 0.784278 coco/AR: 0.771143 coco/AR .5: 0.935611 coco/AR .75: 0.839578 coco/AR (M): 0.726741 coco/AR (L): 0.835229 2022/09/26 19:41:34 - mmengine - INFO - Epoch(train) [171][50/293] lr: 5.000000e-05 eta: 1:38:51 time: 0.621675 data_time: 0.111892 memory: 9504 loss_kpt: 0.000582 acc_pose: 0.855210 loss: 0.000582 2022/09/26 19:42:03 - mmengine - INFO - Epoch(train) [171][100/293] lr: 5.000000e-05 eta: 1:38:26 time: 0.582974 data_time: 0.083941 memory: 9504 loss_kpt: 0.000575 acc_pose: 0.834995 loss: 0.000575 2022/09/26 19:42:32 - mmengine - INFO - Epoch(train) [171][150/293] lr: 5.000000e-05 eta: 1:38:02 time: 0.571177 data_time: 0.126367 memory: 9504 loss_kpt: 0.000569 acc_pose: 0.840675 loss: 0.000569 2022/09/26 19:42:55 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:43:01 - mmengine - INFO - Epoch(train) [171][200/293] lr: 5.000000e-05 eta: 1:37:37 time: 0.585004 data_time: 0.167209 memory: 9504 loss_kpt: 0.000581 acc_pose: 0.864852 loss: 0.000581 2022/09/26 19:43:29 - mmengine - INFO - Epoch(train) [171][250/293] lr: 5.000000e-05 eta: 1:37:12 time: 0.570295 data_time: 0.076147 memory: 9504 loss_kpt: 0.000581 acc_pose: 0.853885 loss: 0.000581 2022/09/26 19:43:54 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:44:24 - mmengine - INFO - Epoch(train) [172][50/293] lr: 5.000000e-05 eta: 1:36:21 time: 0.598061 data_time: 0.117288 memory: 9504 loss_kpt: 0.000577 acc_pose: 0.840509 loss: 0.000577 2022/09/26 19:44:54 - mmengine - INFO - Epoch(train) [172][100/293] lr: 5.000000e-05 eta: 1:35:57 time: 0.590605 data_time: 0.093044 memory: 9504 loss_kpt: 0.000566 acc_pose: 0.860433 loss: 0.000566 2022/09/26 19:45:23 - mmengine - INFO - Epoch(train) [172][150/293] lr: 5.000000e-05 eta: 1:35:32 time: 0.586652 data_time: 0.090498 memory: 9504 loss_kpt: 0.000574 acc_pose: 0.853008 loss: 0.000574 2022/09/26 19:45:50 - mmengine - INFO - Epoch(train) [172][200/293] lr: 5.000000e-05 eta: 1:35:07 time: 0.548528 data_time: 0.098733 memory: 9504 loss_kpt: 0.000565 acc_pose: 0.873506 loss: 0.000565 2022/09/26 19:46:19 - mmengine - INFO - Epoch(train) [172][250/293] lr: 5.000000e-05 eta: 1:34:42 time: 0.564303 data_time: 0.087237 memory: 9504 loss_kpt: 0.000565 acc_pose: 0.860664 loss: 0.000565 2022/09/26 19:46:44 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:47:13 - mmengine - INFO - Epoch(train) [173][50/293] lr: 5.000000e-05 eta: 1:33:51 time: 0.589068 data_time: 0.157329 memory: 9504 loss_kpt: 0.000563 acc_pose: 0.875478 loss: 0.000563 2022/09/26 19:47:42 - mmengine - INFO - Epoch(train) [173][100/293] lr: 5.000000e-05 eta: 1:33:27 time: 0.580615 data_time: 0.173483 memory: 9504 loss_kpt: 0.000566 acc_pose: 0.861139 loss: 0.000566 2022/09/26 19:48:11 - mmengine - INFO - Epoch(train) [173][150/293] lr: 5.000000e-05 eta: 1:33:02 time: 0.576454 data_time: 0.093857 memory: 9504 loss_kpt: 0.000568 acc_pose: 0.870573 loss: 0.000568 2022/09/26 19:48:40 - mmengine - INFO - Epoch(train) [173][200/293] lr: 5.000000e-05 eta: 1:32:37 time: 0.579753 data_time: 0.082418 memory: 9504 loss_kpt: 0.000569 acc_pose: 0.847425 loss: 0.000569 2022/09/26 19:49:09 - mmengine - INFO - Epoch(train) [173][250/293] lr: 5.000000e-05 eta: 1:32:13 time: 0.580838 data_time: 0.079213 memory: 9504 loss_kpt: 0.000562 acc_pose: 0.870295 loss: 0.000562 2022/09/26 19:49:33 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:50:03 - mmengine - INFO - Epoch(train) [174][50/293] lr: 5.000000e-05 eta: 1:31:22 time: 0.610826 data_time: 0.095495 memory: 9504 loss_kpt: 0.000570 acc_pose: 0.882584 loss: 0.000570 2022/09/26 19:50:31 - mmengine - INFO - Epoch(train) [174][100/293] lr: 5.000000e-05 eta: 1:30:57 time: 0.547348 data_time: 0.106484 memory: 9504 loss_kpt: 0.000562 acc_pose: 0.840398 loss: 0.000562 2022/09/26 19:51:00 - mmengine - INFO - Epoch(train) [174][150/293] lr: 5.000000e-05 eta: 1:30:32 time: 0.592108 data_time: 0.084325 memory: 9504 loss_kpt: 0.000558 acc_pose: 0.858461 loss: 0.000558 2022/09/26 19:51:28 - mmengine - INFO - Epoch(train) [174][200/293] lr: 5.000000e-05 eta: 1:30:07 time: 0.546644 data_time: 0.096696 memory: 9504 loss_kpt: 0.000569 acc_pose: 0.869324 loss: 0.000569 2022/09/26 19:51:55 - mmengine - INFO - Epoch(train) [174][250/293] lr: 5.000000e-05 eta: 1:29:42 time: 0.546658 data_time: 0.116429 memory: 9504 loss_kpt: 0.000558 acc_pose: 0.902716 loss: 0.000558 2022/09/26 19:52:19 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:52:30 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:52:49 - mmengine - INFO - Epoch(train) [175][50/293] lr: 5.000000e-05 eta: 1:28:52 time: 0.607157 data_time: 0.155226 memory: 9504 loss_kpt: 0.000554 acc_pose: 0.865982 loss: 0.000554 2022/09/26 19:53:17 - mmengine - INFO - Epoch(train) [175][100/293] lr: 5.000000e-05 eta: 1:28:27 time: 0.550361 data_time: 0.094095 memory: 9504 loss_kpt: 0.000563 acc_pose: 0.850845 loss: 0.000563 2022/09/26 19:53:46 - mmengine - INFO - Epoch(train) [175][150/293] lr: 5.000000e-05 eta: 1:28:02 time: 0.589588 data_time: 0.180241 memory: 9504 loss_kpt: 0.000564 acc_pose: 0.840572 loss: 0.000564 2022/09/26 19:54:15 - mmengine - INFO - Epoch(train) [175][200/293] lr: 5.000000e-05 eta: 1:27:37 time: 0.574991 data_time: 0.106275 memory: 9504 loss_kpt: 0.000558 acc_pose: 0.842980 loss: 0.000558 2022/09/26 19:54:44 - mmengine - INFO - Epoch(train) [175][250/293] lr: 5.000000e-05 eta: 1:27:13 time: 0.586950 data_time: 0.101154 memory: 9504 loss_kpt: 0.000562 acc_pose: 0.882346 loss: 0.000562 2022/09/26 19:55:10 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:55:40 - mmengine - INFO - Epoch(train) [176][50/293] lr: 5.000000e-05 eta: 1:26:22 time: 0.616175 data_time: 0.192560 memory: 9504 loss_kpt: 0.000554 acc_pose: 0.844061 loss: 0.000554 2022/09/26 19:56:09 - mmengine - INFO - Epoch(train) [176][100/293] lr: 5.000000e-05 eta: 1:25:57 time: 0.570682 data_time: 0.134949 memory: 9504 loss_kpt: 0.000556 acc_pose: 0.861911 loss: 0.000556 2022/09/26 19:56:38 - mmengine - INFO - Epoch(train) [176][150/293] lr: 5.000000e-05 eta: 1:25:33 time: 0.573928 data_time: 0.122827 memory: 9504 loss_kpt: 0.000571 acc_pose: 0.806550 loss: 0.000571 2022/09/26 19:57:06 - mmengine - INFO - Epoch(train) [176][200/293] lr: 5.000000e-05 eta: 1:25:08 time: 0.565495 data_time: 0.145007 memory: 9504 loss_kpt: 0.000573 acc_pose: 0.865294 loss: 0.000573 2022/09/26 19:57:35 - mmengine - INFO - Epoch(train) [176][250/293] lr: 5.000000e-05 eta: 1:24:43 time: 0.585338 data_time: 0.136338 memory: 9504 loss_kpt: 0.000568 acc_pose: 0.838712 loss: 0.000568 2022/09/26 19:57:59 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 19:58:29 - mmengine - INFO - Epoch(train) [177][50/293] lr: 5.000000e-05 eta: 1:23:53 time: 0.595639 data_time: 0.094444 memory: 9504 loss_kpt: 0.000557 acc_pose: 0.884267 loss: 0.000557 2022/09/26 19:58:56 - mmengine - INFO - Epoch(train) [177][100/293] lr: 5.000000e-05 eta: 1:23:27 time: 0.538715 data_time: 0.151050 memory: 9504 loss_kpt: 0.000557 acc_pose: 0.815699 loss: 0.000557 2022/09/26 19:59:25 - mmengine - INFO - Epoch(train) [177][150/293] lr: 5.000000e-05 eta: 1:23:03 time: 0.572304 data_time: 0.095564 memory: 9504 loss_kpt: 0.000561 acc_pose: 0.878136 loss: 0.000561 2022/09/26 19:59:53 - mmengine - INFO - Epoch(train) [177][200/293] lr: 5.000000e-05 eta: 1:22:38 time: 0.568809 data_time: 0.087858 memory: 9504 loss_kpt: 0.000553 acc_pose: 0.838133 loss: 0.000553 2022/09/26 20:00:21 - mmengine - INFO - Epoch(train) [177][250/293] lr: 5.000000e-05 eta: 1:22:13 time: 0.554701 data_time: 0.111641 memory: 9504 loss_kpt: 0.000566 acc_pose: 0.876877 loss: 0.000566 2022/09/26 20:00:45 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:01:15 - mmengine - INFO - Epoch(train) [178][50/293] lr: 5.000000e-05 eta: 1:21:23 time: 0.606607 data_time: 0.229783 memory: 9504 loss_kpt: 0.000551 acc_pose: 0.879584 loss: 0.000551 2022/09/26 20:01:44 - mmengine - INFO - Epoch(train) [178][100/293] lr: 5.000000e-05 eta: 1:20:58 time: 0.574067 data_time: 0.168497 memory: 9504 loss_kpt: 0.000556 acc_pose: 0.829249 loss: 0.000556 2022/09/26 20:02:06 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:02:12 - mmengine - INFO - Epoch(train) [178][150/293] lr: 5.000000e-05 eta: 1:20:33 time: 0.566512 data_time: 0.091653 memory: 9504 loss_kpt: 0.000560 acc_pose: 0.844260 loss: 0.000560 2022/09/26 20:02:41 - mmengine - INFO - Epoch(train) [178][200/293] lr: 5.000000e-05 eta: 1:20:08 time: 0.567261 data_time: 0.140691 memory: 9504 loss_kpt: 0.000563 acc_pose: 0.856432 loss: 0.000563 2022/09/26 20:03:10 - mmengine - INFO - Epoch(train) [178][250/293] lr: 5.000000e-05 eta: 1:19:43 time: 0.582038 data_time: 0.119894 memory: 9504 loss_kpt: 0.000559 acc_pose: 0.853279 loss: 0.000559 2022/09/26 20:03:34 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:04:04 - mmengine - INFO - Epoch(train) [179][50/293] lr: 5.000000e-05 eta: 1:18:53 time: 0.592107 data_time: 0.101399 memory: 9504 loss_kpt: 0.000560 acc_pose: 0.832313 loss: 0.000560 2022/09/26 20:04:32 - mmengine - INFO - Epoch(train) [179][100/293] lr: 5.000000e-05 eta: 1:18:28 time: 0.564719 data_time: 0.084430 memory: 9504 loss_kpt: 0.000557 acc_pose: 0.863523 loss: 0.000557 2022/09/26 20:05:01 - mmengine - INFO - Epoch(train) [179][150/293] lr: 5.000000e-05 eta: 1:18:03 time: 0.589438 data_time: 0.080513 memory: 9504 loss_kpt: 0.000565 acc_pose: 0.854347 loss: 0.000565 2022/09/26 20:05:31 - mmengine - INFO - Epoch(train) [179][200/293] lr: 5.000000e-05 eta: 1:17:39 time: 0.586890 data_time: 0.134328 memory: 9504 loss_kpt: 0.000551 acc_pose: 0.868557 loss: 0.000551 2022/09/26 20:05:59 - mmengine - INFO - Epoch(train) [179][250/293] lr: 5.000000e-05 eta: 1:17:14 time: 0.571105 data_time: 0.143075 memory: 9504 loss_kpt: 0.000561 acc_pose: 0.878108 loss: 0.000561 2022/09/26 20:06:25 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:06:55 - mmengine - INFO - Epoch(train) [180][50/293] lr: 5.000000e-05 eta: 1:16:24 time: 0.600726 data_time: 0.175759 memory: 9504 loss_kpt: 0.000559 acc_pose: 0.881101 loss: 0.000559 2022/09/26 20:07:24 - mmengine - INFO - Epoch(train) [180][100/293] lr: 5.000000e-05 eta: 1:15:59 time: 0.590363 data_time: 0.109883 memory: 9504 loss_kpt: 0.000554 acc_pose: 0.850408 loss: 0.000554 2022/09/26 20:07:52 - mmengine - INFO - Epoch(train) [180][150/293] lr: 5.000000e-05 eta: 1:15:34 time: 0.563418 data_time: 0.078065 memory: 9504 loss_kpt: 0.000556 acc_pose: 0.849340 loss: 0.000556 2022/09/26 20:08:21 - mmengine - INFO - Epoch(train) [180][200/293] lr: 5.000000e-05 eta: 1:15:09 time: 0.570826 data_time: 0.125282 memory: 9504 loss_kpt: 0.000554 acc_pose: 0.844438 loss: 0.000554 2022/09/26 20:08:49 - mmengine - INFO - Epoch(train) [180][250/293] lr: 5.000000e-05 eta: 1:14:44 time: 0.562084 data_time: 0.131757 memory: 9504 loss_kpt: 0.000558 acc_pose: 0.880623 loss: 0.000558 2022/09/26 20:09:14 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:09:14 - mmengine - INFO - Saving checkpoint at 180 epochs 2022/09/26 20:09:38 - mmengine - INFO - Epoch(val) [180][50/407] eta: 0:01:53 time: 0.318864 data_time: 0.141820 memory: 9504 2022/09/26 20:09:55 - mmengine - INFO - Epoch(val) [180][100/407] eta: 0:01:41 time: 0.331150 data_time: 0.176655 memory: 1378 2022/09/26 20:10:10 - mmengine - INFO - Epoch(val) [180][150/407] eta: 0:01:18 time: 0.305339 data_time: 0.126484 memory: 1378 2022/09/26 20:10:25 - mmengine - INFO - Epoch(val) [180][200/407] eta: 0:01:04 time: 0.310474 data_time: 0.142657 memory: 1378 2022/09/26 20:10:41 - mmengine - INFO - Epoch(val) [180][250/407] eta: 0:00:48 time: 0.308191 data_time: 0.137278 memory: 1378 2022/09/26 20:10:57 - mmengine - INFO - Epoch(val) [180][300/407] eta: 0:00:34 time: 0.325539 data_time: 0.165395 memory: 1378 2022/09/26 20:11:14 - mmengine - INFO - Epoch(val) [180][350/407] eta: 0:00:19 time: 0.334907 data_time: 0.178582 memory: 1378 2022/09/26 20:11:25 - mmengine - INFO - Epoch(val) [180][400/407] eta: 0:00:01 time: 0.215602 data_time: 0.110009 memory: 1378 2022/09/26 20:11:57 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 20:12:10 - mmengine - INFO - Epoch(val) [180][407/407] coco/AP: 0.724692 coco/AP .5: 0.900180 coco/AP .75: 0.804889 coco/AP (M): 0.688680 coco/AP (L): 0.791726 coco/AR: 0.780337 coco/AR .5: 0.939704 coco/AR .75: 0.850756 coco/AR (M): 0.738459 coco/AR (L): 0.841100 2022/09/26 20:12:10 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_160.pth is removed 2022/09/26 20:12:12 - mmengine - INFO - The best checkpoint with 0.7247 coco/AP at 180 epoch is saved to best_coco/AP_epoch_180.pth. 2022/09/26 20:12:41 - mmengine - INFO - Epoch(train) [181][50/293] lr: 5.000000e-05 eta: 1:13:54 time: 0.572311 data_time: 0.197943 memory: 9504 loss_kpt: 0.000568 acc_pose: 0.824230 loss: 0.000568 2022/09/26 20:13:11 - mmengine - INFO - Epoch(train) [181][100/293] lr: 5.000000e-05 eta: 1:13:29 time: 0.595735 data_time: 0.194356 memory: 9504 loss_kpt: 0.000553 acc_pose: 0.892719 loss: 0.000553 2022/09/26 20:13:40 - mmengine - INFO - Epoch(train) [181][150/293] lr: 5.000000e-05 eta: 1:13:05 time: 0.587167 data_time: 0.129155 memory: 9504 loss_kpt: 0.000570 acc_pose: 0.864573 loss: 0.000570 2022/09/26 20:14:08 - mmengine - INFO - Epoch(train) [181][200/293] lr: 5.000000e-05 eta: 1:12:40 time: 0.564550 data_time: 0.085528 memory: 9504 loss_kpt: 0.000560 acc_pose: 0.854143 loss: 0.000560 2022/09/26 20:14:38 - mmengine - INFO - Epoch(train) [181][250/293] lr: 5.000000e-05 eta: 1:12:15 time: 0.586262 data_time: 0.094473 memory: 9504 loss_kpt: 0.000554 acc_pose: 0.851609 loss: 0.000554 2022/09/26 20:14:43 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:15:03 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:15:33 - mmengine - INFO - Epoch(train) [182][50/293] lr: 5.000000e-05 eta: 1:11:25 time: 0.592690 data_time: 0.104099 memory: 9504 loss_kpt: 0.000563 acc_pose: 0.841823 loss: 0.000563 2022/09/26 20:16:00 - mmengine - INFO - Epoch(train) [182][100/293] lr: 5.000000e-05 eta: 1:11:00 time: 0.551521 data_time: 0.070844 memory: 9504 loss_kpt: 0.000548 acc_pose: 0.850510 loss: 0.000548 2022/09/26 20:16:29 - mmengine - INFO - Epoch(train) [182][150/293] lr: 5.000000e-05 eta: 1:10:35 time: 0.581918 data_time: 0.080657 memory: 9504 loss_kpt: 0.000544 acc_pose: 0.861571 loss: 0.000544 2022/09/26 20:16:59 - mmengine - INFO - Epoch(train) [182][200/293] lr: 5.000000e-05 eta: 1:10:10 time: 0.585975 data_time: 0.098329 memory: 9504 loss_kpt: 0.000548 acc_pose: 0.877975 loss: 0.000548 2022/09/26 20:17:26 - mmengine - INFO - Epoch(train) [182][250/293] lr: 5.000000e-05 eta: 1:09:45 time: 0.552702 data_time: 0.076291 memory: 9504 loss_kpt: 0.000569 acc_pose: 0.861005 loss: 0.000569 2022/09/26 20:17:52 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:18:22 - mmengine - INFO - Epoch(train) [183][50/293] lr: 5.000000e-05 eta: 1:08:56 time: 0.602697 data_time: 0.144483 memory: 9504 loss_kpt: 0.000567 acc_pose: 0.849389 loss: 0.000567 2022/09/26 20:18:51 - mmengine - INFO - Epoch(train) [183][100/293] lr: 5.000000e-05 eta: 1:08:31 time: 0.587488 data_time: 0.078572 memory: 9504 loss_kpt: 0.000545 acc_pose: 0.870472 loss: 0.000545 2022/09/26 20:19:20 - mmengine - INFO - Epoch(train) [183][150/293] lr: 5.000000e-05 eta: 1:08:06 time: 0.570261 data_time: 0.133245 memory: 9504 loss_kpt: 0.000573 acc_pose: 0.859157 loss: 0.000573 2022/09/26 20:19:49 - mmengine - INFO - Epoch(train) [183][200/293] lr: 5.000000e-05 eta: 1:07:41 time: 0.583917 data_time: 0.123761 memory: 9504 loss_kpt: 0.000564 acc_pose: 0.836318 loss: 0.000564 2022/09/26 20:20:18 - mmengine - INFO - Epoch(train) [183][250/293] lr: 5.000000e-05 eta: 1:07:16 time: 0.582681 data_time: 0.066699 memory: 9504 loss_kpt: 0.000556 acc_pose: 0.884371 loss: 0.000556 2022/09/26 20:20:43 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:21:13 - mmengine - INFO - Epoch(train) [184][50/293] lr: 5.000000e-05 eta: 1:06:27 time: 0.606841 data_time: 0.179981 memory: 9504 loss_kpt: 0.000557 acc_pose: 0.875002 loss: 0.000557 2022/09/26 20:21:42 - mmengine - INFO - Epoch(train) [184][100/293] lr: 5.000000e-05 eta: 1:06:02 time: 0.572138 data_time: 0.120693 memory: 9504 loss_kpt: 0.000550 acc_pose: 0.829331 loss: 0.000550 2022/09/26 20:22:11 - mmengine - INFO - Epoch(train) [184][150/293] lr: 5.000000e-05 eta: 1:05:37 time: 0.574234 data_time: 0.256314 memory: 9504 loss_kpt: 0.000552 acc_pose: 0.888461 loss: 0.000552 2022/09/26 20:22:40 - mmengine - INFO - Epoch(train) [184][200/293] lr: 5.000000e-05 eta: 1:05:12 time: 0.585045 data_time: 0.162217 memory: 9504 loss_kpt: 0.000549 acc_pose: 0.859090 loss: 0.000549 2022/09/26 20:23:08 - mmengine - INFO - Epoch(train) [184][250/293] lr: 5.000000e-05 eta: 1:04:47 time: 0.564098 data_time: 0.116220 memory: 9504 loss_kpt: 0.000559 acc_pose: 0.856724 loss: 0.000559 2022/09/26 20:23:33 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:24:03 - mmengine - INFO - Epoch(train) [185][50/293] lr: 5.000000e-05 eta: 1:03:58 time: 0.609734 data_time: 0.099107 memory: 9504 loss_kpt: 0.000556 acc_pose: 0.859617 loss: 0.000556 2022/09/26 20:24:25 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:24:32 - mmengine - INFO - Epoch(train) [185][100/293] lr: 5.000000e-05 eta: 1:03:33 time: 0.570673 data_time: 0.080585 memory: 9504 loss_kpt: 0.000547 acc_pose: 0.902689 loss: 0.000547 2022/09/26 20:25:00 - mmengine - INFO - Epoch(train) [185][150/293] lr: 5.000000e-05 eta: 1:03:08 time: 0.570675 data_time: 0.082503 memory: 9504 loss_kpt: 0.000544 acc_pose: 0.872022 loss: 0.000544 2022/09/26 20:25:29 - mmengine - INFO - Epoch(train) [185][200/293] lr: 5.000000e-05 eta: 1:02:43 time: 0.580197 data_time: 0.099611 memory: 9504 loss_kpt: 0.000557 acc_pose: 0.833066 loss: 0.000557 2022/09/26 20:25:58 - mmengine - INFO - Epoch(train) [185][250/293] lr: 5.000000e-05 eta: 1:02:18 time: 0.578081 data_time: 0.135509 memory: 9504 loss_kpt: 0.000551 acc_pose: 0.836265 loss: 0.000551 2022/09/26 20:26:23 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:26:54 - mmengine - INFO - Epoch(train) [186][50/293] lr: 5.000000e-05 eta: 1:01:29 time: 0.615718 data_time: 0.100200 memory: 9504 loss_kpt: 0.000549 acc_pose: 0.857438 loss: 0.000549 2022/09/26 20:27:23 - mmengine - INFO - Epoch(train) [186][100/293] lr: 5.000000e-05 eta: 1:01:04 time: 0.577070 data_time: 0.102664 memory: 9504 loss_kpt: 0.000546 acc_pose: 0.898841 loss: 0.000546 2022/09/26 20:27:51 - mmengine - INFO - Epoch(train) [186][150/293] lr: 5.000000e-05 eta: 1:00:39 time: 0.558551 data_time: 0.091256 memory: 9504 loss_kpt: 0.000551 acc_pose: 0.900877 loss: 0.000551 2022/09/26 20:28:19 - mmengine - INFO - Epoch(train) [186][200/293] lr: 5.000000e-05 eta: 1:00:14 time: 0.571714 data_time: 0.079301 memory: 9504 loss_kpt: 0.000547 acc_pose: 0.823352 loss: 0.000547 2022/09/26 20:28:48 - mmengine - INFO - Epoch(train) [186][250/293] lr: 5.000000e-05 eta: 0:59:49 time: 0.576148 data_time: 0.089875 memory: 9504 loss_kpt: 0.000550 acc_pose: 0.873781 loss: 0.000550 2022/09/26 20:29:13 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:29:43 - mmengine - INFO - Epoch(train) [187][50/293] lr: 5.000000e-05 eta: 0:58:59 time: 0.588406 data_time: 0.098919 memory: 9504 loss_kpt: 0.000549 acc_pose: 0.859200 loss: 0.000549 2022/09/26 20:30:13 - mmengine - INFO - Epoch(train) [187][100/293] lr: 5.000000e-05 eta: 0:58:35 time: 0.593524 data_time: 0.124245 memory: 9504 loss_kpt: 0.000548 acc_pose: 0.879994 loss: 0.000548 2022/09/26 20:30:41 - mmengine - INFO - Epoch(train) [187][150/293] lr: 5.000000e-05 eta: 0:58:10 time: 0.568401 data_time: 0.152733 memory: 9504 loss_kpt: 0.000553 acc_pose: 0.856388 loss: 0.000553 2022/09/26 20:31:09 - mmengine - INFO - Epoch(train) [187][200/293] lr: 5.000000e-05 eta: 0:57:45 time: 0.561281 data_time: 0.207005 memory: 9504 loss_kpt: 0.000551 acc_pose: 0.836403 loss: 0.000551 2022/09/26 20:31:39 - mmengine - INFO - Epoch(train) [187][250/293] lr: 5.000000e-05 eta: 0:57:20 time: 0.590323 data_time: 0.133394 memory: 9504 loss_kpt: 0.000557 acc_pose: 0.906366 loss: 0.000557 2022/09/26 20:32:03 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:32:34 - mmengine - INFO - Epoch(train) [188][50/293] lr: 5.000000e-05 eta: 0:56:31 time: 0.618502 data_time: 0.103392 memory: 9504 loss_kpt: 0.000561 acc_pose: 0.852804 loss: 0.000561 2022/09/26 20:33:03 - mmengine - INFO - Epoch(train) [188][100/293] lr: 5.000000e-05 eta: 0:56:06 time: 0.582326 data_time: 0.075541 memory: 9504 loss_kpt: 0.000552 acc_pose: 0.875269 loss: 0.000552 2022/09/26 20:33:32 - mmengine - INFO - Epoch(train) [188][150/293] lr: 5.000000e-05 eta: 0:55:41 time: 0.584576 data_time: 0.082185 memory: 9504 loss_kpt: 0.000549 acc_pose: 0.832243 loss: 0.000549 2022/09/26 20:34:02 - mmengine - INFO - Epoch(train) [188][200/293] lr: 5.000000e-05 eta: 0:55:16 time: 0.592144 data_time: 0.083942 memory: 9504 loss_kpt: 0.000551 acc_pose: 0.868746 loss: 0.000551 2022/09/26 20:34:07 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:34:30 - mmengine - INFO - Epoch(train) [188][250/293] lr: 5.000000e-05 eta: 0:54:51 time: 0.577658 data_time: 0.149113 memory: 9504 loss_kpt: 0.000560 acc_pose: 0.878193 loss: 0.000560 2022/09/26 20:34:55 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:35:25 - mmengine - INFO - Epoch(train) [189][50/293] lr: 5.000000e-05 eta: 0:54:02 time: 0.588268 data_time: 0.138906 memory: 9504 loss_kpt: 0.000547 acc_pose: 0.882908 loss: 0.000547 2022/09/26 20:35:53 - mmengine - INFO - Epoch(train) [189][100/293] lr: 5.000000e-05 eta: 0:53:37 time: 0.564467 data_time: 0.087072 memory: 9504 loss_kpt: 0.000546 acc_pose: 0.860758 loss: 0.000546 2022/09/26 20:36:20 - mmengine - INFO - Epoch(train) [189][150/293] lr: 5.000000e-05 eta: 0:53:12 time: 0.553298 data_time: 0.151351 memory: 9504 loss_kpt: 0.000557 acc_pose: 0.878842 loss: 0.000557 2022/09/26 20:36:49 - mmengine - INFO - Epoch(train) [189][200/293] lr: 5.000000e-05 eta: 0:52:47 time: 0.562609 data_time: 0.098585 memory: 9504 loss_kpt: 0.000550 acc_pose: 0.853565 loss: 0.000550 2022/09/26 20:37:17 - mmengine - INFO - Epoch(train) [189][250/293] lr: 5.000000e-05 eta: 0:52:22 time: 0.572219 data_time: 0.154629 memory: 9504 loss_kpt: 0.000555 acc_pose: 0.834899 loss: 0.000555 2022/09/26 20:37:41 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:38:12 - mmengine - INFO - Epoch(train) [190][50/293] lr: 5.000000e-05 eta: 0:51:33 time: 0.615206 data_time: 0.164341 memory: 9504 loss_kpt: 0.000542 acc_pose: 0.886460 loss: 0.000542 2022/09/26 20:38:40 - mmengine - INFO - Epoch(train) [190][100/293] lr: 5.000000e-05 eta: 0:51:08 time: 0.567706 data_time: 0.075341 memory: 9504 loss_kpt: 0.000556 acc_pose: 0.906860 loss: 0.000556 2022/09/26 20:39:10 - mmengine - INFO - Epoch(train) [190][150/293] lr: 5.000000e-05 eta: 0:50:43 time: 0.597424 data_time: 0.119593 memory: 9504 loss_kpt: 0.000543 acc_pose: 0.872363 loss: 0.000543 2022/09/26 20:39:39 - mmengine - INFO - Epoch(train) [190][200/293] lr: 5.000000e-05 eta: 0:50:18 time: 0.592761 data_time: 0.123205 memory: 9504 loss_kpt: 0.000550 acc_pose: 0.865571 loss: 0.000550 2022/09/26 20:40:08 - mmengine - INFO - Epoch(train) [190][250/293] lr: 5.000000e-05 eta: 0:49:53 time: 0.574301 data_time: 0.107718 memory: 9504 loss_kpt: 0.000537 acc_pose: 0.865649 loss: 0.000537 2022/09/26 20:40:33 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:40:33 - mmengine - INFO - Saving checkpoint at 190 epochs 2022/09/26 20:40:57 - mmengine - INFO - Epoch(val) [190][50/407] eta: 0:01:57 time: 0.328729 data_time: 0.165233 memory: 9504 2022/09/26 20:41:13 - mmengine - INFO - Epoch(val) [190][100/407] eta: 0:01:35 time: 0.310083 data_time: 0.142778 memory: 1378 2022/09/26 20:41:29 - mmengine - INFO - Epoch(val) [190][150/407] eta: 0:01:22 time: 0.322255 data_time: 0.143469 memory: 1378 2022/09/26 20:41:43 - mmengine - INFO - Epoch(val) [190][200/407] eta: 0:01:00 time: 0.291253 data_time: 0.114511 memory: 1378 2022/09/26 20:41:59 - mmengine - INFO - Epoch(val) [190][250/407] eta: 0:00:47 time: 0.300864 data_time: 0.136710 memory: 1378 2022/09/26 20:42:15 - mmengine - INFO - Epoch(val) [190][300/407] eta: 0:00:34 time: 0.322578 data_time: 0.157062 memory: 1378 2022/09/26 20:42:31 - mmengine - INFO - Epoch(val) [190][350/407] eta: 0:00:18 time: 0.317435 data_time: 0.147588 memory: 1378 2022/09/26 20:42:42 - mmengine - INFO - Epoch(val) [190][400/407] eta: 0:00:01 time: 0.231008 data_time: 0.107298 memory: 1378 2022/09/26 20:43:15 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 20:43:28 - mmengine - INFO - Epoch(val) [190][407/407] coco/AP: 0.727120 coco/AP .5: 0.903583 coco/AP .75: 0.807717 coco/AP (M): 0.688713 coco/AP (L): 0.794953 coco/AR: 0.781392 coco/AR .5: 0.940176 coco/AR .75: 0.851385 coco/AR (M): 0.737531 coco/AR (L): 0.844407 2022/09/26 20:43:28 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_180.pth is removed 2022/09/26 20:43:31 - mmengine - INFO - The best checkpoint with 0.7271 coco/AP at 190 epoch is saved to best_coco/AP_epoch_190.pth. 2022/09/26 20:44:00 - mmengine - INFO - Epoch(train) [191][50/293] lr: 5.000000e-05 eta: 0:49:04 time: 0.569479 data_time: 0.106960 memory: 9504 loss_kpt: 0.000543 acc_pose: 0.893801 loss: 0.000543 2022/09/26 20:44:29 - mmengine - INFO - Epoch(train) [191][100/293] lr: 5.000000e-05 eta: 0:48:39 time: 0.579896 data_time: 0.164198 memory: 9504 loss_kpt: 0.000544 acc_pose: 0.836169 loss: 0.000544 2022/09/26 20:44:57 - mmengine - INFO - Epoch(train) [191][150/293] lr: 5.000000e-05 eta: 0:48:14 time: 0.562648 data_time: 0.095228 memory: 9504 loss_kpt: 0.000538 acc_pose: 0.891242 loss: 0.000538 2022/09/26 20:45:27 - mmengine - INFO - Epoch(train) [191][200/293] lr: 5.000000e-05 eta: 0:47:49 time: 0.611468 data_time: 0.096458 memory: 9504 loss_kpt: 0.000559 acc_pose: 0.882674 loss: 0.000559 2022/09/26 20:45:57 - mmengine - INFO - Epoch(train) [191][250/293] lr: 5.000000e-05 eta: 0:47:24 time: 0.584172 data_time: 0.129235 memory: 9504 loss_kpt: 0.000546 acc_pose: 0.815335 loss: 0.000546 2022/09/26 20:46:21 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:46:44 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:46:51 - mmengine - INFO - Epoch(train) [192][50/293] lr: 5.000000e-05 eta: 0:46:35 time: 0.603076 data_time: 0.144752 memory: 9504 loss_kpt: 0.000547 acc_pose: 0.885133 loss: 0.000547 2022/09/26 20:47:22 - mmengine - INFO - Epoch(train) [192][100/293] lr: 5.000000e-05 eta: 0:46:10 time: 0.607744 data_time: 0.078215 memory: 9504 loss_kpt: 0.000543 acc_pose: 0.839378 loss: 0.000543 2022/09/26 20:47:51 - mmengine - INFO - Epoch(train) [192][150/293] lr: 5.000000e-05 eta: 0:45:45 time: 0.594558 data_time: 0.181306 memory: 9504 loss_kpt: 0.000543 acc_pose: 0.866488 loss: 0.000543 2022/09/26 20:48:20 - mmengine - INFO - Epoch(train) [192][200/293] lr: 5.000000e-05 eta: 0:45:20 time: 0.565162 data_time: 0.120592 memory: 9504 loss_kpt: 0.000556 acc_pose: 0.899456 loss: 0.000556 2022/09/26 20:48:49 - mmengine - INFO - Epoch(train) [192][250/293] lr: 5.000000e-05 eta: 0:44:55 time: 0.586378 data_time: 0.105400 memory: 9504 loss_kpt: 0.000541 acc_pose: 0.873122 loss: 0.000541 2022/09/26 20:49:13 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:49:42 - mmengine - INFO - Epoch(train) [193][50/293] lr: 5.000000e-05 eta: 0:44:07 time: 0.580439 data_time: 0.102808 memory: 9504 loss_kpt: 0.000550 acc_pose: 0.847656 loss: 0.000550 2022/09/26 20:50:12 - mmengine - INFO - Epoch(train) [193][100/293] lr: 5.000000e-05 eta: 0:43:42 time: 0.597531 data_time: 0.153740 memory: 9504 loss_kpt: 0.000540 acc_pose: 0.874642 loss: 0.000540 2022/09/26 20:50:41 - mmengine - INFO - Epoch(train) [193][150/293] lr: 5.000000e-05 eta: 0:43:17 time: 0.569711 data_time: 0.099007 memory: 9504 loss_kpt: 0.000543 acc_pose: 0.861656 loss: 0.000543 2022/09/26 20:51:10 - mmengine - INFO - Epoch(train) [193][200/293] lr: 5.000000e-05 eta: 0:42:52 time: 0.590427 data_time: 0.084056 memory: 9504 loss_kpt: 0.000552 acc_pose: 0.873422 loss: 0.000552 2022/09/26 20:51:39 - mmengine - INFO - Epoch(train) [193][250/293] lr: 5.000000e-05 eta: 0:42:27 time: 0.581801 data_time: 0.104571 memory: 9504 loss_kpt: 0.000556 acc_pose: 0.839082 loss: 0.000556 2022/09/26 20:52:05 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:52:34 - mmengine - INFO - Epoch(train) [194][50/293] lr: 5.000000e-05 eta: 0:41:38 time: 0.594750 data_time: 0.197938 memory: 9504 loss_kpt: 0.000547 acc_pose: 0.850139 loss: 0.000547 2022/09/26 20:53:03 - mmengine - INFO - Epoch(train) [194][100/293] lr: 5.000000e-05 eta: 0:41:13 time: 0.564914 data_time: 0.213651 memory: 9504 loss_kpt: 0.000550 acc_pose: 0.855050 loss: 0.000550 2022/09/26 20:53:31 - mmengine - INFO - Epoch(train) [194][150/293] lr: 5.000000e-05 eta: 0:40:48 time: 0.576263 data_time: 0.098741 memory: 9504 loss_kpt: 0.000543 acc_pose: 0.875907 loss: 0.000543 2022/09/26 20:54:01 - mmengine - INFO - Epoch(train) [194][200/293] lr: 5.000000e-05 eta: 0:40:23 time: 0.584801 data_time: 0.087287 memory: 9504 loss_kpt: 0.000547 acc_pose: 0.884972 loss: 0.000547 2022/09/26 20:54:29 - mmengine - INFO - Epoch(train) [194][250/293] lr: 5.000000e-05 eta: 0:39:58 time: 0.574807 data_time: 0.075632 memory: 9504 loss_kpt: 0.000548 acc_pose: 0.893878 loss: 0.000548 2022/09/26 20:54:54 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:55:24 - mmengine - INFO - Epoch(train) [195][50/293] lr: 5.000000e-05 eta: 0:39:09 time: 0.590694 data_time: 0.082514 memory: 9504 loss_kpt: 0.000555 acc_pose: 0.866028 loss: 0.000555 2022/09/26 20:55:53 - mmengine - INFO - Epoch(train) [195][100/293] lr: 5.000000e-05 eta: 0:38:44 time: 0.594560 data_time: 0.109478 memory: 9504 loss_kpt: 0.000545 acc_pose: 0.853433 loss: 0.000545 2022/09/26 20:56:22 - mmengine - INFO - Epoch(train) [195][150/293] lr: 5.000000e-05 eta: 0:38:19 time: 0.569145 data_time: 0.078937 memory: 9504 loss_kpt: 0.000550 acc_pose: 0.854574 loss: 0.000550 2022/09/26 20:56:26 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:56:51 - mmengine - INFO - Epoch(train) [195][200/293] lr: 5.000000e-05 eta: 0:37:54 time: 0.576477 data_time: 0.083086 memory: 9504 loss_kpt: 0.000542 acc_pose: 0.836778 loss: 0.000542 2022/09/26 20:57:19 - mmengine - INFO - Epoch(train) [195][250/293] lr: 5.000000e-05 eta: 0:37:29 time: 0.573069 data_time: 0.145244 memory: 9504 loss_kpt: 0.000552 acc_pose: 0.906466 loss: 0.000552 2022/09/26 20:57:44 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 20:58:14 - mmengine - INFO - Epoch(train) [196][50/293] lr: 5.000000e-05 eta: 0:36:41 time: 0.590995 data_time: 0.098520 memory: 9504 loss_kpt: 0.000545 acc_pose: 0.875640 loss: 0.000545 2022/09/26 20:58:43 - mmengine - INFO - Epoch(train) [196][100/293] lr: 5.000000e-05 eta: 0:36:15 time: 0.587773 data_time: 0.083711 memory: 9504 loss_kpt: 0.000546 acc_pose: 0.879063 loss: 0.000546 2022/09/26 20:59:13 - mmengine - INFO - Epoch(train) [196][150/293] lr: 5.000000e-05 eta: 0:35:50 time: 0.588869 data_time: 0.090951 memory: 9504 loss_kpt: 0.000570 acc_pose: 0.880033 loss: 0.000570 2022/09/26 20:59:42 - mmengine - INFO - Epoch(train) [196][200/293] lr: 5.000000e-05 eta: 0:35:25 time: 0.578249 data_time: 0.071872 memory: 9504 loss_kpt: 0.000547 acc_pose: 0.863797 loss: 0.000547 2022/09/26 21:00:11 - mmengine - INFO - Epoch(train) [196][250/293] lr: 5.000000e-05 eta: 0:35:00 time: 0.579741 data_time: 0.082463 memory: 9504 loss_kpt: 0.000542 acc_pose: 0.841107 loss: 0.000542 2022/09/26 21:00:35 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:01:04 - mmengine - INFO - Epoch(train) [197][50/293] lr: 5.000000e-05 eta: 0:34:12 time: 0.587807 data_time: 0.090964 memory: 9504 loss_kpt: 0.000542 acc_pose: 0.873640 loss: 0.000542 2022/09/26 21:01:33 - mmengine - INFO - Epoch(train) [197][100/293] lr: 5.000000e-05 eta: 0:33:47 time: 0.573449 data_time: 0.091193 memory: 9504 loss_kpt: 0.000549 acc_pose: 0.879895 loss: 0.000549 2022/09/26 21:02:03 - mmengine - INFO - Epoch(train) [197][150/293] lr: 5.000000e-05 eta: 0:33:22 time: 0.596262 data_time: 0.128502 memory: 9504 loss_kpt: 0.000540 acc_pose: 0.892544 loss: 0.000540 2022/09/26 21:02:31 - mmengine - INFO - Epoch(train) [197][200/293] lr: 5.000000e-05 eta: 0:32:57 time: 0.571614 data_time: 0.108625 memory: 9504 loss_kpt: 0.000547 acc_pose: 0.836001 loss: 0.000547 2022/09/26 21:03:02 - mmengine - INFO - Epoch(train) [197][250/293] lr: 5.000000e-05 eta: 0:32:32 time: 0.602872 data_time: 0.087755 memory: 9504 loss_kpt: 0.000558 acc_pose: 0.855912 loss: 0.000558 2022/09/26 21:03:26 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:03:56 - mmengine - INFO - Epoch(train) [198][50/293] lr: 5.000000e-05 eta: 0:31:43 time: 0.610878 data_time: 0.107462 memory: 9504 loss_kpt: 0.000548 acc_pose: 0.863441 loss: 0.000548 2022/09/26 21:04:25 - mmengine - INFO - Epoch(train) [198][100/293] lr: 5.000000e-05 eta: 0:31:18 time: 0.573217 data_time: 0.090055 memory: 9504 loss_kpt: 0.000553 acc_pose: 0.845667 loss: 0.000553 2022/09/26 21:04:55 - mmengine - INFO - Epoch(train) [198][150/293] lr: 5.000000e-05 eta: 0:30:53 time: 0.591435 data_time: 0.095977 memory: 9504 loss_kpt: 0.000545 acc_pose: 0.859387 loss: 0.000545 2022/09/26 21:05:25 - mmengine - INFO - Epoch(train) [198][200/293] lr: 5.000000e-05 eta: 0:30:28 time: 0.601854 data_time: 0.151917 memory: 9504 loss_kpt: 0.000555 acc_pose: 0.878716 loss: 0.000555 2022/09/26 21:05:53 - mmengine - INFO - Epoch(train) [198][250/293] lr: 5.000000e-05 eta: 0:30:03 time: 0.563456 data_time: 0.158173 memory: 9504 loss_kpt: 0.000535 acc_pose: 0.856178 loss: 0.000535 2022/09/26 21:06:11 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:06:18 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:06:48 - mmengine - INFO - Epoch(train) [199][50/293] lr: 5.000000e-05 eta: 0:29:15 time: 0.604048 data_time: 0.104561 memory: 9504 loss_kpt: 0.000544 acc_pose: 0.858972 loss: 0.000544 2022/09/26 21:07:17 - mmengine - INFO - Epoch(train) [199][100/293] lr: 5.000000e-05 eta: 0:28:50 time: 0.577645 data_time: 0.092035 memory: 9504 loss_kpt: 0.000538 acc_pose: 0.876008 loss: 0.000538 2022/09/26 21:07:46 - mmengine - INFO - Epoch(train) [199][150/293] lr: 5.000000e-05 eta: 0:28:25 time: 0.578440 data_time: 0.076461 memory: 9504 loss_kpt: 0.000557 acc_pose: 0.857519 loss: 0.000557 2022/09/26 21:08:14 - mmengine - INFO - Epoch(train) [199][200/293] lr: 5.000000e-05 eta: 0:28:00 time: 0.573419 data_time: 0.176540 memory: 9504 loss_kpt: 0.000537 acc_pose: 0.897288 loss: 0.000537 2022/09/26 21:08:44 - mmengine - INFO - Epoch(train) [199][250/293] lr: 5.000000e-05 eta: 0:27:35 time: 0.593896 data_time: 0.092453 memory: 9504 loss_kpt: 0.000546 acc_pose: 0.879971 loss: 0.000546 2022/09/26 21:09:09 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:09:38 - mmengine - INFO - Epoch(train) [200][50/293] lr: 5.000000e-05 eta: 0:26:46 time: 0.589792 data_time: 0.107799 memory: 9504 loss_kpt: 0.000548 acc_pose: 0.863865 loss: 0.000548 2022/09/26 21:10:08 - mmengine - INFO - Epoch(train) [200][100/293] lr: 5.000000e-05 eta: 0:26:21 time: 0.585732 data_time: 0.096325 memory: 9504 loss_kpt: 0.000539 acc_pose: 0.851703 loss: 0.000539 2022/09/26 21:10:36 - mmengine - INFO - Epoch(train) [200][150/293] lr: 5.000000e-05 eta: 0:25:56 time: 0.569788 data_time: 0.120719 memory: 9504 loss_kpt: 0.000549 acc_pose: 0.868071 loss: 0.000549 2022/09/26 21:11:05 - mmengine - INFO - Epoch(train) [200][200/293] lr: 5.000000e-05 eta: 0:25:31 time: 0.575832 data_time: 0.112520 memory: 9504 loss_kpt: 0.000539 acc_pose: 0.858185 loss: 0.000539 2022/09/26 21:11:34 - mmengine - INFO - Epoch(train) [200][250/293] lr: 5.000000e-05 eta: 0:25:06 time: 0.573466 data_time: 0.129652 memory: 9504 loss_kpt: 0.000530 acc_pose: 0.885953 loss: 0.000530 2022/09/26 21:11:58 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:11:58 - mmengine - INFO - Saving checkpoint at 200 epochs 2022/09/26 21:12:23 - mmengine - INFO - Epoch(val) [200][50/407] eta: 0:01:53 time: 0.318468 data_time: 0.151996 memory: 9504 2022/09/26 21:12:39 - mmengine - INFO - Epoch(val) [200][100/407] eta: 0:01:39 time: 0.325442 data_time: 0.154880 memory: 1378 2022/09/26 21:12:55 - mmengine - INFO - Epoch(val) [200][150/407] eta: 0:01:21 time: 0.317144 data_time: 0.136437 memory: 1378 2022/09/26 21:13:11 - mmengine - INFO - Epoch(val) [200][200/407] eta: 0:01:05 time: 0.314295 data_time: 0.154166 memory: 1378 2022/09/26 21:13:27 - mmengine - INFO - Epoch(val) [200][250/407] eta: 0:00:50 time: 0.324507 data_time: 0.162176 memory: 1378 2022/09/26 21:13:43 - mmengine - INFO - Epoch(val) [200][300/407] eta: 0:00:34 time: 0.320844 data_time: 0.157169 memory: 1378 2022/09/26 21:13:59 - mmengine - INFO - Epoch(val) [200][350/407] eta: 0:00:17 time: 0.313649 data_time: 0.137912 memory: 1378 2022/09/26 21:14:08 - mmengine - INFO - Epoch(val) [200][400/407] eta: 0:00:01 time: 0.175063 data_time: 0.084799 memory: 1378 2022/09/26 21:14:41 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 21:14:54 - mmengine - INFO - Epoch(val) [200][407/407] coco/AP: 0.727823 coco/AP .5: 0.904500 coco/AP .75: 0.808990 coco/AP (M): 0.690762 coco/AP (L): 0.794454 coco/AR: 0.782683 coco/AR .5: 0.941908 coco/AR .75: 0.853747 coco/AR (M): 0.739716 coco/AR (L): 0.844519 2022/09/26 21:14:54 - mmengine - INFO - The previous best checkpoint /mnt/lustre/liqikai/work_dirs/20220926/r101_256/best_coco/AP_epoch_190.pth is removed 2022/09/26 21:14:56 - mmengine - INFO - The best checkpoint with 0.7278 coco/AP at 200 epoch is saved to best_coco/AP_epoch_200.pth. 2022/09/26 21:15:27 - mmengine - INFO - Epoch(train) [201][50/293] lr: 5.000000e-06 eta: 0:24:18 time: 0.601599 data_time: 0.152853 memory: 9504 loss_kpt: 0.000561 acc_pose: 0.869762 loss: 0.000561 2022/09/26 21:15:56 - mmengine - INFO - Epoch(train) [201][100/293] lr: 5.000000e-06 eta: 0:23:53 time: 0.580034 data_time: 0.081881 memory: 9504 loss_kpt: 0.000545 acc_pose: 0.900986 loss: 0.000545 2022/09/26 21:16:26 - mmengine - INFO - Epoch(train) [201][150/293] lr: 5.000000e-06 eta: 0:23:28 time: 0.612299 data_time: 0.087350 memory: 9504 loss_kpt: 0.000542 acc_pose: 0.899297 loss: 0.000542 2022/09/26 21:16:56 - mmengine - INFO - Epoch(train) [201][200/293] lr: 5.000000e-06 eta: 0:23:03 time: 0.588617 data_time: 0.114160 memory: 9504 loss_kpt: 0.000544 acc_pose: 0.864782 loss: 0.000544 2022/09/26 21:17:25 - mmengine - INFO - Epoch(train) [201][250/293] lr: 5.000000e-06 eta: 0:22:37 time: 0.589488 data_time: 0.087405 memory: 9504 loss_kpt: 0.000538 acc_pose: 0.888807 loss: 0.000538 2022/09/26 21:17:49 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:18:20 - mmengine - INFO - Epoch(train) [202][50/293] lr: 5.000000e-06 eta: 0:21:50 time: 0.615845 data_time: 0.099254 memory: 9504 loss_kpt: 0.000537 acc_pose: 0.871494 loss: 0.000537 2022/09/26 21:18:48 - mmengine - INFO - Epoch(train) [202][100/293] lr: 5.000000e-06 eta: 0:21:24 time: 0.571942 data_time: 0.095035 memory: 9504 loss_kpt: 0.000543 acc_pose: 0.884176 loss: 0.000543 2022/09/26 21:18:53 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:19:17 - mmengine - INFO - Epoch(train) [202][150/293] lr: 5.000000e-06 eta: 0:20:59 time: 0.575809 data_time: 0.082624 memory: 9504 loss_kpt: 0.000537 acc_pose: 0.893574 loss: 0.000537 2022/09/26 21:19:47 - mmengine - INFO - Epoch(train) [202][200/293] lr: 5.000000e-06 eta: 0:20:34 time: 0.597372 data_time: 0.076717 memory: 9504 loss_kpt: 0.000549 acc_pose: 0.893003 loss: 0.000549 2022/09/26 21:20:15 - mmengine - INFO - Epoch(train) [202][250/293] lr: 5.000000e-06 eta: 0:20:09 time: 0.569476 data_time: 0.075902 memory: 9504 loss_kpt: 0.000532 acc_pose: 0.854874 loss: 0.000532 2022/09/26 21:20:40 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:21:10 - mmengine - INFO - Epoch(train) [203][50/293] lr: 5.000000e-06 eta: 0:19:21 time: 0.611059 data_time: 0.098404 memory: 9504 loss_kpt: 0.000537 acc_pose: 0.832185 loss: 0.000537 2022/09/26 21:21:39 - mmengine - INFO - Epoch(train) [203][100/293] lr: 5.000000e-06 eta: 0:18:56 time: 0.578193 data_time: 0.103708 memory: 9504 loss_kpt: 0.000545 acc_pose: 0.878889 loss: 0.000545 2022/09/26 21:22:08 - mmengine - INFO - Epoch(train) [203][150/293] lr: 5.000000e-06 eta: 0:18:31 time: 0.577147 data_time: 0.101680 memory: 9504 loss_kpt: 0.000546 acc_pose: 0.882344 loss: 0.000546 2022/09/26 21:22:37 - mmengine - INFO - Epoch(train) [203][200/293] lr: 5.000000e-06 eta: 0:18:06 time: 0.572428 data_time: 0.108097 memory: 9504 loss_kpt: 0.000547 acc_pose: 0.906843 loss: 0.000547 2022/09/26 21:23:05 - mmengine - INFO - Epoch(train) [203][250/293] lr: 5.000000e-06 eta: 0:17:40 time: 0.566988 data_time: 0.109166 memory: 9504 loss_kpt: 0.000536 acc_pose: 0.880359 loss: 0.000536 2022/09/26 21:23:30 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:24:00 - mmengine - INFO - Epoch(train) [204][50/293] lr: 5.000000e-06 eta: 0:16:53 time: 0.613889 data_time: 0.133807 memory: 9504 loss_kpt: 0.000545 acc_pose: 0.908409 loss: 0.000545 2022/09/26 21:24:29 - mmengine - INFO - Epoch(train) [204][100/293] lr: 5.000000e-06 eta: 0:16:27 time: 0.572034 data_time: 0.083117 memory: 9504 loss_kpt: 0.000545 acc_pose: 0.858372 loss: 0.000545 2022/09/26 21:24:58 - mmengine - INFO - Epoch(train) [204][150/293] lr: 5.000000e-06 eta: 0:16:02 time: 0.571928 data_time: 0.108034 memory: 9504 loss_kpt: 0.000541 acc_pose: 0.842658 loss: 0.000541 2022/09/26 21:25:27 - mmengine - INFO - Epoch(train) [204][200/293] lr: 5.000000e-06 eta: 0:15:37 time: 0.579066 data_time: 0.108500 memory: 9504 loss_kpt: 0.000534 acc_pose: 0.846572 loss: 0.000534 2022/09/26 21:25:56 - mmengine - INFO - Epoch(train) [204][250/293] lr: 5.000000e-06 eta: 0:15:12 time: 0.589478 data_time: 0.141090 memory: 9504 loss_kpt: 0.000540 acc_pose: 0.843937 loss: 0.000540 2022/09/26 21:26:20 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:26:51 - mmengine - INFO - Epoch(train) [205][50/293] lr: 5.000000e-06 eta: 0:14:24 time: 0.614045 data_time: 0.136803 memory: 9504 loss_kpt: 0.000553 acc_pose: 0.841514 loss: 0.000553 2022/09/26 21:27:21 - mmengine - INFO - Epoch(train) [205][100/293] lr: 5.000000e-06 eta: 0:13:59 time: 0.588657 data_time: 0.070657 memory: 9504 loss_kpt: 0.000550 acc_pose: 0.883236 loss: 0.000550 2022/09/26 21:27:49 - mmengine - INFO - Epoch(train) [205][150/293] lr: 5.000000e-06 eta: 0:13:34 time: 0.577514 data_time: 0.092487 memory: 9504 loss_kpt: 0.000548 acc_pose: 0.849019 loss: 0.000548 2022/09/26 21:28:18 - mmengine - INFO - Epoch(train) [205][200/293] lr: 5.000000e-06 eta: 0:13:09 time: 0.576068 data_time: 0.083454 memory: 9504 loss_kpt: 0.000553 acc_pose: 0.848412 loss: 0.000553 2022/09/26 21:28:35 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:28:48 - mmengine - INFO - Epoch(train) [205][250/293] lr: 5.000000e-06 eta: 0:12:43 time: 0.587317 data_time: 0.095198 memory: 9504 loss_kpt: 0.000535 acc_pose: 0.857564 loss: 0.000535 2022/09/26 21:29:13 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:29:44 - mmengine - INFO - Epoch(train) [206][50/293] lr: 5.000000e-06 eta: 0:11:56 time: 0.616078 data_time: 0.119149 memory: 9504 loss_kpt: 0.000553 acc_pose: 0.898397 loss: 0.000553 2022/09/26 21:30:12 - mmengine - INFO - Epoch(train) [206][100/293] lr: 5.000000e-06 eta: 0:11:31 time: 0.559905 data_time: 0.083536 memory: 9504 loss_kpt: 0.000542 acc_pose: 0.869260 loss: 0.000542 2022/09/26 21:30:42 - mmengine - INFO - Epoch(train) [206][150/293] lr: 5.000000e-06 eta: 0:11:05 time: 0.594235 data_time: 0.098406 memory: 9504 loss_kpt: 0.000545 acc_pose: 0.850091 loss: 0.000545 2022/09/26 21:31:10 - mmengine - INFO - Epoch(train) [206][200/293] lr: 5.000000e-06 eta: 0:10:40 time: 0.564013 data_time: 0.087076 memory: 9504 loss_kpt: 0.000532 acc_pose: 0.849502 loss: 0.000532 2022/09/26 21:31:39 - mmengine - INFO - Epoch(train) [206][250/293] lr: 5.000000e-06 eta: 0:10:15 time: 0.586897 data_time: 0.078023 memory: 9504 loss_kpt: 0.000543 acc_pose: 0.818729 loss: 0.000543 2022/09/26 21:32:04 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:32:34 - mmengine - INFO - Epoch(train) [207][50/293] lr: 5.000000e-06 eta: 0:09:27 time: 0.600259 data_time: 0.110552 memory: 9504 loss_kpt: 0.000539 acc_pose: 0.870380 loss: 0.000539 2022/09/26 21:33:02 - mmengine - INFO - Epoch(train) [207][100/293] lr: 5.000000e-06 eta: 0:09:02 time: 0.569346 data_time: 0.094595 memory: 9504 loss_kpt: 0.000547 acc_pose: 0.864569 loss: 0.000547 2022/09/26 21:33:30 - mmengine - INFO - Epoch(train) [207][150/293] lr: 5.000000e-06 eta: 0:08:37 time: 0.545739 data_time: 0.083922 memory: 9504 loss_kpt: 0.000541 acc_pose: 0.852489 loss: 0.000541 2022/09/26 21:33:59 - mmengine - INFO - Epoch(train) [207][200/293] lr: 5.000000e-06 eta: 0:08:12 time: 0.589332 data_time: 0.117334 memory: 9504 loss_kpt: 0.000543 acc_pose: 0.865032 loss: 0.000543 2022/09/26 21:34:28 - mmengine - INFO - Epoch(train) [207][250/293] lr: 5.000000e-06 eta: 0:07:46 time: 0.571105 data_time: 0.075863 memory: 9504 loss_kpt: 0.000536 acc_pose: 0.841215 loss: 0.000536 2022/09/26 21:34:51 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:35:20 - mmengine - INFO - Epoch(train) [208][50/293] lr: 5.000000e-06 eta: 0:06:59 time: 0.589158 data_time: 0.205551 memory: 9504 loss_kpt: 0.000531 acc_pose: 0.855265 loss: 0.000531 2022/09/26 21:35:49 - mmengine - INFO - Epoch(train) [208][100/293] lr: 5.000000e-06 eta: 0:06:34 time: 0.571660 data_time: 0.136159 memory: 9504 loss_kpt: 0.000535 acc_pose: 0.850622 loss: 0.000535 2022/09/26 21:36:18 - mmengine - INFO - Epoch(train) [208][150/293] lr: 5.000000e-06 eta: 0:06:09 time: 0.585795 data_time: 0.110130 memory: 9504 loss_kpt: 0.000538 acc_pose: 0.865756 loss: 0.000538 2022/09/26 21:36:47 - mmengine - INFO - Epoch(train) [208][200/293] lr: 5.000000e-06 eta: 0:05:43 time: 0.572837 data_time: 0.222573 memory: 9504 loss_kpt: 0.000537 acc_pose: 0.885992 loss: 0.000537 2022/09/26 21:37:16 - mmengine - INFO - Epoch(train) [208][250/293] lr: 5.000000e-06 eta: 0:05:18 time: 0.574639 data_time: 0.170425 memory: 9504 loss_kpt: 0.000536 acc_pose: 0.865045 loss: 0.000536 2022/09/26 21:37:40 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:38:11 - mmengine - INFO - Epoch(train) [209][50/293] lr: 5.000000e-06 eta: 0:04:31 time: 0.619149 data_time: 0.120005 memory: 9504 loss_kpt: 0.000531 acc_pose: 0.887673 loss: 0.000531 2022/09/26 21:38:15 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:38:40 - mmengine - INFO - Epoch(train) [209][100/293] lr: 5.000000e-06 eta: 0:04:05 time: 0.574233 data_time: 0.087259 memory: 9504 loss_kpt: 0.000526 acc_pose: 0.914778 loss: 0.000526 2022/09/26 21:39:09 - mmengine - INFO - Epoch(train) [209][150/293] lr: 5.000000e-06 eta: 0:03:40 time: 0.579946 data_time: 0.075910 memory: 9504 loss_kpt: 0.000531 acc_pose: 0.886095 loss: 0.000531 2022/09/26 21:39:38 - mmengine - INFO - Epoch(train) [209][200/293] lr: 5.000000e-06 eta: 0:03:15 time: 0.569924 data_time: 0.076012 memory: 9504 loss_kpt: 0.000537 acc_pose: 0.854658 loss: 0.000537 2022/09/26 21:40:07 - mmengine - INFO - Epoch(train) [209][250/293] lr: 5.000000e-06 eta: 0:02:50 time: 0.584735 data_time: 0.092360 memory: 9504 loss_kpt: 0.000540 acc_pose: 0.864077 loss: 0.000540 2022/09/26 21:40:32 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:41:04 - mmengine - INFO - Epoch(train) [210][50/293] lr: 5.000000e-06 eta: 0:02:02 time: 0.639400 data_time: 0.145714 memory: 9504 loss_kpt: 0.000536 acc_pose: 0.873054 loss: 0.000536 2022/09/26 21:41:33 - mmengine - INFO - Epoch(train) [210][100/293] lr: 5.000000e-06 eta: 0:01:37 time: 0.584189 data_time: 0.091002 memory: 9504 loss_kpt: 0.000549 acc_pose: 0.858909 loss: 0.000549 2022/09/26 21:42:01 - mmengine - INFO - Epoch(train) [210][150/293] lr: 5.000000e-06 eta: 0:01:12 time: 0.572224 data_time: 0.082701 memory: 9504 loss_kpt: 0.000539 acc_pose: 0.862697 loss: 0.000539 2022/09/26 21:42:30 - mmengine - INFO - Epoch(train) [210][200/293] lr: 5.000000e-06 eta: 0:00:47 time: 0.572854 data_time: 0.085028 memory: 9504 loss_kpt: 0.000545 acc_pose: 0.881037 loss: 0.000545 2022/09/26 21:42:59 - mmengine - INFO - Epoch(train) [210][250/293] lr: 5.000000e-06 eta: 0:00:21 time: 0.587787 data_time: 0.155036 memory: 9504 loss_kpt: 0.000541 acc_pose: 0.892393 loss: 0.000541 2022/09/26 21:43:24 - mmengine - INFO - Exp name: td-hm_res101_8xb64-210e_coco-256x192_20220926_103629 2022/09/26 21:43:25 - mmengine - INFO - Saving checkpoint at 210 epochs 2022/09/26 21:43:48 - mmengine - INFO - Epoch(val) [210][50/407] eta: 0:01:56 time: 0.327139 data_time: 0.161244 memory: 9504 2022/09/26 21:44:05 - mmengine - INFO - Epoch(val) [210][100/407] eta: 0:01:43 time: 0.336933 data_time: 0.172051 memory: 1378 2022/09/26 21:44:21 - mmengine - INFO - Epoch(val) [210][150/407] eta: 0:01:20 time: 0.313428 data_time: 0.135260 memory: 1378 2022/09/26 21:44:38 - mmengine - INFO - Epoch(val) [210][200/407] eta: 0:01:11 time: 0.343066 data_time: 0.180788 memory: 1378 2022/09/26 21:44:52 - mmengine - INFO - Epoch(val) [210][250/407] eta: 0:00:45 time: 0.289485 data_time: 0.116797 memory: 1378 2022/09/26 21:45:08 - mmengine - INFO - Epoch(val) [210][300/407] eta: 0:00:33 time: 0.310647 data_time: 0.141843 memory: 1378 2022/09/26 21:45:24 - mmengine - INFO - Epoch(val) [210][350/407] eta: 0:00:18 time: 0.324258 data_time: 0.161278 memory: 1378 2022/09/26 21:45:35 - mmengine - INFO - Epoch(val) [210][400/407] eta: 0:00:01 time: 0.221448 data_time: 0.107224 memory: 1378 2022/09/26 21:46:08 - mmengine - INFO - Evaluating CocoMetric... 2022/09/26 21:46:21 - mmengine - INFO - Epoch(val) [210][407/407] coco/AP: 0.727488 coco/AP .5: 0.903843 coco/AP .75: 0.805468 coco/AP (M): 0.689653 coco/AP (L): 0.796178 coco/AR: 0.782557 coco/AR .5: 0.941121 coco/AR .75: 0.849654 coco/AR (M): 0.738569 coco/AR (L): 0.845968