2023/03/02 23:00:31 - mmengine - INFO - ------------------------------------------------------------ System environment: sys.platform: linux Python: 3.10.8 (main, Nov 24 2022, 14:13:03) [GCC 11.2.0] CUDA available: True numpy_random_seed: 42 GPU 0,1,2,3,4,5,6,7: NVIDIA A100-SXM4-80GB CUDA_HOME: None GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) 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: 4.7.0 MMEngine: 0.6.0 Runtime environment: cudnn_benchmark: False mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} dist_cfg: {'backend': 'nccl'} seed: 42 Distributed launcher: pytorch Distributed training: True GPU number: 8 ------------------------------------------------------------ 2023/03/02 23:00:33 - mmengine - INFO - Config: custom_imports = dict(imports=['spts'], allow_failed_imports=False) file_client_args = dict(backend='disk') dictionary = dict( type='SPTSDictionary', dict_file= 'mmocr/projects/SPTS/config/spts/../../dicts/spts.txt', with_start=True, with_end=True, with_seq_end=True, same_start_end=False, with_padding=True, with_unknown=True, unknown_token=None) num_bins = 1000 model = dict( type='SPTS', data_preprocessor=dict( type='TextDetDataPreprocessor', mean=[0, 0, 0], std=[255, 255, 255], bgr_to_rgb=True), backbone=dict( type='mmdet.ResNet', depth=50, num_stages=4, out_indices=(3, ), frozen_stages=-1, norm_cfg=dict(type='BN', requires_grad=False), norm_eval=True, init_cfg=dict(type='Pretrained', checkpoint='torchvision://resnet50'), style='pytorch'), encoder=dict(type='SPTSEncoder', d_backbone=2048, d_model=256), decoder=dict( type='SPTSDecoder', dictionary=dict( type='SPTSDictionary', dict_file= 'mmocr/projects/SPTS/config/spts/../../dicts/spts.txt', with_start=True, with_end=True, with_seq_end=True, same_start_end=False, with_padding=True, with_unknown=True, unknown_token=None), num_bins=1000, d_model=256, dropout=0.1, max_num_text=60, module_loss=dict( type='SPTSModuleLoss', num_bins=1000, ignore_first_char=True), postprocessor=dict(type='SPTSPostprocessor', num_bins=1000))) test_pipeline = [ dict(type='LoadImageFromFile', color_type='color_ignore_orientation'), dict( type='RescaleToShortSide', short_side_lens=[1000], long_side_bound=1824), dict( type='LoadOCRAnnotationsWithBezier', with_bbox=True, with_label=True, with_polygon=True, with_text=True), dict( type='PackTextDetInputs', meta_keys=('img_path', 'ori_shape', 'img_shape', 'scale_factor')) ] train_pipeline = [ dict(type='LoadImageFromFile', color_type='color_ignore_orientation'), dict( type='LoadOCRAnnotationsWithBezier', with_bbox=True, with_label=True, with_polygon=True, with_text=True), dict(type='FixInvalidPolygon'), dict(type='RemoveIgnored'), dict(type='RandomCrop', min_side_ratio=0.5), dict( type='RandomApply', transforms=[ dict( type='RandomRotate', max_angle=30, pad_with_fixed_color=True, use_canvas=True) ], prob=0.3), dict(type='FixInvalidPolygon'), dict( type='RandomChoiceResize', scales=[(640, 1600), (672, 1600), (704, 1600), (736, 1600), (768, 1600), (800, 1600), (832, 1600), (864, 1600), (896, 1600)], keep_ratio=True), dict( type='RandomApply', transforms=[ dict( type='TorchVisionWrapper', op='ColorJitter', brightness=0.5, contrast=0.5, saturation=0.5, hue=0.5) ], prob=0.5), dict( type='PackTextDetInputs', meta_keys=('img_path', 'ori_shape', 'img_shape', 'scale_factor')) ] icdar2015_textspotting_data_root = 'mmocr_data/icdar2015' icdar2015_textspotting_train = dict( type='OCRDataset', data_root='mmocr_data/icdar2015', ann_file='textspotting_train.json', pipeline=[ dict(type='LoadImageFromFile', color_type='color_ignore_orientation'), dict( type='LoadOCRAnnotationsWithBezier', with_bbox=True, with_label=True, with_polygon=True, with_text=True), dict(type='FixInvalidPolygon'), dict(type='RemoveIgnored'), dict(type='RandomCrop', min_side_ratio=0.5), dict( type='RandomApply', transforms=[ dict( type='RandomRotate', max_angle=30, pad_with_fixed_color=True, use_canvas=True) ], prob=0.3), dict(type='FixInvalidPolygon'), dict( type='RandomChoiceResize', scales=[(640, 1600), (672, 1600), (704, 1600), (736, 1600), (768, 1600), (800, 1600), (832, 1600), (864, 1600), (896, 1600)], keep_ratio=True), dict( type='RandomApply', transforms=[ dict( type='TorchVisionWrapper', op='ColorJitter', brightness=0.5, contrast=0.5, saturation=0.5, hue=0.5) ], prob=0.5), dict( type='PackTextDetInputs', meta_keys=('img_path', 'ori_shape', 'img_shape', 'scale_factor')) ]) icdar2015_textspotting_test = dict( type='OCRDataset', data_root='mmocr_data/icdar2015', ann_file='textspotting_test.json', test_mode=True, pipeline=[ dict(type='LoadImageFromFile', color_type='color_ignore_orientation'), dict( type='RescaleToShortSide', short_side_lens=[1000], long_side_bound=1824), dict( type='LoadOCRAnnotationsWithBezier', with_bbox=True, with_label=True, with_polygon=True, with_text=True), dict( type='PackTextDetInputs', meta_keys=('img_path', 'ori_shape', 'img_shape', 'scale_factor')) ]) default_scope = 'mmocr' env_cfg = dict( cudnn_benchmark=False, mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), dist_cfg=dict(backend='nccl')) randomness = dict(seed=42) default_hooks = dict( timer=dict(type='IterTimerHook'), logger=dict(type='LoggerHook', interval=10), param_scheduler=dict(type='ParamSchedulerHook'), checkpoint=dict( type='CheckpointHook', save_best='generic/hmean', rule='greater'), sampler_seed=dict(type='DistSamplerSeedHook'), sync_buffer=dict(type='SyncBuffersHook'), visualization=dict( type='VisualizationHook', interval=1, enable=False, show=False, draw_gt=False, draw_pred=False)) log_level = 'INFO' log_processor = dict(type='LogProcessor', window_size=10, by_epoch=True) load_from = 'work_dirs/spts_resnet50_150e_pretrain-spts-2/epoch_150.pth' resume = False val_evaluator = [ dict( type='E2EPointMetric', prefix='generic', lexicon_path='data/icdar2015/lexicons/GenericVocabulary_new.txt', pair_path='data/icdar2015/lexicons/GenericVocabulary_pair_list.txt', match_dist_thr=None), dict( type='E2EPointMetric', prefix='weak', lexicon_path='data/icdar2015/lexicons/ch4_test_vocabulary_new.txt', pair_path='data/icdar2015/lexicons/ch4_test_vocabulary_pair_list.txt', match_dist_thr=None), dict( type='E2EPointMetric', prefix='strong', lexicon_path='data/icdar2015/lexicons/lexicons/', lexicon_mapping=('(.*).jpg', 'new_voc_\\1.txt'), pair_path='data/icdar2015/lexicons/pairs/', pair_mapping=('(.*).jpg', 'pair_voc_\\1.txt'), match_dist_thr=None) ] test_evaluator = [ dict( type='E2EPointMetric', prefix='generic', lexicon_path='data/icdar2015/lexicons/GenericVocabulary_new.txt', pair_path='data/icdar2015/lexicons/GenericVocabulary_pair_list.txt', match_dist_thr=None), dict( type='E2EPointMetric', prefix='weak', lexicon_path='data/icdar2015/lexicons/ch4_test_vocabulary_new.txt', pair_path='data/icdar2015/lexicons/ch4_test_vocabulary_pair_list.txt', match_dist_thr=None), dict( type='E2EPointMetric', prefix='strong', lexicon_path='data/icdar2015/lexicons/lexicons/', lexicon_mapping=('(.*).jpg', 'new_voc_\\1.txt'), pair_path='data/icdar2015/lexicons/pairs/', pair_mapping=('(.*).jpg', 'pair_voc_\\1.txt'), match_dist_thr=None) ] vis_backends = [dict(type='LocalVisBackend')] visualizer = dict( type='TextSpottingLocalVisualizer', name='visualizer', vis_backends=[dict(type='LocalVisBackend')]) num_epochs = 200 lr = 1e-05 optim_wrapper = dict( type='AmpOptimWrapper', optimizer=dict(type='AdamW', lr=1e-05, weight_decay=0.0001), paramwise_cfg=dict(custom_keys=dict(backbone=dict(lr_mult=0.1))), loss_scale='dynamic') train_cfg = dict(type='EpochBasedTrainLoop', max_epochs=200, val_interval=10) val_cfg = dict(type='ValLoop') test_cfg = dict(type='TestLoop') train_dataloader = dict( batch_size=8, num_workers=8, pin_memory=True, persistent_workers=True, sampler=dict(type='RepeatAugSampler', shuffle=True, num_repeats=2), dataset=dict( type='OCRDataset', data_root='mmocr_data/icdar2015', ann_file='textspotting_train.json', pipeline=[ dict( type='LoadImageFromFile', color_type='color_ignore_orientation'), dict( type='LoadOCRAnnotationsWithBezier', with_bbox=True, with_label=True, with_polygon=True, with_text=True), dict(type='FixInvalidPolygon'), dict(type='RemoveIgnored'), dict(type='RandomCrop', min_side_ratio=0.5), dict( type='RandomApply', transforms=[ dict( type='RandomRotate', max_angle=30, pad_with_fixed_color=True, use_canvas=True) ], prob=0.3), dict(type='FixInvalidPolygon'), dict( type='RandomChoiceResize', scales=[(640, 1600), (672, 1600), (704, 1600), (736, 1600), (768, 1600), (800, 1600), (832, 1600), (864, 1600), (896, 1600)], keep_ratio=True), dict( type='RandomApply', transforms=[ dict( type='TorchVisionWrapper', op='ColorJitter', brightness=0.5, contrast=0.5, saturation=0.5, hue=0.5) ], prob=0.5), dict( type='PackTextDetInputs', meta_keys=('img_path', 'ori_shape', 'img_shape', 'scale_factor')) ])) val_dataloader = dict( batch_size=1, num_workers=4, pin_memory=True, persistent_workers=True, sampler=dict(type='DefaultSampler', shuffle=False), dataset=dict( type='OCRDataset', data_root='mmocr_data/icdar2015', ann_file='textspotting_test.json', test_mode=True, pipeline=[ dict( type='LoadImageFromFile', color_type='color_ignore_orientation'), dict( type='RescaleToShortSide', short_side_lens=[1000], long_side_bound=1824), dict( type='LoadOCRAnnotationsWithBezier', with_bbox=True, with_label=True, with_polygon=True, with_text=True), dict( type='PackTextDetInputs', meta_keys=('img_path', 'ori_shape', 'img_shape', 'scale_factor')) ])) test_dataloader = dict( batch_size=1, num_workers=4, pin_memory=True, persistent_workers=True, sampler=dict(type='DefaultSampler', shuffle=False), dataset=dict( type='OCRDataset', data_root='mmocr_data/icdar2015', ann_file='textspotting_test.json', test_mode=True, pipeline=[ dict( type='LoadImageFromFile', color_type='color_ignore_orientation'), dict( type='RescaleToShortSide', short_side_lens=[1000], long_side_bound=1824), dict( type='LoadOCRAnnotationsWithBezier', with_bbox=True, with_label=True, with_polygon=True, with_text=True), dict( type='PackTextDetInputs', meta_keys=('img_path', 'ori_shape', 'img_shape', 'scale_factor')) ])) launcher = 'pytorch' work_dir = './work_dirs/spts_resnet50_350e_icdar2015' 2023/03/02 23:00:33 - mmengine - WARNING - The "visualizer" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead. 2023/03/02 23:00:33 - mmengine - WARNING - The "vis_backend" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead. 2023/03/02 23:00:34 - mmengine - WARNING - The "model" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead. 2023/03/02 23:00:35 - mmengine - WARNING - The "task util" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead. 2023/03/02 23:00:35 - mmengine - WARNING - The "hook" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead. 2023/03/02 23:00:35 - mmengine - INFO - Hooks will be executed in the following order: before_run: (VERY_HIGH ) RuntimeInfoHook (BELOW_NORMAL) LoggerHook -------------------- before_train: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (VERY_LOW ) CheckpointHook -------------------- before_train_epoch: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (NORMAL ) DistSamplerSeedHook -------------------- before_train_iter: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook -------------------- after_train_iter: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (BELOW_NORMAL) LoggerHook (LOW ) ParamSchedulerHook (VERY_LOW ) CheckpointHook -------------------- after_train_epoch: (NORMAL ) IterTimerHook (NORMAL ) SyncBuffersHook (LOW ) ParamSchedulerHook (VERY_LOW ) CheckpointHook -------------------- before_val_epoch: (NORMAL ) IterTimerHook -------------------- before_val_iter: (NORMAL ) IterTimerHook -------------------- after_val_iter: (NORMAL ) IterTimerHook (NORMAL ) VisualizationHook (BELOW_NORMAL) LoggerHook -------------------- after_val_epoch: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (BELOW_NORMAL) LoggerHook (LOW ) ParamSchedulerHook (VERY_LOW ) CheckpointHook -------------------- before_test_epoch: (NORMAL ) IterTimerHook -------------------- before_test_iter: (NORMAL ) IterTimerHook -------------------- after_test_iter: (NORMAL ) IterTimerHook (NORMAL ) VisualizationHook (BELOW_NORMAL) LoggerHook -------------------- after_test_epoch: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (BELOW_NORMAL) LoggerHook -------------------- after_run: (BELOW_NORMAL) LoggerHook -------------------- 2023/03/02 23:00:36 - mmengine - WARNING - The "loop" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead. 2023/03/02 23:00:36 - mmengine - WARNING - The "dataset" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead. 2023/03/02 23:00:36 - mmengine - WARNING - The "transform" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead. 2023/03/02 23:00:37 - mmengine - WARNING - The "data sampler" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead. 2023/03/02 23:00:37 - mmengine - WARNING - The "optimizer constructor" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead. 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.0.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.0.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.0.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.0.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.0.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.0.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.0.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.0.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.0.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.0.downsample.0.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.0.downsample.0.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.0.downsample.0.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.1.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.1.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.1.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.1.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.1.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.1.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.1.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.1.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.1.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.2.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.2.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.2.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.2.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.2.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.2.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.2.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.2.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer1.2.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.0.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.0.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.0.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.0.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.0.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.0.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.0.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.0.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.0.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.0.downsample.0.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.0.downsample.0.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.0.downsample.0.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.1.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.1.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.1.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.1.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.1.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.1.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.1.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.1.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.1.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.2.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.2.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.2.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.2.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.2.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.2.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.2.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.2.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.2.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.3.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.3.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.3.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.3.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.3.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.3.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.3.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.3.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer2.3.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.0.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.0.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.0.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.0.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.0.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.0.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.0.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.0.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.0.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.0.downsample.0.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.0.downsample.0.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.0.downsample.0.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.1.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.1.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.1.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.1.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.1.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.1.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.1.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.1.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.1.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.2.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.2.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.2.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.2.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.2.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.2.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.2.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.2.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.2.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.3.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.3.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.3.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.3.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.3.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.3.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.3.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.3.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.3.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.4.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.4.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.4.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.4.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.4.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.4.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.4.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.4.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.4.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.5.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.5.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.5.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.5.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.5.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.5.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.5.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.5.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer3.5.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.0.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.0.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.0.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.0.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.0.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.0.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.0.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.0.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.0.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.0.downsample.0.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.0.downsample.0.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.0.downsample.0.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.1.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.1.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.1.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.1.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.1.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.1.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.1.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.1.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.1.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.2.conv1.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.2.conv1.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.2.conv1.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.2.conv2.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.2.conv2.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.2.conv2.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.2.conv3.weight:lr=1.0000000000000002e-06 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.2.conv3.weight:weight_decay=0.0001 2023/03/02 23:00:37 - mmengine - INFO - paramwise_options -- backbone.layer4.2.conv3.weight:lr_mult=0.1 2023/03/02 23:00:37 - mmengine - WARNING - The "optimizer" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead. 2023/03/02 23:00:37 - mmengine - WARNING - The "optim wrapper" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead. 2023/03/02 23:00:37 - mmengine - WARNING - The "metric" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead. 2023/03/02 23:00:38 - mmengine - WARNING - The "weight initializer" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead. 2023/03/02 23:00:38 - mmengine - INFO - load model from: torchvision://resnet50 2023/03/02 23:00:38 - mmengine - INFO - Loads checkpoint by torchvision backend from path: torchvision://resnet50 2023/03/02 23:00:38 - mmengine - WARNING - The model and loaded state dict do not match exactly unexpected key in source state_dict: fc.weight, fc.bias Name of parameter - Initialization information backbone.conv1.weight - torch.Size([64, 3, 7, 7]): PretrainedInit: load from torchvision://resnet50 backbone.bn1.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 backbone.bn1.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.0.conv1.weight - torch.Size([64, 64, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.0.bn1.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.0.bn1.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.0.bn2.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.0.bn2.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.0.conv3.weight - torch.Size([256, 64, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.0.bn3.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.0.bn3.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.0.downsample.0.weight - torch.Size([256, 64, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.0.downsample.1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.0.downsample.1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.1.conv1.weight - torch.Size([64, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.1.bn1.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.1.bn1.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.1.bn2.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.1.bn2.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.1.conv3.weight - torch.Size([256, 64, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.1.bn3.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.1.bn3.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.2.conv1.weight - torch.Size([64, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.2.bn1.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.2.bn1.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.2.conv2.weight - torch.Size([64, 64, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.2.bn2.weight - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.2.bn2.bias - torch.Size([64]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.2.conv3.weight - torch.Size([256, 64, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.2.bn3.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer1.2.bn3.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.0.conv1.weight - torch.Size([128, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.0.bn1.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.0.bn1.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.0.bn2.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.0.bn2.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.0.conv3.weight - torch.Size([512, 128, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.0.bn3.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.0.bn3.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.0.downsample.1.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.0.downsample.1.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.1.conv1.weight - torch.Size([128, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.1.bn1.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.1.bn1.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.1.bn2.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.1.bn2.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.1.conv3.weight - torch.Size([512, 128, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.1.bn3.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.1.bn3.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.2.conv1.weight - torch.Size([128, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.2.bn1.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.2.bn1.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.2.conv2.weight - torch.Size([128, 128, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.2.bn2.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.2.bn2.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.2.conv3.weight - torch.Size([512, 128, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.2.bn3.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.2.bn3.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.3.conv1.weight - torch.Size([128, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.3.bn1.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.3.bn1.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.3.conv2.weight - torch.Size([128, 128, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.3.bn2.weight - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.3.bn2.bias - torch.Size([128]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.3.conv3.weight - torch.Size([512, 128, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.3.bn3.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer2.3.bn3.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.0.conv1.weight - torch.Size([256, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.0.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.0.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.0.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.0.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.0.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.0.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.0.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.0.downsample.0.weight - torch.Size([1024, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.0.downsample.1.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.0.downsample.1.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.1.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.1.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.1.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.1.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.1.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.1.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.1.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.1.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.2.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.2.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.2.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.2.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.2.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.2.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.2.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.2.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.2.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.3.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.3.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.3.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.3.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.3.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.3.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.3.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.3.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.3.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.4.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.4.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.4.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.4.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.4.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.4.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.4.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.4.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.4.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.5.conv1.weight - torch.Size([256, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.5.bn1.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.5.bn1.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.5.conv2.weight - torch.Size([256, 256, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.5.bn2.weight - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.5.bn2.bias - torch.Size([256]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.5.conv3.weight - torch.Size([1024, 256, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.5.bn3.weight - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 backbone.layer3.5.bn3.bias - torch.Size([1024]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.0.conv1.weight - torch.Size([512, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.0.bn1.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.0.bn1.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.0.bn2.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.0.bn2.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.0.conv3.weight - torch.Size([2048, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.0.bn3.weight - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.0.bn3.bias - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.0.downsample.0.weight - torch.Size([2048, 1024, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.0.downsample.1.weight - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.0.downsample.1.bias - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.1.conv1.weight - torch.Size([512, 2048, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.1.bn1.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.1.bn1.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.1.bn2.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.1.bn2.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.1.conv3.weight - torch.Size([2048, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.1.bn3.weight - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.1.bn3.bias - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.2.conv1.weight - torch.Size([512, 2048, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.2.bn1.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.2.bn1.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.2.conv2.weight - torch.Size([512, 512, 3, 3]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.2.bn2.weight - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.2.bn2.bias - torch.Size([512]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.2.conv3.weight - torch.Size([2048, 512, 1, 1]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.2.bn3.weight - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 backbone.layer4.2.bn3.bias - torch.Size([2048]): PretrainedInit: load from torchvision://resnet50 encoder.input_proj.weight - torch.Size([256, 2048, 1, 1]): The value is the same before and after calling `init_weights` of SPTS encoder.input_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.embedding.word_embeddings.weight - torch.Size([1100, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.embedding.position_embeddings.weight - torch.Size([1621, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.embedding.LayerNorm.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.embedding.LayerNorm.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.vocab_embed.layer-0.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.vocab_embed.layer-0.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.vocab_embed.layer-1.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.vocab_embed.layer-1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.vocab_embed.layer-2.weight - torch.Size([1100, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.vocab_embed.layer-2.bias - torch.Size([1100]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.0.self_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.0.self_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.0.self_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.0.self_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.0.linear1.weight - torch.Size([1024, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.0.linear1.bias - torch.Size([1024]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.0.linear2.weight - torch.Size([256, 1024]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.0.linear2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.0.norm1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.0.norm1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.0.norm2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.0.norm2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.1.self_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.1.self_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.1.self_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.1.self_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.1.linear1.weight - torch.Size([1024, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.1.linear1.bias - torch.Size([1024]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.1.linear2.weight - torch.Size([256, 1024]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.1.linear2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.1.norm1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.1.norm1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.1.norm2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.1.norm2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.2.self_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.2.self_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.2.self_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.2.self_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.2.linear1.weight - torch.Size([1024, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.2.linear1.bias - torch.Size([1024]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.2.linear2.weight - torch.Size([256, 1024]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.2.linear2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.2.norm1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.2.norm1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.2.norm2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.2.norm2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.3.self_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.3.self_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.3.self_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.3.self_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.3.linear1.weight - torch.Size([1024, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.3.linear1.bias - torch.Size([1024]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.3.linear2.weight - torch.Size([256, 1024]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.3.linear2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.3.norm1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.3.norm1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.3.norm2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.3.norm2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.4.self_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.4.self_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.4.self_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.4.self_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.4.linear1.weight - torch.Size([1024, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.4.linear1.bias - torch.Size([1024]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.4.linear2.weight - torch.Size([256, 1024]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.4.linear2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.4.norm1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.4.norm1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.4.norm2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.4.norm2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.5.self_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.5.self_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.5.self_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.5.self_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.5.linear1.weight - torch.Size([1024, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.5.linear1.bias - torch.Size([1024]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.5.linear2.weight - torch.Size([256, 1024]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.5.linear2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.5.norm1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.5.norm1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.5.norm2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.layers.5.norm2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.norm.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.encoder.norm.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.self_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.self_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.self_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.self_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.multihead_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.multihead_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.multihead_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.multihead_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.linear1.weight - torch.Size([1024, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.linear1.bias - torch.Size([1024]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.linear2.weight - torch.Size([256, 1024]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.linear2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.norm1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.norm1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.norm2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.norm2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.norm3.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.0.norm3.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.self_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.self_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.self_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.self_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.multihead_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.multihead_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.multihead_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.multihead_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.linear1.weight - torch.Size([1024, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.linear1.bias - torch.Size([1024]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.linear2.weight - torch.Size([256, 1024]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.linear2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.norm1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.norm1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.norm2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.norm2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.norm3.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.1.norm3.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.self_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.self_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.self_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.self_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.multihead_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.multihead_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.multihead_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.multihead_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.linear1.weight - torch.Size([1024, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.linear1.bias - torch.Size([1024]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.linear2.weight - torch.Size([256, 1024]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.linear2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.norm1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.norm1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.norm2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.norm2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.norm3.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.2.norm3.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.self_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.self_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.self_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.self_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.multihead_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.multihead_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.multihead_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.multihead_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.linear1.weight - torch.Size([1024, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.linear1.bias - torch.Size([1024]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.linear2.weight - torch.Size([256, 1024]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.linear2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.norm1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.norm1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.norm2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.norm2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.norm3.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.3.norm3.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.self_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.self_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.self_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.self_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.multihead_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.multihead_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.multihead_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.multihead_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.linear1.weight - torch.Size([1024, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.linear1.bias - torch.Size([1024]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.linear2.weight - torch.Size([256, 1024]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.linear2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.norm1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.norm1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.norm2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.norm2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.norm3.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.4.norm3.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.self_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.self_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.self_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.self_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.multihead_attn.in_proj_weight - torch.Size([768, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.multihead_attn.in_proj_bias - torch.Size([768]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.multihead_attn.out_proj.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.multihead_attn.out_proj.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.linear1.weight - torch.Size([1024, 256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.linear1.bias - torch.Size([1024]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.linear2.weight - torch.Size([256, 1024]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.linear2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.norm1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.norm1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.norm2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.norm2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.norm3.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.layers.5.norm3.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.norm.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS decoder.decoder.norm.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SPTS 2023/03/02 23:00:39 - mmengine - INFO - Load checkpoint from work_dirs/spts_resnet50_150e_pretrain-spts-2/epoch_150.pth 2023/03/02 23:00:39 - mmengine - INFO - Checkpoints will be saved to mmocr/projects/SPTS/work_dirs/spts_resnet50_350e_icdar2015. 2023/03/02 23:00:47 - mmengine - INFO - Epoch(train) [1][10/32] lr: 1.0000e-06 eta: 1:24:15 time: 0.7912 data_time: 0.1610 memory: 19749 loss: 0.1665 loss_ce: 0.1665 2023/03/02 23:00:50 - mmengine - INFO - Epoch(train) [1][20/32] lr: 1.0000e-06 eta: 0:57:42 time: 0.2942 data_time: 0.0014 memory: 19146 loss: 0.1653 loss_ce: 0.1653 2023/03/02 23:00:53 - mmengine - INFO - Epoch(train) [1][30/32] lr: 1.0000e-06 eta: 0:49:02 time: 0.3005 data_time: 0.0011 memory: 19146 loss: 0.1427 loss_ce: 0.1427 2023/03/02 23:00:53 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:00:57 - mmengine - INFO - Epoch(train) [2][10/32] lr: 1.0000e-06 eta: 0:45:31 time: 0.3729 data_time: 0.0586 memory: 18071 loss: 0.1396 loss_ce: 0.1396 2023/03/02 23:01:00 - mmengine - INFO - Epoch(train) [2][20/32] lr: 1.0000e-06 eta: 0:42:07 time: 0.2664 data_time: 0.0013 memory: 18238 loss: 0.1302 loss_ce: 0.1302 2023/03/02 23:01:02 - mmengine - INFO - Epoch(train) [2][30/32] lr: 1.0000e-06 eta: 0:39:55 time: 0.2726 data_time: 0.0011 memory: 17911 loss: 0.1453 loss_ce: 0.1453 2023/03/02 23:01:03 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:01:07 - mmengine - INFO - Epoch(train) [3][10/32] lr: 1.0000e-06 eta: 0:39:22 time: 0.3735 data_time: 0.0341 memory: 19146 loss: 0.1365 loss_ce: 0.1365 2023/03/02 23:01:10 - mmengine - INFO - Epoch(train) [3][20/32] lr: 1.0000e-06 eta: 0:38:22 time: 0.2987 data_time: 0.0013 memory: 19195 loss: 0.1322 loss_ce: 0.1322 2023/03/02 23:01:13 - mmengine - INFO - Epoch(train) [3][30/32] lr: 1.0000e-06 eta: 0:37:34 time: 0.2990 data_time: 0.0011 memory: 19980 loss: 0.1316 loss_ce: 0.1316 2023/03/02 23:01:13 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:01:17 - mmengine - INFO - Epoch(train) [4][10/32] lr: 1.0000e-06 eta: 0:37:43 time: 0.4024 data_time: 0.0602 memory: 18353 loss: 0.1343 loss_ce: 0.1343 2023/03/02 23:01:22 - mmengine - INFO - Epoch(train) [4][20/32] lr: 1.0000e-06 eta: 0:38:26 time: 0.4452 data_time: 0.0014 memory: 17912 loss: 0.1357 loss_ce: 0.1357 2023/03/02 23:01:26 - mmengine - INFO - Epoch(train) [4][30/32] lr: 1.0000e-06 eta: 0:39:19 time: 0.4807 data_time: 0.0013 memory: 25450 loss: 0.1197 loss_ce: 0.1197 2023/03/02 23:01:27 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:01:32 - mmengine - INFO - Epoch(train) [5][10/32] lr: 1.0000e-06 eta: 0:40:22 time: 0.5439 data_time: 0.0543 memory: 17913 loss: 0.1396 loss_ce: 0.1396 2023/03/02 23:01:37 - mmengine - INFO - Epoch(train) [5][20/32] lr: 1.0000e-06 eta: 0:40:56 time: 0.4774 data_time: 0.0012 memory: 18585 loss: 0.1066 loss_ce: 0.1066 2023/03/02 23:01:42 - mmengine - INFO - Epoch(train) [5][30/32] lr: 1.0000e-06 eta: 0:41:15 time: 0.4512 data_time: 0.0011 memory: 19952 loss: 0.1138 loss_ce: 0.1138 2023/03/02 23:01:42 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:01:48 - mmengine - INFO - Epoch(train) [6][10/32] lr: 1.0000e-06 eta: 0:42:11 time: 0.5655 data_time: 0.0709 memory: 19134 loss: 0.1466 loss_ce: 0.1466 2023/03/02 23:01:53 - mmengine - INFO - Epoch(train) [6][20/32] lr: 1.0000e-06 eta: 0:42:21 time: 0.4457 data_time: 0.0013 memory: 17913 loss: 0.1316 loss_ce: 0.1316 2023/03/02 23:01:58 - mmengine - INFO - Epoch(train) [6][30/32] lr: 1.0000e-06 eta: 0:42:45 time: 0.4967 data_time: 0.0010 memory: 20376 loss: 0.1190 loss_ce: 0.1190 2023/03/02 23:01:58 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:02:03 - mmengine - INFO - Epoch(train) [7][10/32] lr: 1.0000e-06 eta: 0:42:55 time: 0.4934 data_time: 0.0624 memory: 17730 loss: 0.1359 loss_ce: 0.1359 2023/03/02 23:02:08 - mmengine - INFO - Epoch(train) [7][20/32] lr: 1.0000e-06 eta: 0:43:21 time: 0.5168 data_time: 0.0011 memory: 24870 loss: 0.1217 loss_ce: 0.1217 2023/03/02 23:02:13 - mmengine - INFO - Epoch(train) [7][30/32] lr: 1.0000e-06 eta: 0:43:28 time: 0.4630 data_time: 0.0012 memory: 17731 loss: 0.1198 loss_ce: 0.1198 2023/03/02 23:02:13 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:02:17 - mmengine - INFO - Epoch(train) [8][10/32] lr: 1.0000e-06 eta: 0:43:01 time: 0.3741 data_time: 0.0456 memory: 18079 loss: 0.1452 loss_ce: 0.1452 2023/03/02 23:02:20 - mmengine - INFO - Epoch(train) [8][20/32] lr: 1.0000e-06 eta: 0:42:38 time: 0.3466 data_time: 0.0011 memory: 19146 loss: 0.1367 loss_ce: 0.1367 2023/03/02 23:02:26 - mmengine - INFO - Epoch(train) [8][30/32] lr: 1.0000e-06 eta: 0:43:16 time: 0.5884 data_time: 0.0012 memory: 19247 loss: 0.1050 loss_ce: 0.1050 2023/03/02 23:02:27 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:02:33 - mmengine - INFO - Epoch(train) [9][10/32] lr: 1.0000e-06 eta: 0:43:58 time: 0.6323 data_time: 0.0663 memory: 18407 loss: 0.1180 loss_ce: 0.1180 2023/03/02 23:02:39 - mmengine - INFO - Epoch(train) [9][20/32] lr: 1.0000e-06 eta: 0:44:18 time: 0.5394 data_time: 0.0013 memory: 17423 loss: 0.1320 loss_ce: 0.1320 2023/03/02 23:02:44 - mmengine - INFO - Epoch(train) [9][30/32] lr: 1.0000e-06 eta: 0:44:38 time: 0.5459 data_time: 0.0011 memory: 19953 loss: 0.1463 loss_ce: 0.1463 2023/03/02 23:02:45 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:02:51 - mmengine - INFO - Epoch(train) [10][10/32] lr: 1.0000e-06 eta: 0:45:06 time: 0.6143 data_time: 0.0389 memory: 19953 loss: 0.1239 loss_ce: 0.1239 2023/03/02 23:02:57 - mmengine - INFO - Epoch(train) [10][20/32] lr: 1.0000e-06 eta: 0:45:25 time: 0.5629 data_time: 0.0011 memory: 20816 loss: 0.1295 loss_ce: 0.1295 2023/03/02 23:03:03 - mmengine - INFO - Epoch(train) [10][30/32] lr: 1.0000e-06 eta: 0:45:47 time: 0.5837 data_time: 0.0012 memory: 23554 loss: 0.1146 loss_ce: 0.1146 2023/03/02 23:03:03 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:03:14 - mmengine - INFO - Epoch(val) [10][10/63] eta: 0:00:56 time: 1.0750 data_time: 0.0263 memory: 17137 2023/03/02 23:03:28 - mmengine - INFO - Epoch(val) [10][20/63] eta: 0:00:53 time: 1.3947 data_time: 0.0004 memory: 1075 2023/03/02 23:03:45 - mmengine - INFO - Epoch(val) [10][30/63] eta: 0:00:45 time: 1.6986 data_time: 0.0004 memory: 1075 2023/03/02 23:03:52 - mmengine - INFO - Epoch(val) [10][40/63] eta: 0:00:27 time: 0.6642 data_time: 0.0003 memory: 1075 2023/03/02 23:04:05 - mmengine - INFO - Epoch(val) [10][50/63] eta: 0:00:15 time: 1.2806 data_time: 0.0003 memory: 1075 2023/03/02 23:04:13 - mmengine - INFO - Epoch(val) [10][60/63] eta: 0:00:03 time: 0.8040 data_time: 0.0003 memory: 1075 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.80, recall: 0.5970, precision: 0.7515, hmean: 0.6654 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.81, recall: 0.5961, precision: 0.7590, hmean: 0.6677 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.82, recall: 0.5936, precision: 0.7644, hmean: 0.6683 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.83, recall: 0.5927, precision: 0.7713, hmean: 0.6703 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.84, recall: 0.5908, precision: 0.7800, hmean: 0.6723 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.85, recall: 0.5883, precision: 0.7894, hmean: 0.6742 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.86, recall: 0.5855, precision: 0.7974, hmean: 0.6752 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.87, recall: 0.5792, precision: 0.8025, hmean: 0.6728 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.88, recall: 0.5739, precision: 0.8076, hmean: 0.6710 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.89, recall: 0.5676, precision: 0.8165, hmean: 0.6697 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.90, recall: 0.5628, precision: 0.8250, hmean: 0.6691 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.91, recall: 0.5551, precision: 0.8343, hmean: 0.6667 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.92, recall: 0.5484, precision: 0.8400, hmean: 0.6636 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.93, recall: 0.5368, precision: 0.8441, hmean: 0.6563 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.94, recall: 0.5248, precision: 0.8536, hmean: 0.6500 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.95, recall: 0.5094, precision: 0.8609, hmean: 0.6400 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.96, recall: 0.4906, precision: 0.8695, hmean: 0.6273 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.97, recall: 0.4661, precision: 0.8752, hmean: 0.6082 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.98, recall: 0.4377, precision: 0.8886, hmean: 0.5865 2023/03/02 23:05:13 - mmengine - INFO - text score threshold: 0.99, recall: 0.3991, precision: 0.9021, hmean: 0.5534 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.80, recall: 0.6432, precision: 0.8097, hmean: 0.7169 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.81, recall: 0.6389, precision: 0.8136, hmean: 0.7157 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.82, recall: 0.6360, precision: 0.8190, hmean: 0.7160 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.83, recall: 0.6341, precision: 0.8252, hmean: 0.7171 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.84, recall: 0.6293, precision: 0.8309, hmean: 0.7162 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.85, recall: 0.6245, precision: 0.8379, hmean: 0.7156 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.86, recall: 0.6196, precision: 0.8439, hmean: 0.7146 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.87, recall: 0.6115, precision: 0.8472, hmean: 0.7103 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.88, recall: 0.6042, precision: 0.8503, hmean: 0.7064 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.89, recall: 0.5975, precision: 0.8594, hmean: 0.7049 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.90, recall: 0.5912, precision: 0.8666, hmean: 0.7029 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.91, recall: 0.5797, precision: 0.8712, hmean: 0.6962 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.92, recall: 0.5720, precision: 0.8761, hmean: 0.6921 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.93, recall: 0.5585, precision: 0.8781, hmean: 0.6828 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.94, recall: 0.5436, precision: 0.8841, hmean: 0.6732 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.95, recall: 0.5262, precision: 0.8893, hmean: 0.6612 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.96, recall: 0.5051, precision: 0.8951, hmean: 0.6457 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.97, recall: 0.4781, precision: 0.8978, hmean: 0.6239 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.98, recall: 0.4463, precision: 0.9062, hmean: 0.5981 2023/03/02 23:05:22 - mmengine - INFO - text score threshold: 0.99, recall: 0.4039, precision: 0.9129, hmean: 0.5601 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.80, recall: 0.7000, precision: 0.8812, hmean: 0.7803 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.81, recall: 0.6938, precision: 0.8835, hmean: 0.7772 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.82, recall: 0.6890, precision: 0.8872, hmean: 0.7756 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.83, recall: 0.6861, precision: 0.8929, hmean: 0.7759 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.84, recall: 0.6798, precision: 0.8976, hmean: 0.7737 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.85, recall: 0.6731, precision: 0.9031, hmean: 0.7713 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.86, recall: 0.6659, precision: 0.9069, hmean: 0.7679 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.87, recall: 0.6562, precision: 0.9093, hmean: 0.7623 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.88, recall: 0.6476, precision: 0.9112, hmean: 0.7571 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.89, recall: 0.6360, precision: 0.9148, hmean: 0.7504 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.90, recall: 0.6264, precision: 0.9181, hmean: 0.7447 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.91, recall: 0.6129, precision: 0.9211, hmean: 0.7361 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.92, recall: 0.6038, precision: 0.9248, hmean: 0.7306 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.93, recall: 0.5879, precision: 0.9243, hmean: 0.7187 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.94, recall: 0.5696, precision: 0.9264, hmean: 0.7054 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.95, recall: 0.5513, precision: 0.9317, hmean: 0.6927 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.96, recall: 0.5286, precision: 0.9369, hmean: 0.6759 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.97, recall: 0.4988, precision: 0.9367, hmean: 0.6510 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.98, recall: 0.4656, precision: 0.9453, hmean: 0.6239 2023/03/02 23:05:31 - mmengine - INFO - text score threshold: 0.99, recall: 0.4213, precision: 0.9521, hmean: 0.5841 2023/03/02 23:05:31 - mmengine - INFO - Epoch(val) [10][63/63] generic/precision: 0.7974 generic/recall: 0.5855 generic/hmean: 0.6752 weak/precision: 0.8252 weak/recall: 0.6341 weak/hmean: 0.7171 strong/precision: 0.8812 strong/recall: 0.7000 strong/hmean: 0.7803 2023/03/02 23:05:33 - mmengine - INFO - The best checkpoint with 0.6752 generic/hmean at 10 epoch is saved to best_generic/hmean_epoch_10.pth. 2023/03/02 23:05:36 - mmengine - INFO - Epoch(train) [11][10/32] lr: 1.0000e-06 eta: 0:45:16 time: 0.3363 data_time: 0.0573 memory: 18413 loss: 0.1370 loss_ce: 0.1370 2023/03/02 23:05:39 - mmengine - INFO - Epoch(train) [11][20/32] lr: 1.0000e-06 eta: 0:44:47 time: 0.3079 data_time: 0.0011 memory: 17583 loss: 0.1178 loss_ce: 0.1178 2023/03/02 23:05:42 - mmengine - INFO - Epoch(train) [11][30/32] lr: 1.0000e-06 eta: 0:44:13 time: 0.2699 data_time: 0.0009 memory: 24319 loss: 0.1161 loss_ce: 0.1161 2023/03/02 23:05:42 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:05:46 - mmengine - INFO - Epoch(train) [12][10/32] lr: 1.0000e-06 eta: 0:43:48 time: 0.3644 data_time: 0.0591 memory: 18966 loss: 0.1230 loss_ce: 0.1230 2023/03/02 23:05:49 - mmengine - INFO - Epoch(train) [12][20/32] lr: 1.0000e-06 eta: 0:43:22 time: 0.3050 data_time: 0.0011 memory: 17583 loss: 0.1155 loss_ce: 0.1155 2023/03/02 23:05:52 - mmengine - INFO - Epoch(train) [12][30/32] lr: 1.0000e-06 eta: 0:42:52 time: 0.2703 data_time: 0.0009 memory: 20172 loss: 0.1277 loss_ce: 0.1277 2023/03/02 23:05:52 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:05:56 - mmengine - INFO - Epoch(train) [13][10/32] lr: 1.0000e-06 eta: 0:42:29 time: 0.3492 data_time: 0.0290 memory: 18779 loss: 0.1258 loss_ce: 0.1258 2023/03/02 23:05:59 - mmengine - INFO - Epoch(train) [13][20/32] lr: 1.0000e-06 eta: 0:42:09 time: 0.3204 data_time: 0.0011 memory: 24545 loss: 0.1213 loss_ce: 0.1213 2023/03/02 23:06:02 - mmengine - INFO - Epoch(train) [13][30/32] lr: 1.0000e-06 eta: 0:41:44 time: 0.2770 data_time: 0.0009 memory: 20999 loss: 0.1228 loss_ce: 0.1228 2023/03/02 23:06:02 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:06:06 - mmengine - INFO - Epoch(train) [14][10/32] lr: 1.0000e-06 eta: 0:41:25 time: 0.3624 data_time: 0.0293 memory: 18246 loss: 0.1299 loss_ce: 0.1299 2023/03/02 23:06:09 - mmengine - INFO - Epoch(train) [14][20/32] lr: 1.0000e-06 eta: 0:41:05 time: 0.2972 data_time: 0.0011 memory: 18083 loss: 0.1245 loss_ce: 0.1245 2023/03/02 23:06:12 - mmengine - INFO - Epoch(train) [14][30/32] lr: 1.0000e-06 eta: 0:40:48 time: 0.3162 data_time: 0.0009 memory: 23004 loss: 0.1134 loss_ce: 0.1134 2023/03/02 23:06:12 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:06:16 - mmengine - INFO - Epoch(train) [15][10/32] lr: 1.0000e-06 eta: 0:40:35 time: 0.3937 data_time: 0.0523 memory: 19958 loss: 0.1231 loss_ce: 0.1231 2023/03/02 23:06:19 - mmengine - INFO - Epoch(train) [15][20/32] lr: 1.0000e-06 eta: 0:40:13 time: 0.2692 data_time: 0.0011 memory: 21046 loss: 0.1250 loss_ce: 0.1250 2023/03/02 23:06:22 - mmengine - INFO - Epoch(train) [15][30/32] lr: 1.0000e-06 eta: 0:39:56 time: 0.2986 data_time: 0.0010 memory: 20382 loss: 0.1236 loss_ce: 0.1236 2023/03/02 23:06:22 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:06:26 - mmengine - INFO - Epoch(train) [16][10/32] lr: 1.0000e-06 eta: 0:39:43 time: 0.3745 data_time: 0.0548 memory: 18083 loss: 0.1133 loss_ce: 0.1133 2023/03/02 23:06:29 - mmengine - INFO - Epoch(train) [16][20/32] lr: 1.0000e-06 eta: 0:39:26 time: 0.2943 data_time: 0.0011 memory: 21742 loss: 0.1222 loss_ce: 0.1222 2023/03/02 23:06:32 - mmengine - INFO - Epoch(train) [16][30/32] lr: 1.0000e-06 eta: 0:39:09 time: 0.2920 data_time: 0.0010 memory: 19558 loss: 0.1206 loss_ce: 0.1206 2023/03/02 23:06:33 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:06:36 - mmengine - INFO - Epoch(train) [17][10/32] lr: 1.0000e-06 eta: 0:38:58 time: 0.3756 data_time: 0.0414 memory: 19958 loss: 0.1076 loss_ce: 0.1076 2023/03/02 23:06:39 - mmengine - INFO - Epoch(train) [17][20/32] lr: 1.0000e-06 eta: 0:38:41 time: 0.2763 data_time: 0.0012 memory: 20070 loss: 0.1166 loss_ce: 0.1166 2023/03/02 23:06:42 - mmengine - INFO - Epoch(train) [17][30/32] lr: 1.0000e-06 eta: 0:38:26 time: 0.2972 data_time: 0.0010 memory: 19552 loss: 0.1160 loss_ce: 0.1160 2023/03/02 23:06:42 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:06:46 - mmengine - INFO - Epoch(train) [18][10/32] lr: 1.0000e-06 eta: 0:38:16 time: 0.3756 data_time: 0.0647 memory: 19346 loss: 0.1274 loss_ce: 0.1274 2023/03/02 23:06:49 - mmengine - INFO - Epoch(train) [18][20/32] lr: 1.0000e-06 eta: 0:38:00 time: 0.2792 data_time: 0.0012 memory: 19958 loss: 0.1181 loss_ce: 0.1181 2023/03/02 23:06:52 - mmengine - INFO - Epoch(train) [18][30/32] lr: 1.0000e-06 eta: 0:37:45 time: 0.2755 data_time: 0.0010 memory: 20297 loss: 0.1245 loss_ce: 0.1245 2023/03/02 23:06:52 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:06:56 - mmengine - INFO - Epoch(train) [19][10/32] lr: 1.0000e-06 eta: 0:37:35 time: 0.3660 data_time: 0.0788 memory: 17583 loss: 0.1216 loss_ce: 0.1216 2023/03/02 23:06:59 - mmengine - INFO - Epoch(train) [19][20/32] lr: 1.0000e-06 eta: 0:37:23 time: 0.3031 data_time: 0.0012 memory: 18962 loss: 0.1331 loss_ce: 0.1331 2023/03/02 23:07:02 - mmengine - INFO - Epoch(train) [19][30/32] lr: 1.0000e-06 eta: 0:37:10 time: 0.2896 data_time: 0.0010 memory: 21092 loss: 0.1311 loss_ce: 0.1311 2023/03/02 23:07:02 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:07:06 - mmengine - INFO - Epoch(train) [20][10/32] lr: 1.0000e-06 eta: 0:37:02 time: 0.3832 data_time: 0.0569 memory: 17919 loss: 0.1283 loss_ce: 0.1283 2023/03/02 23:07:09 - mmengine - INFO - Epoch(train) [20][20/32] lr: 1.0000e-06 eta: 0:36:51 time: 0.3122 data_time: 0.0012 memory: 18083 loss: 0.1108 loss_ce: 0.1108 2023/03/02 23:07:13 - mmengine - INFO - Epoch(train) [20][30/32] lr: 1.0000e-06 eta: 0:36:49 time: 0.3993 data_time: 0.0011 memory: 18940 loss: 0.1282 loss_ce: 0.1282 2023/03/02 23:07:14 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:07:30 - mmengine - INFO - Epoch(val) [20][10/63] eta: 0:01:23 time: 1.5784 data_time: 0.0037 memory: 16516 2023/03/02 23:07:56 - mmengine - INFO - Epoch(val) [20][20/63] eta: 0:01:31 time: 2.6673 data_time: 0.0005 memory: 1075 2023/03/02 23:08:25 - mmengine - INFO - Epoch(val) [20][30/63] eta: 0:01:18 time: 2.8866 data_time: 0.0004 memory: 1075 2023/03/02 23:08:37 - mmengine - INFO - Epoch(val) [20][40/63] eta: 0:00:47 time: 1.1617 data_time: 0.0004 memory: 1075 2023/03/02 23:08:59 - mmengine - INFO - Epoch(val) [20][50/63] eta: 0:00:27 time: 2.1917 data_time: 0.0004 memory: 1075 2023/03/02 23:09:13 - mmengine - INFO - Epoch(val) [20][60/63] eta: 0:00:05 time: 1.4662 data_time: 0.0004 memory: 1075 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.80, recall: 0.6033, precision: 0.7548, hmean: 0.6706 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.81, recall: 0.6018, precision: 0.7613, hmean: 0.6722 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.82, recall: 0.6004, precision: 0.7674, hmean: 0.6737 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.83, recall: 0.5980, precision: 0.7762, hmean: 0.6756 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.84, recall: 0.5946, precision: 0.7846, hmean: 0.6765 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.85, recall: 0.5922, precision: 0.7910, hmean: 0.6773 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.86, recall: 0.5888, precision: 0.7973, hmean: 0.6774 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.87, recall: 0.5845, precision: 0.8050, hmean: 0.6773 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.88, recall: 0.5773, precision: 0.8134, hmean: 0.6753 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.89, recall: 0.5720, precision: 0.8233, hmean: 0.6750 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.90, recall: 0.5667, precision: 0.8312, hmean: 0.6739 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.91, recall: 0.5590, precision: 0.8365, hmean: 0.6701 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.92, recall: 0.5465, precision: 0.8451, hmean: 0.6637 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.93, recall: 0.5378, precision: 0.8527, hmean: 0.6596 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.94, recall: 0.5282, precision: 0.8651, hmean: 0.6559 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.95, recall: 0.5118, precision: 0.8649, hmean: 0.6431 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.96, recall: 0.4964, precision: 0.8737, hmean: 0.6331 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.97, recall: 0.4738, precision: 0.8833, hmean: 0.6167 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.98, recall: 0.4429, precision: 0.8949, hmean: 0.5926 2023/03/02 23:10:13 - mmengine - INFO - text score threshold: 0.99, recall: 0.3991, precision: 0.9060, hmean: 0.5541 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.80, recall: 0.6461, precision: 0.8084, hmean: 0.7182 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.81, recall: 0.6437, precision: 0.8143, hmean: 0.7190 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.82, recall: 0.6418, precision: 0.8203, hmean: 0.7202 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.83, recall: 0.6370, precision: 0.8269, hmean: 0.7196 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.84, recall: 0.6317, precision: 0.8335, hmean: 0.7187 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.85, recall: 0.6273, precision: 0.8379, hmean: 0.7175 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.86, recall: 0.6230, precision: 0.8435, hmean: 0.7167 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.87, recall: 0.6182, precision: 0.8515, hmean: 0.7163 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.88, recall: 0.6086, precision: 0.8575, hmean: 0.7119 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.89, recall: 0.6013, precision: 0.8656, hmean: 0.7097 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.90, recall: 0.5941, precision: 0.8715, hmean: 0.7066 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.91, recall: 0.5840, precision: 0.8739, hmean: 0.7001 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.92, recall: 0.5696, precision: 0.8809, hmean: 0.6918 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.93, recall: 0.5590, precision: 0.8863, hmean: 0.6856 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.94, recall: 0.5460, precision: 0.8943, hmean: 0.6780 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.95, recall: 0.5291, precision: 0.8942, hmean: 0.6649 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.96, recall: 0.5113, precision: 0.9000, hmean: 0.6521 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.97, recall: 0.4839, precision: 0.9022, hmean: 0.6299 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.98, recall: 0.4511, precision: 0.9115, hmean: 0.6035 2023/03/02 23:10:22 - mmengine - INFO - text score threshold: 0.99, recall: 0.4039, precision: 0.9169, hmean: 0.5608 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.80, recall: 0.6996, precision: 0.8753, hmean: 0.7776 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.81, recall: 0.6957, precision: 0.8800, hmean: 0.7771 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.82, recall: 0.6919, precision: 0.8843, hmean: 0.7763 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.83, recall: 0.6856, precision: 0.8900, hmean: 0.7745 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.84, recall: 0.6793, precision: 0.8964, hmean: 0.7729 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.85, recall: 0.6740, precision: 0.9003, hmean: 0.7709 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.86, recall: 0.6683, precision: 0.9048, hmean: 0.7688 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.87, recall: 0.6610, precision: 0.9105, hmean: 0.7660 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.88, recall: 0.6495, precision: 0.9152, hmean: 0.7598 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.89, recall: 0.6375, precision: 0.9175, hmean: 0.7523 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.90, recall: 0.6283, precision: 0.9216, hmean: 0.7472 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.91, recall: 0.6177, precision: 0.9244, hmean: 0.7405 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.92, recall: 0.5999, precision: 0.9278, hmean: 0.7287 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.93, recall: 0.5864, precision: 0.9298, hmean: 0.7192 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.94, recall: 0.5720, precision: 0.9369, hmean: 0.7103 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.95, recall: 0.5546, precision: 0.9373, hmean: 0.6969 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.96, recall: 0.5344, precision: 0.9407, hmean: 0.6816 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.97, recall: 0.5065, precision: 0.9443, hmean: 0.6594 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.98, recall: 0.4709, precision: 0.9514, hmean: 0.6300 2023/03/02 23:10:31 - mmengine - INFO - text score threshold: 0.99, recall: 0.4213, precision: 0.9563, hmean: 0.5849 2023/03/02 23:10:31 - mmengine - INFO - Epoch(val) [20][63/63] generic/precision: 0.7973 generic/recall: 0.5888 generic/hmean: 0.6774 weak/precision: 0.8203 weak/recall: 0.6418 weak/hmean: 0.7202 strong/precision: 0.8753 strong/recall: 0.6996 strong/hmean: 0.7776 2023/03/02 23:10:31 - mmengine - INFO - The previous best checkpoint mmocr/projects/SPTS/work_dirs/spts_resnet50_350e_icdar2015/best_generic/hmean_epoch_10.pth is removed 2023/03/02 23:10:34 - mmengine - INFO - The best checkpoint with 0.6774 generic/hmean at 20 epoch is saved to best_generic/hmean_epoch_20.pth. 2023/03/02 23:10:39 - mmengine - INFO - Epoch(train) [21][10/32] lr: 1.0000e-06 eta: 0:36:57 time: 0.5498 data_time: 0.0505 memory: 26130 loss: 0.1156 loss_ce: 0.1156 2023/03/02 23:10:44 - mmengine - INFO - Epoch(train) [21][20/32] lr: 1.0000e-06 eta: 0:37:00 time: 0.4605 data_time: 0.0013 memory: 22766 loss: 0.1208 loss_ce: 0.1208 2023/03/02 23:10:48 - mmengine - INFO - Epoch(train) [21][30/32] lr: 1.0000e-06 eta: 0:36:58 time: 0.4092 data_time: 0.0012 memory: 18246 loss: 0.1239 loss_ce: 0.1239 2023/03/02 23:10:48 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:10:54 - mmengine - INFO - Epoch(train) [22][10/32] lr: 1.0000e-06 eta: 0:37:03 time: 0.5325 data_time: 0.0999 memory: 17919 loss: 0.1330 loss_ce: 0.1330 2023/03/02 23:10:58 - mmengine - INFO - Epoch(train) [22][20/32] lr: 1.0000e-06 eta: 0:37:05 time: 0.4614 data_time: 0.0012 memory: 21656 loss: 0.1158 loss_ce: 0.1158 2023/03/02 23:11:03 - mmengine - INFO - Epoch(train) [22][30/32] lr: 1.0000e-06 eta: 0:37:07 time: 0.4572 data_time: 0.0012 memory: 18597 loss: 0.1251 loss_ce: 0.1251 2023/03/02 23:11:03 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:11:09 - mmengine - INFO - Epoch(train) [23][10/32] lr: 1.0000e-06 eta: 0:37:17 time: 0.5864 data_time: 0.0757 memory: 24544 loss: 0.0996 loss_ce: 0.0996 2023/03/02 23:11:14 - mmengine - INFO - Epoch(train) [23][20/32] lr: 1.0000e-06 eta: 0:37:17 time: 0.4331 data_time: 0.0012 memory: 20820 loss: 0.1016 loss_ce: 0.1016 2023/03/02 23:11:18 - mmengine - INFO - Epoch(train) [23][30/32] lr: 1.0000e-06 eta: 0:37:18 time: 0.4579 data_time: 0.0011 memory: 19726 loss: 0.1074 loss_ce: 0.1074 2023/03/02 23:11:19 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:11:24 - mmengine - INFO - Epoch(train) [24][10/32] lr: 1.0000e-06 eta: 0:37:20 time: 0.4928 data_time: 0.0340 memory: 20172 loss: 0.1098 loss_ce: 0.1098 2023/03/02 23:11:28 - mmengine - INFO - Epoch(train) [24][20/32] lr: 1.0000e-06 eta: 0:37:20 time: 0.4432 data_time: 0.0011 memory: 19552 loss: 0.1090 loss_ce: 0.1090 2023/03/02 23:11:33 - mmengine - INFO - Epoch(train) [24][30/32] lr: 1.0000e-06 eta: 0:37:18 time: 0.4229 data_time: 0.0012 memory: 18413 loss: 0.1091 loss_ce: 0.1091 2023/03/02 23:11:33 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:11:39 - mmengine - INFO - Epoch(train) [25][10/32] lr: 1.0000e-06 eta: 0:37:22 time: 0.5422 data_time: 0.0610 memory: 20853 loss: 0.1397 loss_ce: 0.1397 2023/03/02 23:11:43 - mmengine - INFO - Epoch(train) [25][20/32] lr: 1.0000e-06 eta: 0:37:21 time: 0.4395 data_time: 0.0013 memory: 18779 loss: 0.1047 loss_ce: 0.1047 2023/03/02 23:11:48 - mmengine - INFO - Epoch(train) [25][30/32] lr: 1.0000e-06 eta: 0:37:21 time: 0.4529 data_time: 0.0010 memory: 19346 loss: 0.1184 loss_ce: 0.1184 2023/03/02 23:11:48 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:11:53 - mmengine - INFO - Epoch(train) [26][10/32] lr: 1.0000e-06 eta: 0:37:23 time: 0.5326 data_time: 0.0638 memory: 20172 loss: 0.1178 loss_ce: 0.1178 2023/03/02 23:11:58 - mmengine - INFO - Epoch(train) [26][20/32] lr: 1.0000e-06 eta: 0:37:21 time: 0.4320 data_time: 0.0012 memory: 18597 loss: 0.1133 loss_ce: 0.1133 2023/03/02 23:12:02 - mmengine - INFO - Epoch(train) [26][30/32] lr: 1.0000e-06 eta: 0:37:21 time: 0.4547 data_time: 0.0012 memory: 24544 loss: 0.1003 loss_ce: 0.1003 2023/03/02 23:12:03 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:12:09 - mmengine - INFO - Epoch(train) [27][10/32] lr: 1.0000e-06 eta: 0:37:25 time: 0.5385 data_time: 0.0588 memory: 21363 loss: 0.1105 loss_ce: 0.1105 2023/03/02 23:12:13 - mmengine - INFO - Epoch(train) [27][20/32] lr: 1.0000e-06 eta: 0:37:25 time: 0.4686 data_time: 0.0013 memory: 18246 loss: 0.1227 loss_ce: 0.1227 2023/03/02 23:12:17 - mmengine - INFO - Epoch(train) [27][30/32] lr: 1.0000e-06 eta: 0:37:22 time: 0.4204 data_time: 0.0011 memory: 18246 loss: 0.1517 loss_ce: 0.1517 2023/03/02 23:12:18 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:12:24 - mmengine - INFO - Epoch(train) [28][10/32] lr: 1.0000e-06 eta: 0:37:26 time: 0.5719 data_time: 0.0885 memory: 20540 loss: 0.1114 loss_ce: 0.1114 2023/03/02 23:12:28 - mmengine - INFO - Epoch(train) [28][20/32] lr: 1.0000e-06 eta: 0:37:25 time: 0.4552 data_time: 0.0014 memory: 17919 loss: 0.1205 loss_ce: 0.1205 2023/03/02 23:12:33 - mmengine - INFO - Epoch(train) [28][30/32] lr: 1.0000e-06 eta: 0:37:23 time: 0.4453 data_time: 0.0011 memory: 18155 loss: 0.1347 loss_ce: 0.1347 2023/03/02 23:12:33 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:12:39 - mmengine - INFO - Epoch(train) [29][10/32] lr: 1.0000e-06 eta: 0:37:25 time: 0.5216 data_time: 0.0627 memory: 18940 loss: 0.1263 loss_ce: 0.1263 2023/03/02 23:12:43 - mmengine - INFO - Epoch(train) [29][20/32] lr: 1.0000e-06 eta: 0:37:21 time: 0.4035 data_time: 0.0013 memory: 17919 loss: 0.1242 loss_ce: 0.1242 2023/03/02 23:12:47 - mmengine - INFO - Epoch(train) [29][30/32] lr: 1.0000e-06 eta: 0:37:18 time: 0.4417 data_time: 0.0012 memory: 20820 loss: 0.1001 loss_ce: 0.1001 2023/03/02 23:12:48 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:12:53 - mmengine - INFO - Epoch(train) [30][10/32] lr: 1.0000e-06 eta: 0:37:19 time: 0.5353 data_time: 0.0923 memory: 17936 loss: 0.1373 loss_ce: 0.1373 2023/03/02 23:12:57 - mmengine - INFO - Epoch(train) [30][20/32] lr: 1.0000e-06 eta: 0:37:17 time: 0.4437 data_time: 0.0013 memory: 18966 loss: 0.1230 loss_ce: 0.1230 2023/03/02 23:13:02 - mmengine - INFO - Epoch(train) [30][30/32] lr: 1.0000e-06 eta: 0:37:13 time: 0.4086 data_time: 0.0013 memory: 19958 loss: 0.1136 loss_ce: 0.1136 2023/03/02 23:13:02 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:13:17 - mmengine - INFO - Epoch(val) [30][10/63] eta: 0:01:16 time: 1.4396 data_time: 0.0030 memory: 17743 2023/03/02 23:13:42 - mmengine - INFO - Epoch(val) [30][20/63] eta: 0:01:25 time: 2.5467 data_time: 0.0004 memory: 1075 2023/03/02 23:14:06 - mmengine - INFO - Epoch(val) [30][30/63] eta: 0:01:09 time: 2.3323 data_time: 0.0004 memory: 1075 2023/03/02 23:14:17 - mmengine - INFO - Epoch(val) [30][40/63] eta: 0:00:43 time: 1.1854 data_time: 0.0004 memory: 1075 2023/03/02 23:14:40 - mmengine - INFO - Epoch(val) [30][50/63] eta: 0:00:25 time: 2.2843 data_time: 0.0003 memory: 1075 2023/03/02 23:14:54 - mmengine - INFO - Epoch(val) [30][60/63] eta: 0:00:05 time: 1.3639 data_time: 0.0004 memory: 1075 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.80, recall: 0.6042, precision: 0.7588, hmean: 0.6727 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.81, recall: 0.6033, precision: 0.7650, hmean: 0.6746 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.82, recall: 0.6013, precision: 0.7677, hmean: 0.6744 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.83, recall: 0.5989, precision: 0.7732, hmean: 0.6750 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.84, recall: 0.5980, precision: 0.7811, hmean: 0.6774 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.85, recall: 0.5956, precision: 0.7889, hmean: 0.6787 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.86, recall: 0.5936, precision: 0.7965, hmean: 0.6803 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.87, recall: 0.5903, precision: 0.8034, hmean: 0.6805 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.88, recall: 0.5840, precision: 0.8146, hmean: 0.6803 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.89, recall: 0.5773, precision: 0.8190, hmean: 0.6772 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.90, recall: 0.5705, precision: 0.8287, hmean: 0.6758 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.91, recall: 0.5604, precision: 0.8398, hmean: 0.6722 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.92, recall: 0.5489, precision: 0.8470, hmean: 0.6661 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.93, recall: 0.5426, precision: 0.8538, hmean: 0.6635 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.94, recall: 0.5272, precision: 0.8575, hmean: 0.6530 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.95, recall: 0.5152, precision: 0.8671, hmean: 0.6463 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.96, recall: 0.5002, precision: 0.8738, hmean: 0.6363 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.97, recall: 0.4738, precision: 0.8801, hmean: 0.6160 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.98, recall: 0.4482, precision: 0.8969, hmean: 0.5978 2023/03/02 23:15:53 - mmengine - INFO - text score threshold: 0.99, recall: 0.4059, precision: 0.9084, hmean: 0.5611 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.80, recall: 0.6476, precision: 0.8132, hmean: 0.7210 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.81, recall: 0.6461, precision: 0.8193, hmean: 0.7225 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.82, recall: 0.6442, precision: 0.8224, hmean: 0.7225 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.83, recall: 0.6403, precision: 0.8266, hmean: 0.7216 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.84, recall: 0.6375, precision: 0.8327, hmean: 0.7221 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.85, recall: 0.6317, precision: 0.8367, hmean: 0.7199 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.86, recall: 0.6283, precision: 0.8430, hmean: 0.7200 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.87, recall: 0.6245, precision: 0.8499, hmean: 0.7200 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.88, recall: 0.6158, precision: 0.8590, hmean: 0.7173 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.89, recall: 0.6086, precision: 0.8634, hmean: 0.7139 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.90, recall: 0.5985, precision: 0.8692, hmean: 0.7089 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.91, recall: 0.5835, precision: 0.8745, hmean: 0.7000 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.92, recall: 0.5715, precision: 0.8819, hmean: 0.6935 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.93, recall: 0.5638, precision: 0.8871, hmean: 0.6894 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.94, recall: 0.5465, precision: 0.8888, hmean: 0.6768 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.95, recall: 0.5330, precision: 0.8971, hmean: 0.6687 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.96, recall: 0.5152, precision: 0.8999, hmean: 0.6552 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.97, recall: 0.4868, precision: 0.9043, hmean: 0.6329 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.98, recall: 0.4569, precision: 0.9143, hmean: 0.6093 2023/03/02 23:16:03 - mmengine - INFO - text score threshold: 0.99, recall: 0.4121, precision: 0.9224, hmean: 0.5697 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.80, recall: 0.7000, precision: 0.8791, hmean: 0.7794 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.81, recall: 0.6976, precision: 0.8846, hmean: 0.7801 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.82, recall: 0.6952, precision: 0.8875, hmean: 0.7797 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.83, recall: 0.6899, precision: 0.8906, hmean: 0.7775 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.84, recall: 0.6866, precision: 0.8969, hmean: 0.7777 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.85, recall: 0.6798, precision: 0.9005, hmean: 0.7748 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.86, recall: 0.6745, precision: 0.9050, hmean: 0.7730 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.87, recall: 0.6688, precision: 0.9102, hmean: 0.7710 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.88, recall: 0.6562, precision: 0.9154, hmean: 0.7644 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.89, recall: 0.6466, precision: 0.9173, hmean: 0.7585 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.90, recall: 0.6326, precision: 0.9189, hmean: 0.7494 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.91, recall: 0.6163, precision: 0.9235, hmean: 0.7392 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.92, recall: 0.5999, precision: 0.9257, hmean: 0.7280 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.93, recall: 0.5888, precision: 0.9265, hmean: 0.7200 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.94, recall: 0.5715, precision: 0.9295, hmean: 0.7078 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.95, recall: 0.5566, precision: 0.9368, hmean: 0.6983 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.96, recall: 0.5373, precision: 0.9386, hmean: 0.6834 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.97, recall: 0.5070, precision: 0.9419, hmean: 0.6592 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.98, recall: 0.4747, precision: 0.9499, hmean: 0.6331 2023/03/02 23:16:11 - mmengine - INFO - text score threshold: 0.99, recall: 0.4266, precision: 0.9547, hmean: 0.5897 2023/03/02 23:16:11 - mmengine - INFO - Epoch(val) [30][63/63] generic/precision: 0.8034 generic/recall: 0.5903 generic/hmean: 0.6805 weak/precision: 0.8193 weak/recall: 0.6461 weak/hmean: 0.7225 strong/precision: 0.8846 strong/recall: 0.6976 strong/hmean: 0.7801 2023/03/02 23:16:11 - mmengine - INFO - The previous best checkpoint mmocr/projects/SPTS/work_dirs/spts_resnet50_350e_icdar2015/best_generic/hmean_epoch_20.pth is removed 2023/03/02 23:16:14 - mmengine - INFO - The best checkpoint with 0.6805 generic/hmean at 30 epoch is saved to best_generic/hmean_epoch_30.pth. 2023/03/02 23:16:19 - mmengine - INFO - Epoch(train) [31][10/32] lr: 1.0000e-06 eta: 0:37:13 time: 0.5060 data_time: 0.0505 memory: 24394 loss: 0.1149 loss_ce: 0.1149 2023/03/02 23:16:23 - mmengine - INFO - Epoch(train) [31][20/32] lr: 1.0000e-06 eta: 0:37:11 time: 0.4353 data_time: 0.0012 memory: 19346 loss: 0.1113 loss_ce: 0.1113 2023/03/02 23:16:27 - mmengine - INFO - Epoch(train) [31][30/32] lr: 1.0000e-06 eta: 0:37:08 time: 0.4389 data_time: 0.0011 memory: 17919 loss: 0.1286 loss_ce: 0.1286 2023/03/02 23:16:28 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:16:33 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:16:34 - mmengine - INFO - Epoch(train) [32][10/32] lr: 1.0000e-06 eta: 0:37:09 time: 0.5586 data_time: 0.0955 memory: 24319 loss: 0.1096 loss_ce: 0.1096 2023/03/02 23:16:38 - mmengine - INFO - Epoch(train) [32][20/32] lr: 1.0000e-06 eta: 0:37:09 time: 0.4765 data_time: 0.0012 memory: 19751 loss: 0.1171 loss_ce: 0.1171 2023/03/02 23:16:43 - mmengine - INFO - Epoch(train) [32][30/32] lr: 1.0000e-06 eta: 0:37:06 time: 0.4432 data_time: 0.0011 memory: 19958 loss: 0.1245 loss_ce: 0.1245 2023/03/02 23:16:44 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:16:49 - mmengine - INFO - Epoch(train) [33][10/32] lr: 1.0000e-06 eta: 0:37:05 time: 0.4935 data_time: 0.0502 memory: 18246 loss: 0.1201 loss_ce: 0.1201 2023/03/02 23:16:53 - mmengine - INFO - Epoch(train) [33][20/32] lr: 1.0000e-06 eta: 0:37:03 time: 0.4509 data_time: 0.0012 memory: 18487 loss: 0.1132 loss_ce: 0.1132 2023/03/02 23:16:58 - mmengine - INFO - Epoch(train) [33][30/32] lr: 1.0000e-06 eta: 0:37:02 time: 0.4846 data_time: 0.0010 memory: 18966 loss: 0.1220 loss_ce: 0.1220 2023/03/02 23:16:58 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:17:04 - mmengine - INFO - Epoch(train) [34][10/32] lr: 1.0000e-06 eta: 0:37:05 time: 0.5958 data_time: 0.0532 memory: 27104 loss: 0.1152 loss_ce: 0.1152 2023/03/02 23:17:09 - mmengine - INFO - Epoch(train) [34][20/32] lr: 1.0000e-06 eta: 0:37:02 time: 0.4533 data_time: 0.0014 memory: 19958 loss: 0.1073 loss_ce: 0.1073 2023/03/02 23:17:13 - mmengine - INFO - Epoch(train) [34][30/32] lr: 1.0000e-06 eta: 0:36:59 time: 0.4403 data_time: 0.0013 memory: 21046 loss: 0.1172 loss_ce: 0.1172 2023/03/02 23:17:14 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:17:20 - mmengine - INFO - Epoch(train) [35][10/32] lr: 1.0000e-06 eta: 0:37:00 time: 0.5209 data_time: 0.0739 memory: 19844 loss: 0.1198 loss_ce: 0.1198 2023/03/02 23:17:24 - mmengine - INFO - Epoch(train) [35][20/32] lr: 1.0000e-06 eta: 0:36:57 time: 0.4360 data_time: 0.0012 memory: 20820 loss: 0.1050 loss_ce: 0.1050 2023/03/02 23:17:28 - mmengine - INFO - Epoch(train) [35][30/32] lr: 1.0000e-06 eta: 0:36:53 time: 0.4285 data_time: 0.0014 memory: 18325 loss: 0.1199 loss_ce: 0.1199 2023/03/02 23:17:29 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:17:34 - mmengine - INFO - Epoch(train) [36][10/32] lr: 1.0000e-06 eta: 0:36:52 time: 0.5260 data_time: 0.0466 memory: 18779 loss: 0.1193 loss_ce: 0.1193 2023/03/02 23:17:39 - mmengine - INFO - Epoch(train) [36][20/32] lr: 1.0000e-06 eta: 0:36:50 time: 0.4762 data_time: 0.0012 memory: 18779 loss: 0.1034 loss_ce: 0.1034 2023/03/02 23:17:43 - mmengine - INFO - Epoch(train) [36][30/32] lr: 1.0000e-06 eta: 0:36:48 time: 0.4551 data_time: 0.0012 memory: 20238 loss: 0.1225 loss_ce: 0.1225 2023/03/02 23:17:44 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:17:49 - mmengine - INFO - Epoch(train) [37][10/32] lr: 1.0000e-06 eta: 0:36:47 time: 0.5323 data_time: 0.0705 memory: 17777 loss: 0.1092 loss_ce: 0.1092 2023/03/02 23:17:54 - mmengine - INFO - Epoch(train) [37][20/32] lr: 1.0000e-06 eta: 0:36:44 time: 0.4530 data_time: 0.0012 memory: 18083 loss: 0.1165 loss_ce: 0.1165 2023/03/02 23:17:59 - mmengine - INFO - Epoch(train) [37][30/32] lr: 1.0000e-06 eta: 0:36:42 time: 0.4733 data_time: 0.0012 memory: 19344 loss: 0.1063 loss_ce: 0.1063 2023/03/02 23:17:59 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:18:04 - mmengine - INFO - Epoch(train) [38][10/32] lr: 1.0000e-06 eta: 0:36:40 time: 0.5201 data_time: 0.0614 memory: 19958 loss: 0.1123 loss_ce: 0.1123 2023/03/02 23:18:09 - mmengine - INFO - Epoch(train) [38][20/32] lr: 1.0000e-06 eta: 0:36:39 time: 0.4855 data_time: 0.0012 memory: 22545 loss: 0.1270 loss_ce: 0.1270 2023/03/02 23:18:14 - mmengine - INFO - Epoch(train) [38][30/32] lr: 1.0000e-06 eta: 0:36:35 time: 0.4357 data_time: 0.0012 memory: 21046 loss: 0.1184 loss_ce: 0.1184 2023/03/02 23:18:14 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:18:20 - mmengine - INFO - Epoch(train) [39][10/32] lr: 1.0000e-06 eta: 0:36:35 time: 0.5428 data_time: 0.0569 memory: 18799 loss: 0.1037 loss_ce: 0.1037 2023/03/02 23:18:25 - mmengine - INFO - Epoch(train) [39][20/32] lr: 1.0000e-06 eta: 0:36:33 time: 0.4833 data_time: 0.0014 memory: 24319 loss: 0.1103 loss_ce: 0.1103 2023/03/02 23:18:29 - mmengine - INFO - Epoch(train) [39][30/32] lr: 1.0000e-06 eta: 0:36:29 time: 0.4216 data_time: 0.0013 memory: 17919 loss: 0.1370 loss_ce: 0.1370 2023/03/02 23:18:30 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:18:35 - mmengine - INFO - Epoch(train) [40][10/32] lr: 1.0000e-06 eta: 0:36:26 time: 0.5172 data_time: 0.1008 memory: 18413 loss: 0.1265 loss_ce: 0.1265 2023/03/02 23:18:39 - mmengine - INFO - Epoch(train) [40][20/32] lr: 1.0000e-06 eta: 0:36:24 time: 0.4651 data_time: 0.0014 memory: 19958 loss: 0.1445 loss_ce: 0.1445 2023/03/02 23:18:44 - mmengine - INFO - Epoch(train) [40][30/32] lr: 1.0000e-06 eta: 0:36:20 time: 0.4315 data_time: 0.0011 memory: 19346 loss: 0.1192 loss_ce: 0.1192 2023/03/02 23:18:44 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:19:00 - mmengine - INFO - Epoch(val) [40][10/63] eta: 0:01:20 time: 1.5283 data_time: 0.0030 memory: 16226 2023/03/02 23:19:24 - mmengine - INFO - Epoch(val) [40][20/63] eta: 0:01:25 time: 2.4601 data_time: 0.0004 memory: 1075 2023/03/02 23:19:51 - mmengine - INFO - Epoch(val) [40][30/63] eta: 0:01:13 time: 2.6875 data_time: 0.0004 memory: 1075 2023/03/02 23:20:03 - mmengine - INFO - Epoch(val) [40][40/63] eta: 0:00:45 time: 1.1874 data_time: 0.0004 memory: 1075 2023/03/02 23:20:25 - mmengine - INFO - Epoch(val) [40][50/63] eta: 0:00:26 time: 2.2507 data_time: 0.0004 memory: 1075 2023/03/02 23:20:39 - mmengine - INFO - Epoch(val) [40][60/63] eta: 0:00:05 time: 1.4040 data_time: 0.0003 memory: 1075 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.80, recall: 0.6115, precision: 0.7582, hmean: 0.6770 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.81, recall: 0.6081, precision: 0.7641, hmean: 0.6772 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.82, recall: 0.6052, precision: 0.7707, hmean: 0.6780 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.83, recall: 0.6028, precision: 0.7791, hmean: 0.6797 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.84, recall: 0.6004, precision: 0.7828, hmean: 0.6796 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.85, recall: 0.5989, precision: 0.7913, hmean: 0.6818 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.86, recall: 0.5956, precision: 0.7970, hmean: 0.6817 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.87, recall: 0.5903, precision: 0.8076, hmean: 0.6821 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.88, recall: 0.5859, precision: 0.8151, hmean: 0.6818 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.89, recall: 0.5831, precision: 0.8249, hmean: 0.6832 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.90, recall: 0.5782, precision: 0.8311, hmean: 0.6820 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.91, recall: 0.5691, precision: 0.8425, hmean: 0.6793 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.92, recall: 0.5595, precision: 0.8507, hmean: 0.6750 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.93, recall: 0.5508, precision: 0.8621, hmean: 0.6722 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.94, recall: 0.5359, precision: 0.8648, hmean: 0.6617 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.95, recall: 0.5229, precision: 0.8723, hmean: 0.6538 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.96, recall: 0.5065, precision: 0.8781, hmean: 0.6424 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.97, recall: 0.4858, precision: 0.8906, hmean: 0.6287 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.98, recall: 0.4588, precision: 0.9008, hmean: 0.6080 2023/03/02 23:21:39 - mmengine - INFO - text score threshold: 0.99, recall: 0.4141, precision: 0.9149, hmean: 0.5701 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.80, recall: 0.6553, precision: 0.8125, hmean: 0.7255 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.81, recall: 0.6509, precision: 0.8179, hmean: 0.7249 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.82, recall: 0.6471, precision: 0.8240, hmean: 0.7249 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.83, recall: 0.6423, precision: 0.8301, hmean: 0.7242 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.84, recall: 0.6389, precision: 0.8330, hmean: 0.7232 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.85, recall: 0.6351, precision: 0.8391, hmean: 0.7229 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.86, recall: 0.6293, precision: 0.8421, hmean: 0.7203 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.87, recall: 0.6235, precision: 0.8531, hmean: 0.7204 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.88, recall: 0.6182, precision: 0.8600, hmean: 0.7193 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.89, recall: 0.6139, precision: 0.8685, hmean: 0.7193 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.90, recall: 0.6076, precision: 0.8734, hmean: 0.7166 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.91, recall: 0.5956, precision: 0.8817, hmean: 0.7109 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.92, recall: 0.5840, precision: 0.8880, hmean: 0.7046 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.93, recall: 0.5715, precision: 0.8945, hmean: 0.6974 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.94, recall: 0.5556, precision: 0.8967, hmean: 0.6861 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.95, recall: 0.5388, precision: 0.8988, hmean: 0.6737 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.96, recall: 0.5205, precision: 0.9023, hmean: 0.6602 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.97, recall: 0.4969, precision: 0.9109, hmean: 0.6430 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.98, recall: 0.4685, precision: 0.9197, hmean: 0.6207 2023/03/02 23:21:48 - mmengine - INFO - text score threshold: 0.99, recall: 0.4203, precision: 0.9287, hmean: 0.5787 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.80, recall: 0.7130, precision: 0.8842, hmean: 0.7894 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.81, recall: 0.7068, precision: 0.8881, hmean: 0.7871 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.82, recall: 0.7020, precision: 0.8939, hmean: 0.7864 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.83, recall: 0.6943, precision: 0.8973, hmean: 0.7828 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.84, recall: 0.6899, precision: 0.8996, hmean: 0.7809 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.85, recall: 0.6846, precision: 0.9046, hmean: 0.7794 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.86, recall: 0.6779, precision: 0.9072, hmean: 0.7760 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.87, recall: 0.6692, precision: 0.9157, hmean: 0.7733 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.88, recall: 0.6615, precision: 0.9203, hmean: 0.7697 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.89, recall: 0.6524, precision: 0.9230, hmean: 0.7645 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.90, recall: 0.6447, precision: 0.9266, hmean: 0.7604 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.91, recall: 0.6278, precision: 0.9294, hmean: 0.7494 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.92, recall: 0.6124, precision: 0.9312, hmean: 0.7389 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.93, recall: 0.5985, precision: 0.9367, hmean: 0.7303 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.94, recall: 0.5816, precision: 0.9386, hmean: 0.7182 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.95, recall: 0.5643, precision: 0.9414, hmean: 0.7056 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.96, recall: 0.5445, precision: 0.9441, hmean: 0.6907 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.97, recall: 0.5185, precision: 0.9506, hmean: 0.6710 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.98, recall: 0.4872, precision: 0.9565, hmean: 0.6456 2023/03/02 23:21:58 - mmengine - INFO - text score threshold: 0.99, recall: 0.4362, precision: 0.9638, hmean: 0.6006 2023/03/02 23:21:58 - mmengine - INFO - Epoch(val) [40][63/63] generic/precision: 0.8249 generic/recall: 0.5831 generic/hmean: 0.6832 weak/precision: 0.8125 weak/recall: 0.6553 weak/hmean: 0.7255 strong/precision: 0.8842 strong/recall: 0.7130 strong/hmean: 0.7894 2023/03/02 23:21:58 - mmengine - INFO - The previous best checkpoint mmocr/projects/SPTS/work_dirs/spts_resnet50_350e_icdar2015/best_generic/hmean_epoch_30.pth is removed 2023/03/02 23:22:00 - mmengine - INFO - The best checkpoint with 0.6832 generic/hmean at 40 epoch is saved to best_generic/hmean_epoch_40.pth. 2023/03/02 23:22:05 - mmengine - INFO - Epoch(train) [41][10/32] lr: 1.0000e-06 eta: 0:36:18 time: 0.5424 data_time: 0.0724 memory: 20424 loss: 0.1270 loss_ce: 0.1270 2023/03/02 23:22:10 - mmengine - INFO - Epoch(train) [41][20/32] lr: 1.0000e-06 eta: 0:36:14 time: 0.4293 data_time: 0.0014 memory: 17919 loss: 0.1181 loss_ce: 0.1181 2023/03/02 23:22:14 - mmengine - INFO - Epoch(train) [41][30/32] lr: 1.0000e-06 eta: 0:36:11 time: 0.4637 data_time: 0.0014 memory: 17919 loss: 0.0991 loss_ce: 0.0991 2023/03/02 23:22:15 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:22:20 - mmengine - INFO - Epoch(train) [42][10/32] lr: 1.0000e-06 eta: 0:36:09 time: 0.5088 data_time: 0.0676 memory: 19958 loss: 0.1103 loss_ce: 0.1103 2023/03/02 23:22:24 - mmengine - INFO - Epoch(train) [42][20/32] lr: 1.0000e-06 eta: 0:36:04 time: 0.4134 data_time: 0.0013 memory: 19751 loss: 0.1218 loss_ce: 0.1218 2023/03/02 23:22:28 - mmengine - INFO - Epoch(train) [42][30/32] lr: 1.0000e-06 eta: 0:35:59 time: 0.4231 data_time: 0.0012 memory: 19150 loss: 0.1151 loss_ce: 0.1151 2023/03/02 23:22:29 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:22:34 - mmengine - INFO - Epoch(train) [43][10/32] lr: 1.0000e-06 eta: 0:35:58 time: 0.5224 data_time: 0.0763 memory: 18083 loss: 0.1152 loss_ce: 0.1152 2023/03/02 23:22:39 - mmengine - INFO - Epoch(train) [43][20/32] lr: 1.0000e-06 eta: 0:35:55 time: 0.4623 data_time: 0.0014 memory: 22104 loss: 0.1205 loss_ce: 0.1205 2023/03/02 23:22:44 - mmengine - INFO - Epoch(train) [43][30/32] lr: 1.0000e-06 eta: 0:35:52 time: 0.4670 data_time: 0.0012 memory: 18746 loss: 0.1136 loss_ce: 0.1136 2023/03/02 23:22:44 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:22:50 - mmengine - INFO - Epoch(train) [44][10/32] lr: 1.0000e-06 eta: 0:35:50 time: 0.5274 data_time: 0.0722 memory: 18083 loss: 0.1222 loss_ce: 0.1222 2023/03/02 23:22:54 - mmengine - INFO - Epoch(train) [44][20/32] lr: 1.0000e-06 eta: 0:35:47 time: 0.4696 data_time: 0.0014 memory: 20582 loss: 0.1203 loss_ce: 0.1203 2023/03/02 23:22:59 - mmengine - INFO - Epoch(train) [44][30/32] lr: 1.0000e-06 eta: 0:35:44 time: 0.4708 data_time: 0.0012 memory: 19142 loss: 0.1022 loss_ce: 0.1022 2023/03/02 23:23:00 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:23:05 - mmengine - INFO - Epoch(train) [45][10/32] lr: 1.0000e-06 eta: 0:35:41 time: 0.5113 data_time: 0.0519 memory: 20716 loss: 0.0984 loss_ce: 0.0984 2023/03/02 23:23:10 - mmengine - INFO - Epoch(train) [45][20/32] lr: 1.0000e-06 eta: 0:35:38 time: 0.4761 data_time: 0.0015 memory: 18654 loss: 0.0959 loss_ce: 0.0959 2023/03/02 23:23:14 - mmengine - INFO - Epoch(train) [45][30/32] lr: 1.0000e-06 eta: 0:35:35 time: 0.4744 data_time: 0.0011 memory: 24875 loss: 0.1186 loss_ce: 0.1186 2023/03/02 23:23:15 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:23:20 - mmengine - INFO - Epoch(train) [46][10/32] lr: 1.0000e-06 eta: 0:35:31 time: 0.4844 data_time: 0.0646 memory: 17428 loss: 0.1232 loss_ce: 0.1232 2023/03/02 23:23:25 - mmengine - INFO - Epoch(train) [46][20/32] lr: 1.0000e-06 eta: 0:35:29 time: 0.4740 data_time: 0.0012 memory: 18741 loss: 0.1221 loss_ce: 0.1221 2023/03/02 23:23:29 - mmengine - INFO - Epoch(train) [46][30/32] lr: 1.0000e-06 eta: 0:35:25 time: 0.4632 data_time: 0.0013 memory: 24058 loss: 0.0974 loss_ce: 0.0974 2023/03/02 23:23:30 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:23:35 - mmengine - INFO - Epoch(train) [47][10/32] lr: 1.0000e-06 eta: 0:35:23 time: 0.5396 data_time: 0.0584 memory: 19958 loss: 0.1116 loss_ce: 0.1116 2023/03/02 23:23:40 - mmengine - INFO - Epoch(train) [47][20/32] lr: 1.0000e-06 eta: 0:35:19 time: 0.4550 data_time: 0.0012 memory: 22984 loss: 0.1000 loss_ce: 0.1000 2023/03/02 23:23:44 - mmengine - INFO - Epoch(train) [47][30/32] lr: 1.0000e-06 eta: 0:35:16 time: 0.4623 data_time: 0.0011 memory: 22545 loss: 0.0938 loss_ce: 0.0938 2023/03/02 23:23:45 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:23:51 - mmengine - INFO - Epoch(train) [48][10/32] lr: 1.0000e-06 eta: 0:35:14 time: 0.5486 data_time: 0.0889 memory: 18774 loss: 0.1019 loss_ce: 0.1019 2023/03/02 23:23:55 - mmengine - INFO - Epoch(train) [48][20/32] lr: 1.0000e-06 eta: 0:35:10 time: 0.4423 data_time: 0.0012 memory: 25455 loss: 0.1164 loss_ce: 0.1164 2023/03/02 23:23:59 - mmengine - INFO - Epoch(train) [48][30/32] lr: 1.0000e-06 eta: 0:35:06 time: 0.4408 data_time: 0.0013 memory: 17936 loss: 0.1031 loss_ce: 0.1031 2023/03/02 23:24:00 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:24:05 - mmengine - INFO - Epoch(train) [49][10/32] lr: 1.0000e-06 eta: 0:35:02 time: 0.5015 data_time: 0.0773 memory: 19958 loss: 0.1178 loss_ce: 0.1178 2023/03/02 23:24:10 - mmengine - INFO - Epoch(train) [49][20/32] lr: 1.0000e-06 eta: 0:34:59 time: 0.4660 data_time: 0.0016 memory: 22545 loss: 0.0966 loss_ce: 0.0966 2023/03/02 23:24:14 - mmengine - INFO - Epoch(train) [49][30/32] lr: 1.0000e-06 eta: 0:34:54 time: 0.4129 data_time: 0.0013 memory: 18597 loss: 0.1115 loss_ce: 0.1115 2023/03/02 23:24:14 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:24:20 - mmengine - INFO - Epoch(train) [50][10/32] lr: 1.0000e-06 eta: 0:34:50 time: 0.5187 data_time: 0.0700 memory: 18535 loss: 0.1316 loss_ce: 0.1316 2023/03/02 23:24:24 - mmengine - INFO - Epoch(train) [50][20/32] lr: 1.0000e-06 eta: 0:34:46 time: 0.4435 data_time: 0.0014 memory: 18246 loss: 0.1101 loss_ce: 0.1101 2023/03/02 23:24:29 - mmengine - INFO - Epoch(train) [50][30/32] lr: 1.0000e-06 eta: 0:34:44 time: 0.4935 data_time: 0.0013 memory: 19958 loss: 0.1038 loss_ce: 0.1038 2023/03/02 23:24:30 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:24:45 - mmengine - INFO - Epoch(val) [50][10/63] eta: 0:01:22 time: 1.5488 data_time: 0.0029 memory: 18413 2023/03/02 23:25:10 - mmengine - INFO - Epoch(val) [50][20/63] eta: 0:01:27 time: 2.5020 data_time: 0.0004 memory: 1075 2023/03/02 23:25:36 - mmengine - INFO - Epoch(val) [50][30/63] eta: 0:01:13 time: 2.6337 data_time: 0.0005 memory: 1075 2023/03/02 23:25:47 - mmengine - INFO - Epoch(val) [50][40/63] eta: 0:00:44 time: 1.0939 data_time: 0.0003 memory: 1075 2023/03/02 23:26:11 - mmengine - INFO - Epoch(val) [50][50/63] eta: 0:00:26 time: 2.3524 data_time: 0.0005 memory: 1075 2023/03/02 23:26:24 - mmengine - INFO - Epoch(val) [50][60/63] eta: 0:00:05 time: 1.3445 data_time: 0.0003 memory: 1075 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.80, recall: 0.6086, precision: 0.7569, hmean: 0.6747 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.81, recall: 0.6081, precision: 0.7641, hmean: 0.6772 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.82, recall: 0.6057, precision: 0.7689, hmean: 0.6776 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.83, recall: 0.6042, precision: 0.7766, hmean: 0.6797 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.84, recall: 0.6023, precision: 0.7843, hmean: 0.6814 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.85, recall: 0.5989, precision: 0.7893, hmean: 0.6811 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.86, recall: 0.5970, precision: 0.7959, hmean: 0.6823 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.87, recall: 0.5908, precision: 0.8062, hmean: 0.6819 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.88, recall: 0.5864, precision: 0.8142, hmean: 0.6818 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.89, recall: 0.5821, precision: 0.8253, hmean: 0.6827 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.90, recall: 0.5753, precision: 0.8281, hmean: 0.6790 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.91, recall: 0.5701, precision: 0.8409, hmean: 0.6795 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.92, recall: 0.5580, precision: 0.8448, hmean: 0.6721 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.93, recall: 0.5469, precision: 0.8522, hmean: 0.6663 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.94, recall: 0.5349, precision: 0.8626, hmean: 0.6603 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.95, recall: 0.5243, precision: 0.8712, hmean: 0.6546 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.96, recall: 0.5070, precision: 0.8775, hmean: 0.6427 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.97, recall: 0.4868, precision: 0.8845, hmean: 0.6280 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.98, recall: 0.4569, precision: 0.8995, hmean: 0.6060 2023/03/02 23:27:27 - mmengine - INFO - text score threshold: 0.99, recall: 0.4160, precision: 0.9114, hmean: 0.5712 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.80, recall: 0.6562, precision: 0.8162, hmean: 0.7275 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.81, recall: 0.6553, precision: 0.8234, hmean: 0.7298 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.82, recall: 0.6519, precision: 0.8276, hmean: 0.7293 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.83, recall: 0.6481, precision: 0.8329, hmean: 0.7289 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.84, recall: 0.6428, precision: 0.8370, hmean: 0.7271 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.85, recall: 0.6384, precision: 0.8414, hmean: 0.7260 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.86, recall: 0.6346, precision: 0.8460, hmean: 0.7252 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.87, recall: 0.6264, precision: 0.8548, hmean: 0.7230 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.88, recall: 0.6206, precision: 0.8616, hmean: 0.7215 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.89, recall: 0.6129, precision: 0.8689, hmean: 0.7188 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.90, recall: 0.6057, precision: 0.8718, hmean: 0.7148 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.91, recall: 0.5985, precision: 0.8828, hmean: 0.7133 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.92, recall: 0.5855, precision: 0.8863, hmean: 0.7051 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.93, recall: 0.5705, precision: 0.8890, hmean: 0.6950 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.94, recall: 0.5542, precision: 0.8936, hmean: 0.6841 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.95, recall: 0.5402, precision: 0.8976, hmean: 0.6745 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.96, recall: 0.5214, precision: 0.9025, hmean: 0.6610 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.97, recall: 0.4993, precision: 0.9073, hmean: 0.6441 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.98, recall: 0.4670, precision: 0.9194, hmean: 0.6194 2023/03/02 23:27:36 - mmengine - INFO - text score threshold: 0.99, recall: 0.4222, precision: 0.9251, hmean: 0.5798 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.80, recall: 0.7116, precision: 0.8850, hmean: 0.7889 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.81, recall: 0.7097, precision: 0.8917, hmean: 0.7903 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.82, recall: 0.7058, precision: 0.8961, hmean: 0.7897 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.83, recall: 0.7005, precision: 0.9004, hmean: 0.7880 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.84, recall: 0.6938, precision: 0.9034, hmean: 0.7849 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.85, recall: 0.6890, precision: 0.9080, hmean: 0.7835 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.86, recall: 0.6837, precision: 0.9114, hmean: 0.7813 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.87, recall: 0.6712, precision: 0.9159, hmean: 0.7747 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.88, recall: 0.6635, precision: 0.9211, hmean: 0.7713 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.89, recall: 0.6524, precision: 0.9249, hmean: 0.7651 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.90, recall: 0.6447, precision: 0.9279, hmean: 0.7608 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.91, recall: 0.6307, precision: 0.9304, hmean: 0.7518 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.92, recall: 0.6158, precision: 0.9322, hmean: 0.7417 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.93, recall: 0.5994, precision: 0.9340, hmean: 0.7302 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.94, recall: 0.5811, precision: 0.9371, hmean: 0.7174 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.95, recall: 0.5662, precision: 0.9408, hmean: 0.7069 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.96, recall: 0.5450, precision: 0.9433, hmean: 0.6909 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.97, recall: 0.5214, precision: 0.9475, hmean: 0.6727 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.98, recall: 0.4848, precision: 0.9545, hmean: 0.6430 2023/03/02 23:27:45 - mmengine - INFO - text score threshold: 0.99, recall: 0.4381, precision: 0.9599, hmean: 0.6017 2023/03/02 23:27:45 - mmengine - INFO - Epoch(val) [50][63/63] generic/precision: 0.8253 generic/recall: 0.5821 generic/hmean: 0.6827 weak/precision: 0.8234 weak/recall: 0.6553 weak/hmean: 0.7298 strong/precision: 0.8917 strong/recall: 0.7097 strong/hmean: 0.7903 2023/03/02 23:27:50 - mmengine - INFO - Epoch(train) [51][10/32] lr: 1.0000e-06 eta: 0:34:41 time: 0.5451 data_time: 0.0726 memory: 19150 loss: 0.1180 loss_ce: 0.1180 2023/03/02 23:27:55 - mmengine - INFO - Epoch(train) [51][20/32] lr: 1.0000e-06 eta: 0:34:36 time: 0.4149 data_time: 0.0012 memory: 17919 loss: 0.1255 loss_ce: 0.1255 2023/03/02 23:27:59 - mmengine - INFO - Epoch(train) [51][30/32] lr: 1.0000e-06 eta: 0:34:32 time: 0.4333 data_time: 0.0012 memory: 18413 loss: 0.1202 loss_ce: 0.1202 2023/03/02 23:28:00 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:28:05 - mmengine - INFO - Epoch(train) [52][10/32] lr: 1.0000e-06 eta: 0:34:29 time: 0.5015 data_time: 0.0323 memory: 24261 loss: 0.1087 loss_ce: 0.1087 2023/03/02 23:28:09 - mmengine - INFO - Epoch(train) [52][20/32] lr: 1.0000e-06 eta: 0:34:24 time: 0.4323 data_time: 0.0012 memory: 19958 loss: 0.1068 loss_ce: 0.1068 2023/03/02 23:28:13 - mmengine - INFO - Epoch(train) [52][30/32] lr: 1.0000e-06 eta: 0:34:20 time: 0.4378 data_time: 0.0013 memory: 19552 loss: 0.1293 loss_ce: 0.1293 2023/03/02 23:28:14 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:28:20 - mmengine - INFO - Epoch(train) [53][10/32] lr: 1.0000e-06 eta: 0:34:17 time: 0.5513 data_time: 0.0840 memory: 18083 loss: 0.1024 loss_ce: 0.1024 2023/03/02 23:28:24 - mmengine - INFO - Epoch(train) [53][20/32] lr: 1.0000e-06 eta: 0:34:13 time: 0.4375 data_time: 0.0014 memory: 18597 loss: 0.1255 loss_ce: 0.1255 2023/03/02 23:28:28 - mmengine - INFO - Epoch(train) [53][30/32] lr: 1.0000e-06 eta: 0:34:09 time: 0.4500 data_time: 0.0012 memory: 22104 loss: 0.1198 loss_ce: 0.1198 2023/03/02 23:28:29 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:28:34 - mmengine - INFO - Epoch(train) [54][10/32] lr: 1.0000e-06 eta: 0:34:06 time: 0.5283 data_time: 0.0779 memory: 17935 loss: 0.1227 loss_ce: 0.1227 2023/03/02 23:28:39 - mmengine - INFO - Epoch(train) [54][20/32] lr: 1.0000e-06 eta: 0:34:01 time: 0.4263 data_time: 0.0012 memory: 18083 loss: 0.1272 loss_ce: 0.1272 2023/03/02 23:28:43 - mmengine - INFO - Epoch(train) [54][30/32] lr: 1.0000e-06 eta: 0:33:57 time: 0.4344 data_time: 0.0012 memory: 18597 loss: 0.1191 loss_ce: 0.1191 2023/03/02 23:28:44 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:28:49 - mmengine - INFO - Epoch(train) [55][10/32] lr: 1.0000e-06 eta: 0:33:54 time: 0.5568 data_time: 0.0678 memory: 24319 loss: 0.1082 loss_ce: 0.1082 2023/03/02 23:28:54 - mmengine - INFO - Epoch(train) [55][20/32] lr: 1.0000e-06 eta: 0:33:50 time: 0.4642 data_time: 0.0015 memory: 18325 loss: 0.1066 loss_ce: 0.1066 2023/03/02 23:28:58 - mmengine - INFO - Epoch(train) [55][30/32] lr: 1.0000e-06 eta: 0:33:46 time: 0.4500 data_time: 0.0011 memory: 20820 loss: 0.0995 loss_ce: 0.0995 2023/03/02 23:28:59 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:29:04 - mmengine - INFO - Epoch(train) [56][10/32] lr: 1.0000e-06 eta: 0:33:42 time: 0.5106 data_time: 0.0422 memory: 25455 loss: 0.1111 loss_ce: 0.1111 2023/03/02 23:29:09 - mmengine - INFO - Epoch(train) [56][20/32] lr: 1.0000e-06 eta: 0:33:39 time: 0.4706 data_time: 0.0015 memory: 19958 loss: 0.0981 loss_ce: 0.0981 2023/03/02 23:29:13 - mmengine - INFO - Epoch(train) [56][30/32] lr: 1.0000e-06 eta: 0:33:34 time: 0.4269 data_time: 0.0013 memory: 20172 loss: 0.1006 loss_ce: 0.1006 2023/03/02 23:29:14 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:29:19 - mmengine - INFO - Epoch(train) [57][10/32] lr: 1.0000e-06 eta: 0:33:32 time: 0.5645 data_time: 0.0908 memory: 25344 loss: 0.1219 loss_ce: 0.1219 2023/03/02 23:29:24 - mmengine - INFO - Epoch(train) [57][20/32] lr: 1.0000e-06 eta: 0:33:27 time: 0.4211 data_time: 0.0014 memory: 18083 loss: 0.1392 loss_ce: 0.1392 2023/03/02 23:29:28 - mmengine - INFO - Epoch(train) [57][30/32] lr: 1.0000e-06 eta: 0:33:24 time: 0.4809 data_time: 0.0011 memory: 18246 loss: 0.1061 loss_ce: 0.1061 2023/03/02 23:29:29 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:29:34 - mmengine - INFO - Epoch(train) [58][10/32] lr: 1.0000e-06 eta: 0:33:20 time: 0.5179 data_time: 0.0513 memory: 18966 loss: 0.1183 loss_ce: 0.1183 2023/03/02 23:29:39 - mmengine - INFO - Epoch(train) [58][20/32] lr: 1.0000e-06 eta: 0:33:15 time: 0.4392 data_time: 0.0014 memory: 19174 loss: 0.1138 loss_ce: 0.1138 2023/03/02 23:29:43 - mmengine - INFO - Epoch(train) [58][30/32] lr: 1.0000e-06 eta: 0:33:12 time: 0.4711 data_time: 0.0013 memory: 24875 loss: 0.1075 loss_ce: 0.1075 2023/03/02 23:29:44 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:29:50 - mmengine - INFO - Epoch(train) [59][10/32] lr: 1.0000e-06 eta: 0:33:09 time: 0.5669 data_time: 0.0995 memory: 18535 loss: 0.1156 loss_ce: 0.1156 2023/03/02 23:29:54 - mmengine - INFO - Epoch(train) [59][20/32] lr: 1.0000e-06 eta: 0:33:05 time: 0.4546 data_time: 0.0014 memory: 20600 loss: 0.1096 loss_ce: 0.1096 2023/03/02 23:29:59 - mmengine - INFO - Epoch(train) [59][30/32] lr: 1.0000e-06 eta: 0:33:01 time: 0.4659 data_time: 0.0013 memory: 19552 loss: 0.1077 loss_ce: 0.1077 2023/03/02 23:29:59 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:30:05 - mmengine - INFO - Epoch(train) [60][10/32] lr: 1.0000e-06 eta: 0:32:58 time: 0.5510 data_time: 0.0852 memory: 18982 loss: 0.1105 loss_ce: 0.1105 2023/03/02 23:30:09 - mmengine - INFO - Epoch(train) [60][20/32] lr: 1.0000e-06 eta: 0:32:53 time: 0.4313 data_time: 0.0014 memory: 19958 loss: 0.1157 loss_ce: 0.1157 2023/03/02 23:30:14 - mmengine - INFO - Epoch(train) [60][30/32] lr: 1.0000e-06 eta: 0:32:50 time: 0.4756 data_time: 0.0012 memory: 19751 loss: 0.1089 loss_ce: 0.1089 2023/03/02 23:30:15 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:30:29 - mmengine - INFO - Epoch(val) [60][10/63] eta: 0:01:15 time: 1.4183 data_time: 0.0028 memory: 17137 2023/03/02 23:30:53 - mmengine - INFO - Epoch(val) [60][20/63] eta: 0:01:22 time: 2.4337 data_time: 0.0003 memory: 1075 2023/03/02 23:31:18 - mmengine - INFO - Epoch(val) [60][30/63] eta: 0:01:09 time: 2.4601 data_time: 0.0004 memory: 1075 2023/03/02 23:31:29 - mmengine - INFO - Epoch(val) [60][40/63] eta: 0:00:42 time: 1.1419 data_time: 0.0004 memory: 1075 2023/03/02 23:31:52 - mmengine - INFO - Epoch(val) [60][50/63] eta: 0:00:25 time: 2.3137 data_time: 0.0006 memory: 1075 2023/03/02 23:32:06 - mmengine - INFO - Epoch(val) [60][60/63] eta: 0:00:05 time: 1.3283 data_time: 0.0003 memory: 1075 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.80, recall: 0.6086, precision: 0.7519, hmean: 0.6727 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.81, recall: 0.6081, precision: 0.7613, hmean: 0.6761 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.82, recall: 0.6071, precision: 0.7675, hmean: 0.6780 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.83, recall: 0.6047, precision: 0.7763, hmean: 0.6798 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.84, recall: 0.6018, precision: 0.7837, hmean: 0.6808 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.85, recall: 0.5985, precision: 0.7927, hmean: 0.6820 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.86, recall: 0.5956, precision: 0.8012, hmean: 0.6832 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.87, recall: 0.5927, precision: 0.8083, hmean: 0.6839 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.88, recall: 0.5869, precision: 0.8138, hmean: 0.6820 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.89, recall: 0.5816, precision: 0.8212, hmean: 0.6809 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.90, recall: 0.5763, precision: 0.8301, hmean: 0.6803 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.91, recall: 0.5705, precision: 0.8404, hmean: 0.6797 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.92, recall: 0.5614, precision: 0.8480, hmean: 0.6756 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.93, recall: 0.5513, precision: 0.8551, hmean: 0.6704 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.94, recall: 0.5388, precision: 0.8628, hmean: 0.6633 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.95, recall: 0.5248, precision: 0.8699, hmean: 0.6547 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.96, recall: 0.5084, precision: 0.8778, hmean: 0.6439 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.97, recall: 0.4868, precision: 0.8814, hmean: 0.6272 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.98, recall: 0.4636, precision: 0.8950, hmean: 0.6108 2023/03/02 23:33:06 - mmengine - INFO - text score threshold: 0.99, recall: 0.4218, precision: 0.9135, hmean: 0.5771 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.80, recall: 0.6558, precision: 0.8102, hmean: 0.7249 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.81, recall: 0.6543, precision: 0.8192, hmean: 0.7275 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.82, recall: 0.6529, precision: 0.8253, hmean: 0.7290 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.83, recall: 0.6485, precision: 0.8325, hmean: 0.7291 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.84, recall: 0.6437, precision: 0.8382, hmean: 0.7282 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.85, recall: 0.6370, precision: 0.8438, hmean: 0.7259 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.86, recall: 0.6331, precision: 0.8517, hmean: 0.7263 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.87, recall: 0.6298, precision: 0.8588, hmean: 0.7267 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.88, recall: 0.6216, precision: 0.8618, hmean: 0.7222 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.89, recall: 0.6148, precision: 0.8681, hmean: 0.7198 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.90, recall: 0.6071, precision: 0.8745, hmean: 0.7167 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.91, recall: 0.5985, precision: 0.8816, hmean: 0.7129 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.92, recall: 0.5869, precision: 0.8865, hmean: 0.7063 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.93, recall: 0.5739, precision: 0.8902, hmean: 0.6979 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.94, recall: 0.5585, precision: 0.8944, hmean: 0.6876 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.95, recall: 0.5412, precision: 0.8970, hmean: 0.6751 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.96, recall: 0.5214, precision: 0.9002, hmean: 0.6604 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.97, recall: 0.4998, precision: 0.9050, hmean: 0.6439 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.98, recall: 0.4728, precision: 0.9126, hmean: 0.6229 2023/03/02 23:33:15 - mmengine - INFO - text score threshold: 0.99, recall: 0.4275, precision: 0.9260, hmean: 0.5850 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.80, recall: 0.7116, precision: 0.8792, hmean: 0.7866 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.81, recall: 0.7087, precision: 0.8873, hmean: 0.7880 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.82, recall: 0.7063, precision: 0.8929, hmean: 0.7887 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.83, recall: 0.7000, precision: 0.8986, hmean: 0.7870 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.84, recall: 0.6938, precision: 0.9034, hmean: 0.7849 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.85, recall: 0.6861, precision: 0.9088, hmean: 0.7819 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.86, recall: 0.6798, precision: 0.9145, hmean: 0.7799 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.87, recall: 0.6736, precision: 0.9186, hmean: 0.7772 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.88, recall: 0.6635, precision: 0.9199, hmean: 0.7709 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.89, recall: 0.6533, precision: 0.9225, hmean: 0.7649 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.90, recall: 0.6428, precision: 0.9258, hmean: 0.7587 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.91, recall: 0.6322, precision: 0.9312, hmean: 0.7531 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.92, recall: 0.6192, precision: 0.9353, hmean: 0.7451 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.93, recall: 0.6033, precision: 0.9358, hmean: 0.7336 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.94, recall: 0.5859, precision: 0.9383, hmean: 0.7214 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.95, recall: 0.5681, precision: 0.9417, hmean: 0.7087 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.96, recall: 0.5484, precision: 0.9468, hmean: 0.6945 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.97, recall: 0.5248, precision: 0.9503, hmean: 0.6762 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.98, recall: 0.4940, precision: 0.9535, hmean: 0.6508 2023/03/02 23:33:24 - mmengine - INFO - text score threshold: 0.99, recall: 0.4449, precision: 0.9635, hmean: 0.6087 2023/03/02 23:33:24 - mmengine - INFO - Epoch(val) [60][63/63] generic/precision: 0.8083 generic/recall: 0.5927 generic/hmean: 0.6839 weak/precision: 0.8325 weak/recall: 0.6485 weak/hmean: 0.7291 strong/precision: 0.8929 strong/recall: 0.7063 strong/hmean: 0.7887 2023/03/02 23:33:24 - mmengine - INFO - The previous best checkpoint mmocr/projects/SPTS/work_dirs/spts_resnet50_350e_icdar2015/best_generic/hmean_epoch_40.pth is removed 2023/03/02 23:33:27 - mmengine - INFO - The best checkpoint with 0.6839 generic/hmean at 60 epoch is saved to best_generic/hmean_epoch_60.pth. 2023/03/02 23:33:32 - mmengine - INFO - Epoch(train) [61][10/32] lr: 1.0000e-06 eta: 0:32:46 time: 0.5376 data_time: 0.0452 memory: 25455 loss: 0.1089 loss_ce: 0.1089 2023/03/02 23:33:36 - mmengine - INFO - Epoch(train) [61][20/32] lr: 1.0000e-06 eta: 0:32:41 time: 0.4183 data_time: 0.0013 memory: 17428 loss: 0.1225 loss_ce: 0.1225 2023/03/02 23:33:41 - mmengine - INFO - Epoch(train) [61][30/32] lr: 1.0000e-06 eta: 0:32:37 time: 0.4535 data_time: 0.0012 memory: 18165 loss: 0.1028 loss_ce: 0.1028 2023/03/02 23:33:41 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:33:47 - mmengine - INFO - Epoch(train) [62][10/32] lr: 1.0000e-06 eta: 0:32:35 time: 0.5858 data_time: 0.0881 memory: 25344 loss: 0.0977 loss_ce: 0.0977 2023/03/02 23:33:52 - mmengine - INFO - Epoch(train) [62][20/32] lr: 1.0000e-06 eta: 0:32:31 time: 0.4552 data_time: 0.0012 memory: 23431 loss: 0.1219 loss_ce: 0.1219 2023/03/02 23:33:56 - mmengine - INFO - Epoch(train) [62][30/32] lr: 1.0000e-06 eta: 0:32:27 time: 0.4658 data_time: 0.0010 memory: 19344 loss: 0.1094 loss_ce: 0.1094 2023/03/02 23:33:57 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:34:03 - mmengine - INFO - Epoch(train) [63][10/32] lr: 1.0000e-06 eta: 0:32:25 time: 0.6258 data_time: 0.0683 memory: 18737 loss: 0.1230 loss_ce: 0.1230 2023/03/02 23:34:06 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:34:08 - mmengine - INFO - Epoch(train) [63][20/32] lr: 1.0000e-06 eta: 0:32:21 time: 0.4461 data_time: 0.0012 memory: 18535 loss: 0.1246 loss_ce: 0.1246 2023/03/02 23:34:12 - mmengine - INFO - Epoch(train) [63][30/32] lr: 1.0000e-06 eta: 0:32:17 time: 0.4671 data_time: 0.0010 memory: 19679 loss: 0.1080 loss_ce: 0.1080 2023/03/02 23:34:13 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:34:19 - mmengine - INFO - Epoch(train) [64][10/32] lr: 1.0000e-06 eta: 0:32:14 time: 0.5866 data_time: 0.0508 memory: 19958 loss: 0.1021 loss_ce: 0.1021 2023/03/02 23:34:23 - mmengine - INFO - Epoch(train) [64][20/32] lr: 1.0000e-06 eta: 0:32:10 time: 0.4288 data_time: 0.0012 memory: 17733 loss: 0.1299 loss_ce: 0.1299 2023/03/02 23:34:28 - mmengine - INFO - Epoch(train) [64][30/32] lr: 1.0000e-06 eta: 0:32:05 time: 0.4468 data_time: 0.0014 memory: 18246 loss: 0.1166 loss_ce: 0.1166 2023/03/02 23:34:28 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:34:34 - mmengine - INFO - Epoch(train) [65][10/32] lr: 1.0000e-06 eta: 0:32:02 time: 0.5592 data_time: 0.0952 memory: 17919 loss: 0.1069 loss_ce: 0.1069 2023/03/02 23:34:38 - mmengine - INFO - Epoch(train) [65][20/32] lr: 1.0000e-06 eta: 0:31:58 time: 0.4528 data_time: 0.0017 memory: 22984 loss: 0.0967 loss_ce: 0.0967 2023/03/02 23:34:43 - mmengine - INFO - Epoch(train) [65][30/32] lr: 1.0000e-06 eta: 0:31:53 time: 0.4533 data_time: 0.0014 memory: 20172 loss: 0.1168 loss_ce: 0.1168 2023/03/02 23:34:44 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:34:49 - mmengine - INFO - Epoch(train) [66][10/32] lr: 1.0000e-06 eta: 0:31:50 time: 0.5712 data_time: 0.1207 memory: 20172 loss: 0.0930 loss_ce: 0.0930 2023/03/02 23:34:54 - mmengine - INFO - Epoch(train) [66][20/32] lr: 1.0000e-06 eta: 0:31:46 time: 0.4379 data_time: 0.0013 memory: 19958 loss: 0.1048 loss_ce: 0.1048 2023/03/02 23:34:58 - mmengine - INFO - Epoch(train) [66][30/32] lr: 1.0000e-06 eta: 0:31:41 time: 0.4409 data_time: 0.0013 memory: 17583 loss: 0.1078 loss_ce: 0.1078 2023/03/02 23:34:59 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:35:05 - mmengine - INFO - Epoch(train) [67][10/32] lr: 1.0000e-06 eta: 0:31:38 time: 0.5519 data_time: 0.0787 memory: 19958 loss: 0.0964 loss_ce: 0.0964 2023/03/02 23:35:09 - mmengine - INFO - Epoch(train) [67][20/32] lr: 1.0000e-06 eta: 0:31:34 time: 0.4751 data_time: 0.0013 memory: 18830 loss: 0.1196 loss_ce: 0.1196 2023/03/02 23:35:14 - mmengine - INFO - Epoch(train) [67][30/32] lr: 1.0000e-06 eta: 0:31:31 time: 0.5070 data_time: 0.0011 memory: 27446 loss: 0.1048 loss_ce: 0.1048 2023/03/02 23:35:15 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:35:20 - mmengine - INFO - Epoch(train) [68][10/32] lr: 1.0000e-06 eta: 0:31:27 time: 0.5421 data_time: 0.0225 memory: 24319 loss: 0.1308 loss_ce: 0.1308 2023/03/02 23:35:25 - mmengine - INFO - Epoch(train) [68][20/32] lr: 1.0000e-06 eta: 0:31:23 time: 0.4639 data_time: 0.0015 memory: 21276 loss: 0.1094 loss_ce: 0.1094 2023/03/02 23:35:29 - mmengine - INFO - Epoch(train) [68][30/32] lr: 1.0000e-06 eta: 0:31:18 time: 0.4122 data_time: 0.0011 memory: 18246 loss: 0.1150 loss_ce: 0.1150 2023/03/02 23:35:30 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:35:36 - mmengine - INFO - Epoch(train) [69][10/32] lr: 1.0000e-06 eta: 0:31:15 time: 0.5867 data_time: 0.0501 memory: 21276 loss: 0.1014 loss_ce: 0.1014 2023/03/02 23:35:40 - mmengine - INFO - Epoch(train) [69][20/32] lr: 1.0000e-06 eta: 0:31:11 time: 0.4950 data_time: 0.0012 memory: 23088 loss: 0.1173 loss_ce: 0.1173 2023/03/02 23:35:45 - mmengine - INFO - Epoch(train) [69][30/32] lr: 1.0000e-06 eta: 0:31:07 time: 0.4395 data_time: 0.0011 memory: 18966 loss: 0.1093 loss_ce: 0.1093 2023/03/02 23:35:46 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:35:51 - mmengine - INFO - Epoch(train) [70][10/32] lr: 1.0000e-06 eta: 0:31:03 time: 0.5644 data_time: 0.0550 memory: 18940 loss: 0.1187 loss_ce: 0.1187 2023/03/02 23:35:56 - mmengine - INFO - Epoch(train) [70][20/32] lr: 1.0000e-06 eta: 0:30:59 time: 0.4468 data_time: 0.0014 memory: 19958 loss: 0.1277 loss_ce: 0.1277 2023/03/02 23:36:00 - mmengine - INFO - Epoch(train) [70][30/32] lr: 1.0000e-06 eta: 0:30:55 time: 0.4501 data_time: 0.0012 memory: 18349 loss: 0.0997 loss_ce: 0.0997 2023/03/02 23:36:01 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:36:17 - mmengine - INFO - Epoch(val) [70][10/63] eta: 0:01:22 time: 1.5503 data_time: 0.0046 memory: 16988 2023/03/02 23:36:42 - mmengine - INFO - Epoch(val) [70][20/63] eta: 0:01:27 time: 2.5087 data_time: 0.0004 memory: 1075 2023/03/02 23:37:10 - mmengine - INFO - Epoch(val) [70][30/63] eta: 0:01:15 time: 2.7885 data_time: 0.0004 memory: 1075 2023/03/02 23:37:20 - mmengine - INFO - Epoch(val) [70][40/63] eta: 0:00:45 time: 1.0239 data_time: 0.0004 memory: 1075 2023/03/02 23:37:43 - mmengine - INFO - Epoch(val) [70][50/63] eta: 0:00:26 time: 2.2857 data_time: 0.0004 memory: 1075 2023/03/02 23:37:56 - mmengine - INFO - Epoch(val) [70][60/63] eta: 0:00:05 time: 1.3222 data_time: 0.0004 memory: 1075 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.80, recall: 0.6095, precision: 0.7599, hmean: 0.6765 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.81, recall: 0.6071, precision: 0.7638, hmean: 0.6765 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.82, recall: 0.6047, precision: 0.7710, hmean: 0.6778 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.83, recall: 0.6023, precision: 0.7761, hmean: 0.6782 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.84, recall: 0.5985, precision: 0.7832, hmean: 0.6785 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.85, recall: 0.5965, precision: 0.7897, hmean: 0.6796 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.86, recall: 0.5946, precision: 0.7983, hmean: 0.6816 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.87, recall: 0.5912, precision: 0.8042, hmean: 0.6815 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.88, recall: 0.5874, precision: 0.8144, hmean: 0.6825 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.89, recall: 0.5787, precision: 0.8199, hmean: 0.6785 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.90, recall: 0.5739, precision: 0.8289, hmean: 0.6782 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.91, recall: 0.5672, precision: 0.8378, hmean: 0.6764 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.92, recall: 0.5595, precision: 0.8451, hmean: 0.6732 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.93, recall: 0.5513, precision: 0.8519, hmean: 0.6694 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.94, recall: 0.5412, precision: 0.8626, hmean: 0.6651 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.95, recall: 0.5262, precision: 0.8723, hmean: 0.6565 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.96, recall: 0.5099, precision: 0.8796, hmean: 0.6455 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.97, recall: 0.4887, precision: 0.8849, hmean: 0.6297 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.98, recall: 0.4622, precision: 0.8964, hmean: 0.6099 2023/03/02 23:38:56 - mmengine - INFO - text score threshold: 0.99, recall: 0.4218, precision: 0.9163, hmean: 0.5776 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.80, recall: 0.6586, precision: 0.8211, hmean: 0.7310 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.81, recall: 0.6558, precision: 0.8250, hmean: 0.7307 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.82, recall: 0.6524, precision: 0.8318, hmean: 0.7312 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.83, recall: 0.6495, precision: 0.8368, hmean: 0.7314 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.84, recall: 0.6437, precision: 0.8425, hmean: 0.7298 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.85, recall: 0.6389, precision: 0.8458, hmean: 0.7279 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.86, recall: 0.6346, precision: 0.8520, hmean: 0.7274 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.87, recall: 0.6293, precision: 0.8559, hmean: 0.7253 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.88, recall: 0.6230, precision: 0.8638, hmean: 0.7239 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.89, recall: 0.6129, precision: 0.8683, hmean: 0.7186 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.90, recall: 0.6066, precision: 0.8762, hmean: 0.7169 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.91, recall: 0.5970, precision: 0.8819, hmean: 0.7120 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.92, recall: 0.5874, precision: 0.8873, hmean: 0.7068 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.93, recall: 0.5773, precision: 0.8921, hmean: 0.7010 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.94, recall: 0.5623, precision: 0.8964, hmean: 0.6911 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.95, recall: 0.5436, precision: 0.9010, hmean: 0.6781 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.96, recall: 0.5243, precision: 0.9045, hmean: 0.6638 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.97, recall: 0.5017, precision: 0.9085, hmean: 0.6464 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.98, recall: 0.4718, precision: 0.9150, hmean: 0.6226 2023/03/02 23:39:05 - mmengine - INFO - text score threshold: 0.99, recall: 0.4275, precision: 0.9289, hmean: 0.5856 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.80, recall: 0.7097, precision: 0.8848, hmean: 0.7876 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.81, recall: 0.7058, precision: 0.8879, hmean: 0.7865 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.82, recall: 0.7015, precision: 0.8944, hmean: 0.7863 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.83, recall: 0.6976, precision: 0.8989, hmean: 0.7856 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.84, recall: 0.6914, precision: 0.9049, hmean: 0.7838 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.85, recall: 0.6846, precision: 0.9063, hmean: 0.7800 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.86, recall: 0.6793, precision: 0.9121, hmean: 0.7787 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.87, recall: 0.6716, precision: 0.9136, hmean: 0.7741 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.88, recall: 0.6625, precision: 0.9186, hmean: 0.7698 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.89, recall: 0.6490, precision: 0.9195, hmean: 0.7609 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.90, recall: 0.6413, precision: 0.9263, hmean: 0.7579 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.91, recall: 0.6298, precision: 0.9303, hmean: 0.7511 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.92, recall: 0.6187, precision: 0.9345, hmean: 0.7445 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.93, recall: 0.6066, precision: 0.9375, hmean: 0.7366 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.94, recall: 0.5908, precision: 0.9417, hmean: 0.7260 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.95, recall: 0.5696, precision: 0.9441, hmean: 0.7105 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.96, recall: 0.5498, precision: 0.9485, hmean: 0.6961 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.97, recall: 0.5248, precision: 0.9503, hmean: 0.6762 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.98, recall: 0.4935, precision: 0.9570, hmean: 0.6512 2023/03/02 23:39:14 - mmengine - INFO - text score threshold: 0.99, recall: 0.4444, precision: 0.9655, hmean: 0.6086 2023/03/02 23:39:14 - mmengine - INFO - Epoch(val) [70][63/63] generic/precision: 0.8144 generic/recall: 0.5874 generic/hmean: 0.6825 weak/precision: 0.8368 weak/recall: 0.6495 weak/hmean: 0.7314 strong/precision: 0.8848 strong/recall: 0.7097 strong/hmean: 0.7876 2023/03/02 23:39:17 - mmengine - INFO - Epoch(train) [71][10/32] lr: 1.0000e-06 eta: 0:30:48 time: 0.3809 data_time: 0.0448 memory: 21202 loss: 0.1006 loss_ce: 0.1006 2023/03/02 23:39:20 - mmengine - INFO - Epoch(train) [71][20/32] lr: 1.0000e-06 eta: 0:30:41 time: 0.2949 data_time: 0.0012 memory: 20832 loss: 0.1071 loss_ce: 0.1071 2023/03/02 23:39:23 - mmengine - INFO - Epoch(train) [71][30/32] lr: 1.0000e-06 eta: 0:30:33 time: 0.3000 data_time: 0.0010 memory: 18966 loss: 0.1171 loss_ce: 0.1171 2023/03/02 23:39:24 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:39:27 - mmengine - INFO - Epoch(train) [72][10/32] lr: 1.0000e-06 eta: 0:30:26 time: 0.3699 data_time: 0.0718 memory: 18790 loss: 0.1153 loss_ce: 0.1153 2023/03/02 23:39:30 - mmengine - INFO - Epoch(train) [72][20/32] lr: 1.0000e-06 eta: 0:30:19 time: 0.2909 data_time: 0.0012 memory: 18757 loss: 0.1221 loss_ce: 0.1221 2023/03/02 23:39:33 - mmengine - INFO - Epoch(train) [72][30/32] lr: 1.0000e-06 eta: 0:30:12 time: 0.2983 data_time: 0.0010 memory: 18535 loss: 0.1066 loss_ce: 0.1066 2023/03/02 23:39:34 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:39:38 - mmengine - INFO - Epoch(train) [73][10/32] lr: 1.0000e-06 eta: 0:30:05 time: 0.3817 data_time: 0.0437 memory: 20820 loss: 0.1054 loss_ce: 0.1054 2023/03/02 23:39:40 - mmengine - INFO - Epoch(train) [73][20/32] lr: 1.0000e-06 eta: 0:29:57 time: 0.2816 data_time: 0.0012 memory: 23431 loss: 0.1183 loss_ce: 0.1183 2023/03/02 23:39:43 - mmengine - INFO - Epoch(train) [73][30/32] lr: 1.0000e-06 eta: 0:29:50 time: 0.2927 data_time: 0.0010 memory: 19382 loss: 0.1231 loss_ce: 0.1231 2023/03/02 23:39:44 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:39:48 - mmengine - INFO - Epoch(train) [74][10/32] lr: 1.0000e-06 eta: 0:29:44 time: 0.3921 data_time: 0.0616 memory: 19751 loss: 0.1079 loss_ce: 0.1079 2023/03/02 23:39:51 - mmengine - INFO - Epoch(train) [74][20/32] lr: 1.0000e-06 eta: 0:29:37 time: 0.2903 data_time: 0.0012 memory: 21980 loss: 0.1142 loss_ce: 0.1142 2023/03/02 23:39:53 - mmengine - INFO - Epoch(train) [74][30/32] lr: 1.0000e-06 eta: 0:29:30 time: 0.2842 data_time: 0.0010 memory: 20382 loss: 0.1048 loss_ce: 0.1048 2023/03/02 23:39:54 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:39:58 - mmengine - INFO - Epoch(train) [75][10/32] lr: 1.0000e-06 eta: 0:29:22 time: 0.3609 data_time: 0.0674 memory: 19958 loss: 0.1044 loss_ce: 0.1044 2023/03/02 23:40:00 - mmengine - INFO - Epoch(train) [75][20/32] lr: 1.0000e-06 eta: 0:29:15 time: 0.2912 data_time: 0.0012 memory: 21742 loss: 0.1120 loss_ce: 0.1120 2023/03/02 23:40:03 - mmengine - INFO - Epoch(train) [75][30/32] lr: 1.0000e-06 eta: 0:29:08 time: 0.2852 data_time: 0.0010 memory: 18083 loss: 0.1089 loss_ce: 0.1089 2023/03/02 23:40:04 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:40:08 - mmengine - INFO - Epoch(train) [76][10/32] lr: 1.0000e-06 eta: 0:29:02 time: 0.3862 data_time: 0.0906 memory: 19751 loss: 0.0909 loss_ce: 0.0909 2023/03/02 23:40:10 - mmengine - INFO - Epoch(train) [76][20/32] lr: 1.0000e-06 eta: 0:28:55 time: 0.2796 data_time: 0.0013 memory: 19448 loss: 0.1123 loss_ce: 0.1123 2023/03/02 23:40:13 - mmengine - INFO - Epoch(train) [76][30/32] lr: 1.0000e-06 eta: 0:28:48 time: 0.2759 data_time: 0.0010 memory: 22104 loss: 0.0921 loss_ce: 0.0921 2023/03/02 23:40:14 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:40:17 - mmengine - INFO - Epoch(train) [77][10/32] lr: 1.0000e-06 eta: 0:28:41 time: 0.3884 data_time: 0.1088 memory: 17919 loss: 0.1201 loss_ce: 0.1201 2023/03/02 23:40:20 - mmengine - INFO - Epoch(train) [77][20/32] lr: 1.0000e-06 eta: 0:28:34 time: 0.2976 data_time: 0.0012 memory: 18597 loss: 0.1025 loss_ce: 0.1025 2023/03/02 23:40:23 - mmengine - INFO - Epoch(train) [77][30/32] lr: 1.0000e-06 eta: 0:28:28 time: 0.2960 data_time: 0.0010 memory: 19751 loss: 0.1041 loss_ce: 0.1041 2023/03/02 23:40:24 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:40:27 - mmengine - INFO - Epoch(train) [78][10/32] lr: 1.0000e-06 eta: 0:28:21 time: 0.3516 data_time: 0.0508 memory: 18746 loss: 0.1044 loss_ce: 0.1044 2023/03/02 23:40:30 - mmengine - INFO - Epoch(train) [78][20/32] lr: 1.0000e-06 eta: 0:28:14 time: 0.2993 data_time: 0.0013 memory: 23431 loss: 0.0990 loss_ce: 0.0990 2023/03/02 23:40:33 - mmengine - INFO - Epoch(train) [78][30/32] lr: 1.0000e-06 eta: 0:28:07 time: 0.2697 data_time: 0.0011 memory: 19958 loss: 0.1077 loss_ce: 0.1077 2023/03/02 23:40:34 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:40:37 - mmengine - INFO - Epoch(train) [79][10/32] lr: 1.0000e-06 eta: 0:28:01 time: 0.3765 data_time: 0.0865 memory: 22984 loss: 0.1083 loss_ce: 0.1083 2023/03/02 23:40:40 - mmengine - INFO - Epoch(train) [79][20/32] lr: 1.0000e-06 eta: 0:27:54 time: 0.2919 data_time: 0.0012 memory: 18790 loss: 0.1083 loss_ce: 0.1083 2023/03/02 23:40:43 - mmengine - INFO - Epoch(train) [79][30/32] lr: 1.0000e-06 eta: 0:27:48 time: 0.2816 data_time: 0.0010 memory: 18597 loss: 0.1040 loss_ce: 0.1040 2023/03/02 23:40:43 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:40:47 - mmengine - INFO - Epoch(train) [80][10/32] lr: 1.0000e-06 eta: 0:27:41 time: 0.3793 data_time: 0.0500 memory: 19958 loss: 0.1096 loss_ce: 0.1096 2023/03/02 23:40:50 - mmengine - INFO - Epoch(train) [80][20/32] lr: 1.0000e-06 eta: 0:27:35 time: 0.2949 data_time: 0.0013 memory: 19751 loss: 0.1045 loss_ce: 0.1045 2023/03/02 23:40:53 - mmengine - INFO - Epoch(train) [80][30/32] lr: 1.0000e-06 eta: 0:27:28 time: 0.2884 data_time: 0.0010 memory: 18940 loss: 0.1146 loss_ce: 0.1146 2023/03/02 23:40:53 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:41:01 - mmengine - INFO - Epoch(val) [80][10/63] eta: 0:00:38 time: 0.7346 data_time: 0.0031 memory: 15914 2023/03/02 23:41:14 - mmengine - INFO - Epoch(val) [80][20/63] eta: 0:00:45 time: 1.3608 data_time: 0.0003 memory: 1075 2023/03/02 23:41:27 - mmengine - INFO - Epoch(val) [80][30/63] eta: 0:00:36 time: 1.2231 data_time: 0.0003 memory: 1075 2023/03/02 23:41:32 - mmengine - INFO - Epoch(val) [80][40/63] eta: 0:00:22 time: 0.5755 data_time: 0.0003 memory: 1075 2023/03/02 23:41:45 - mmengine - INFO - Epoch(val) [80][50/63] eta: 0:00:13 time: 1.2169 data_time: 0.0003 memory: 1075 2023/03/02 23:41:52 - mmengine - INFO - Epoch(val) [80][60/63] eta: 0:00:02 time: 0.7529 data_time: 0.0002 memory: 1075 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.80, recall: 0.6134, precision: 0.7543, hmean: 0.6766 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.81, recall: 0.6119, precision: 0.7579, hmean: 0.6771 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.82, recall: 0.6100, precision: 0.7642, hmean: 0.6784 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.83, recall: 0.6066, precision: 0.7706, hmean: 0.6789 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.84, recall: 0.6038, precision: 0.7794, hmean: 0.6804 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.85, recall: 0.5999, precision: 0.7846, hmean: 0.6799 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.86, recall: 0.5980, precision: 0.7926, hmean: 0.6817 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.87, recall: 0.5951, precision: 0.7974, hmean: 0.6816 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.88, recall: 0.5898, precision: 0.8065, hmean: 0.6813 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.89, recall: 0.5864, precision: 0.8147, hmean: 0.6820 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.90, recall: 0.5773, precision: 0.8212, hmean: 0.6780 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.91, recall: 0.5720, precision: 0.8331, hmean: 0.6783 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.92, recall: 0.5652, precision: 0.8398, hmean: 0.6757 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.93, recall: 0.5542, precision: 0.8501, hmean: 0.6709 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.94, recall: 0.5455, precision: 0.8629, hmean: 0.6684 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.95, recall: 0.5354, precision: 0.8722, hmean: 0.6635 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.96, recall: 0.5195, precision: 0.8787, hmean: 0.6530 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.97, recall: 0.4959, precision: 0.8864, hmean: 0.6360 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.98, recall: 0.4752, precision: 0.8948, hmean: 0.6208 2023/03/02 23:42:54 - mmengine - INFO - text score threshold: 0.99, recall: 0.4314, precision: 0.9134, hmean: 0.5860 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.80, recall: 0.6625, precision: 0.8147, hmean: 0.7307 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.81, recall: 0.6601, precision: 0.8175, hmean: 0.7304 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.82, recall: 0.6577, precision: 0.8239, hmean: 0.7315 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.83, recall: 0.6533, precision: 0.8300, hmean: 0.7311 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.84, recall: 0.6495, precision: 0.8384, hmean: 0.7320 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.85, recall: 0.6437, precision: 0.8419, hmean: 0.7296 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.86, recall: 0.6399, precision: 0.8481, hmean: 0.7294 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.87, recall: 0.6355, precision: 0.8516, hmean: 0.7279 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.88, recall: 0.6283, precision: 0.8591, hmean: 0.7258 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.89, recall: 0.6240, precision: 0.8669, hmean: 0.7256 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.90, recall: 0.6115, precision: 0.8699, hmean: 0.7181 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.91, recall: 0.6028, precision: 0.8780, hmean: 0.7148 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.92, recall: 0.5946, precision: 0.8834, hmean: 0.7108 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.93, recall: 0.5792, precision: 0.8885, hmean: 0.7013 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.94, recall: 0.5662, precision: 0.8957, hmean: 0.6938 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.95, recall: 0.5522, precision: 0.8996, hmean: 0.6844 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.96, recall: 0.5344, precision: 0.9039, hmean: 0.6717 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.97, recall: 0.5079, precision: 0.9079, hmean: 0.6514 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.98, recall: 0.4848, precision: 0.9130, hmean: 0.6333 2023/03/02 23:43:04 - mmengine - INFO - text score threshold: 0.99, recall: 0.4381, precision: 0.9276, hmean: 0.5952 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.80, recall: 0.7193, precision: 0.8845, hmean: 0.7934 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.81, recall: 0.7169, precision: 0.8879, hmean: 0.7933 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.82, recall: 0.7106, precision: 0.8902, hmean: 0.7904 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.83, recall: 0.7044, precision: 0.8948, hmean: 0.7883 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.84, recall: 0.6996, precision: 0.9030, hmean: 0.7884 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.85, recall: 0.6919, precision: 0.9049, hmean: 0.7842 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.86, recall: 0.6861, precision: 0.9094, hmean: 0.7821 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.87, recall: 0.6813, precision: 0.9129, hmean: 0.7803 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.88, recall: 0.6712, precision: 0.9177, hmean: 0.7753 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.89, recall: 0.6635, precision: 0.9217, hmean: 0.7716 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.90, recall: 0.6490, precision: 0.9233, hmean: 0.7622 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.91, recall: 0.6370, precision: 0.9278, hmean: 0.7554 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.92, recall: 0.6269, precision: 0.9313, hmean: 0.7494 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.93, recall: 0.6095, precision: 0.9350, hmean: 0.7380 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.94, recall: 0.5936, precision: 0.9391, hmean: 0.7274 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.95, recall: 0.5787, precision: 0.9427, hmean: 0.7172 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.96, recall: 0.5590, precision: 0.9454, hmean: 0.7026 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.97, recall: 0.5306, precision: 0.9484, hmean: 0.6805 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.98, recall: 0.5046, precision: 0.9501, hmean: 0.6591 2023/03/02 23:43:12 - mmengine - INFO - text score threshold: 0.99, recall: 0.4545, precision: 0.9623, hmean: 0.6174 2023/03/02 23:43:12 - mmengine - INFO - Epoch(val) [80][63/63] generic/precision: 0.8147 generic/recall: 0.5864 generic/hmean: 0.6820 weak/precision: 0.8384 weak/recall: 0.6495 weak/hmean: 0.7320 strong/precision: 0.8845 strong/recall: 0.7193 strong/hmean: 0.7934 2023/03/02 23:43:16 - mmengine - INFO - Epoch(train) [81][10/32] lr: 1.0000e-06 eta: 0:27:22 time: 0.3788 data_time: 0.1027 memory: 18246 loss: 0.1141 loss_ce: 0.1141 2023/03/02 23:43:19 - mmengine - INFO - Epoch(train) [81][20/32] lr: 1.0000e-06 eta: 0:27:15 time: 0.2943 data_time: 0.0012 memory: 18597 loss: 0.1174 loss_ce: 0.1174 2023/03/02 23:43:22 - mmengine - INFO - Epoch(train) [81][30/32] lr: 1.0000e-06 eta: 0:27:09 time: 0.2733 data_time: 0.0010 memory: 20820 loss: 0.1061 loss_ce: 0.1061 2023/03/02 23:43:22 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:43:26 - mmengine - INFO - Epoch(train) [82][10/32] lr: 1.0000e-06 eta: 0:27:02 time: 0.3615 data_time: 0.0711 memory: 18597 loss: 0.1117 loss_ce: 0.1117 2023/03/02 23:43:28 - mmengine - INFO - Epoch(train) [82][20/32] lr: 1.0000e-06 eta: 0:26:56 time: 0.2748 data_time: 0.0012 memory: 18128 loss: 0.1098 loss_ce: 0.1098 2023/03/02 23:43:31 - mmengine - INFO - Epoch(train) [82][30/32] lr: 1.0000e-06 eta: 0:26:49 time: 0.2822 data_time: 0.0010 memory: 19958 loss: 0.1025 loss_ce: 0.1025 2023/03/02 23:43:32 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:43:36 - mmengine - INFO - Epoch(train) [83][10/32] lr: 1.0000e-06 eta: 0:26:43 time: 0.4045 data_time: 0.0872 memory: 18597 loss: 0.0999 loss_ce: 0.0999 2023/03/02 23:43:39 - mmengine - INFO - Epoch(train) [83][20/32] lr: 1.0000e-06 eta: 0:26:38 time: 0.3184 data_time: 0.0012 memory: 19875 loss: 0.1096 loss_ce: 0.1096 2023/03/02 23:43:43 - mmengine - INFO - Epoch(train) [83][30/32] lr: 1.0000e-06 eta: 0:26:33 time: 0.4121 data_time: 0.0010 memory: 26338 loss: 0.1207 loss_ce: 0.1207 2023/03/02 23:43:44 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:43:49 - mmengine - INFO - Epoch(train) [84][10/32] lr: 1.0000e-06 eta: 0:26:29 time: 0.5277 data_time: 0.0582 memory: 18967 loss: 0.1138 loss_ce: 0.1138 2023/03/02 23:43:53 - mmengine - INFO - Epoch(train) [84][20/32] lr: 1.0000e-06 eta: 0:26:25 time: 0.4262 data_time: 0.0016 memory: 19751 loss: 0.1063 loss_ce: 0.1063 2023/03/02 23:43:58 - mmengine - INFO - Epoch(train) [84][30/32] lr: 1.0000e-06 eta: 0:26:21 time: 0.4487 data_time: 0.0012 memory: 24319 loss: 0.1095 loss_ce: 0.1095 2023/03/02 23:43:58 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:44:04 - mmengine - INFO - Epoch(train) [85][10/32] lr: 1.0000e-06 eta: 0:26:17 time: 0.5749 data_time: 0.0475 memory: 19692 loss: 0.0990 loss_ce: 0.0990 2023/03/02 23:44:09 - mmengine - INFO - Epoch(train) [85][20/32] lr: 1.0000e-06 eta: 0:26:14 time: 0.4735 data_time: 0.0014 memory: 19115 loss: 0.0888 loss_ce: 0.0888 2023/03/02 23:44:13 - mmengine - INFO - Epoch(train) [85][30/32] lr: 1.0000e-06 eta: 0:26:10 time: 0.4267 data_time: 0.0010 memory: 17656 loss: 0.1226 loss_ce: 0.1226 2023/03/02 23:44:14 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:44:19 - mmengine - INFO - Epoch(train) [86][10/32] lr: 1.0000e-06 eta: 0:26:06 time: 0.5771 data_time: 0.0972 memory: 24319 loss: 0.0957 loss_ce: 0.0957 2023/03/02 23:44:24 - mmengine - INFO - Epoch(train) [86][20/32] lr: 1.0000e-06 eta: 0:26:02 time: 0.4540 data_time: 0.0012 memory: 19346 loss: 0.1084 loss_ce: 0.1084 2023/03/02 23:44:28 - mmengine - INFO - Epoch(train) [86][30/32] lr: 1.0000e-06 eta: 0:25:58 time: 0.4521 data_time: 0.0012 memory: 17733 loss: 0.1088 loss_ce: 0.1088 2023/03/02 23:44:29 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:44:35 - mmengine - INFO - Epoch(train) [87][10/32] lr: 1.0000e-06 eta: 0:25:54 time: 0.5623 data_time: 0.0662 memory: 18083 loss: 0.1093 loss_ce: 0.1093 2023/03/02 23:44:39 - mmengine - INFO - Epoch(train) [87][20/32] lr: 1.0000e-06 eta: 0:25:50 time: 0.4447 data_time: 0.0014 memory: 19150 loss: 0.1152 loss_ce: 0.1152 2023/03/02 23:44:44 - mmengine - INFO - Epoch(train) [87][30/32] lr: 1.0000e-06 eta: 0:25:46 time: 0.4513 data_time: 0.0012 memory: 24319 loss: 0.1144 loss_ce: 0.1144 2023/03/02 23:44:44 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:44:50 - mmengine - INFO - Epoch(train) [88][10/32] lr: 1.0000e-06 eta: 0:25:43 time: 0.5324 data_time: 0.0577 memory: 19958 loss: 0.1005 loss_ce: 0.1005 2023/03/02 23:44:54 - mmengine - INFO - Epoch(train) [88][20/32] lr: 1.0000e-06 eta: 0:25:39 time: 0.4599 data_time: 0.0014 memory: 19142 loss: 0.1156 loss_ce: 0.1156 2023/03/02 23:44:59 - mmengine - INFO - Epoch(train) [88][30/32] lr: 1.0000e-06 eta: 0:25:35 time: 0.4810 data_time: 0.0011 memory: 25747 loss: 0.0918 loss_ce: 0.0918 2023/03/02 23:45:00 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:45:05 - mmengine - INFO - Epoch(train) [89][10/32] lr: 1.0000e-06 eta: 0:25:31 time: 0.5670 data_time: 0.0956 memory: 19142 loss: 0.1049 loss_ce: 0.1049 2023/03/02 23:45:10 - mmengine - INFO - Epoch(train) [89][20/32] lr: 1.0000e-06 eta: 0:25:27 time: 0.4254 data_time: 0.0013 memory: 17583 loss: 0.1187 loss_ce: 0.1187 2023/03/02 23:45:14 - mmengine - INFO - Epoch(train) [89][30/32] lr: 1.0000e-06 eta: 0:25:23 time: 0.4695 data_time: 0.0011 memory: 20655 loss: 0.1195 loss_ce: 0.1195 2023/03/02 23:45:15 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:45:21 - mmengine - INFO - Epoch(train) [90][10/32] lr: 1.0000e-06 eta: 0:25:19 time: 0.5551 data_time: 0.0801 memory: 18550 loss: 0.1100 loss_ce: 0.1100 2023/03/02 23:45:25 - mmengine - INFO - Epoch(train) [90][20/32] lr: 1.0000e-06 eta: 0:25:15 time: 0.4456 data_time: 0.0013 memory: 17733 loss: 0.1208 loss_ce: 0.1208 2023/03/02 23:45:30 - mmengine - INFO - Epoch(train) [90][30/32] lr: 1.0000e-06 eta: 0:25:12 time: 0.4711 data_time: 0.0011 memory: 19142 loss: 0.1055 loss_ce: 0.1055 2023/03/02 23:45:30 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:45:43 - mmengine - INFO - Epoch(val) [90][10/63] eta: 0:01:07 time: 1.2692 data_time: 0.0040 memory: 17137 2023/03/02 23:46:06 - mmengine - INFO - Epoch(val) [90][20/63] eta: 0:01:17 time: 2.3358 data_time: 0.0004 memory: 1075 2023/03/02 23:46:29 - mmengine - INFO - Epoch(val) [90][30/63] eta: 0:01:04 time: 2.2756 data_time: 0.0003 memory: 1075 2023/03/02 23:46:40 - mmengine - INFO - Epoch(val) [90][40/63] eta: 0:00:39 time: 1.0559 data_time: 0.0003 memory: 1075 2023/03/02 23:47:00 - mmengine - INFO - Epoch(val) [90][50/63] eta: 0:00:23 time: 2.0939 data_time: 0.0003 memory: 1075 2023/03/02 23:47:13 - mmengine - INFO - Epoch(val) [90][60/63] eta: 0:00:05 time: 1.2485 data_time: 0.0003 memory: 1075 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.80, recall: 0.6139, precision: 0.7544, hmean: 0.6769 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.81, recall: 0.6115, precision: 0.7600, hmean: 0.6777 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.82, recall: 0.6100, precision: 0.7628, hmean: 0.6779 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.83, recall: 0.6081, precision: 0.7701, hmean: 0.6796 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.84, recall: 0.6052, precision: 0.7764, hmean: 0.6802 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.85, recall: 0.6028, precision: 0.7864, hmean: 0.6825 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.86, recall: 0.6009, precision: 0.7944, hmean: 0.6842 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.87, recall: 0.5970, precision: 0.8005, hmean: 0.6839 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.88, recall: 0.5912, precision: 0.8100, hmean: 0.6836 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.89, recall: 0.5874, precision: 0.8210, hmean: 0.6848 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.90, recall: 0.5806, precision: 0.8272, hmean: 0.6823 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.91, recall: 0.5729, precision: 0.8345, hmean: 0.6794 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.92, recall: 0.5667, precision: 0.8425, hmean: 0.6776 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.93, recall: 0.5542, precision: 0.8513, hmean: 0.6713 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.94, recall: 0.5436, precision: 0.8605, hmean: 0.6663 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.95, recall: 0.5315, precision: 0.8734, hmean: 0.6609 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.96, recall: 0.5166, precision: 0.8781, hmean: 0.6505 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.97, recall: 0.5012, precision: 0.8852, hmean: 0.6400 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.98, recall: 0.4709, precision: 0.8940, hmean: 0.6168 2023/03/02 23:48:12 - mmengine - INFO - text score threshold: 0.99, recall: 0.4357, precision: 0.9095, hmean: 0.5892 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.80, recall: 0.6615, precision: 0.8130, hmean: 0.7295 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.81, recall: 0.6586, precision: 0.8187, hmean: 0.7300 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.82, recall: 0.6562, precision: 0.8206, hmean: 0.7293 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.83, recall: 0.6529, precision: 0.8268, hmean: 0.7296 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.84, recall: 0.6481, precision: 0.8314, hmean: 0.7284 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.85, recall: 0.6432, precision: 0.8392, hmean: 0.7283 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.86, recall: 0.6394, precision: 0.8453, hmean: 0.7281 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.87, recall: 0.6346, precision: 0.8509, hmean: 0.7270 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.88, recall: 0.6254, precision: 0.8569, hmean: 0.7231 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.89, recall: 0.6182, precision: 0.8641, hmean: 0.7207 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.90, recall: 0.6115, precision: 0.8711, hmean: 0.7185 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.91, recall: 0.6028, precision: 0.8780, hmean: 0.7148 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.92, recall: 0.5946, precision: 0.8840, hmean: 0.7110 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.93, recall: 0.5811, precision: 0.8928, hmean: 0.7040 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.94, recall: 0.5657, precision: 0.8956, hmean: 0.6934 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.95, recall: 0.5489, precision: 0.9019, hmean: 0.6824 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.96, recall: 0.5320, precision: 0.9043, hmean: 0.6699 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.97, recall: 0.5142, precision: 0.9082, hmean: 0.6566 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.98, recall: 0.4800, precision: 0.9113, hmean: 0.6288 2023/03/02 23:48:22 - mmengine - INFO - text score threshold: 0.99, recall: 0.4425, precision: 0.9236, hmean: 0.5983 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.80, recall: 0.7164, precision: 0.8805, hmean: 0.7900 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.81, recall: 0.7111, precision: 0.8839, hmean: 0.7882 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.82, recall: 0.7087, precision: 0.8862, hmean: 0.7876 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.83, recall: 0.7044, precision: 0.8921, hmean: 0.7872 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.84, recall: 0.6986, precision: 0.8962, hmean: 0.7852 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.85, recall: 0.6919, precision: 0.9026, hmean: 0.7833 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.86, recall: 0.6861, precision: 0.9071, hmean: 0.7812 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.87, recall: 0.6803, precision: 0.9122, hmean: 0.7794 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.88, recall: 0.6678, precision: 0.9149, hmean: 0.7721 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.89, recall: 0.6572, precision: 0.9186, hmean: 0.7662 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.90, recall: 0.6476, precision: 0.9225, hmean: 0.7610 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.91, recall: 0.6360, precision: 0.9264, hmean: 0.7542 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.92, recall: 0.6259, precision: 0.9306, hmean: 0.7484 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.93, recall: 0.6086, precision: 0.9349, hmean: 0.7372 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.94, recall: 0.5922, precision: 0.9375, hmean: 0.7259 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.95, recall: 0.5729, precision: 0.9415, hmean: 0.7124 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.96, recall: 0.5556, precision: 0.9444, hmean: 0.6996 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.97, recall: 0.5359, precision: 0.9464, hmean: 0.6843 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.98, recall: 0.4998, precision: 0.9488, hmean: 0.6547 2023/03/02 23:48:30 - mmengine - INFO - text score threshold: 0.99, recall: 0.4588, precision: 0.9578, hmean: 0.6204 2023/03/02 23:48:30 - mmengine - INFO - Epoch(val) [90][63/63] generic/precision: 0.8210 generic/recall: 0.5874 generic/hmean: 0.6848 weak/precision: 0.8187 weak/recall: 0.6586 weak/hmean: 0.7300 strong/precision: 0.8805 strong/recall: 0.7164 strong/hmean: 0.7900 2023/03/02 23:48:30 - mmengine - INFO - The previous best checkpoint mmocr/projects/SPTS/work_dirs/spts_resnet50_350e_icdar2015/best_generic/hmean_epoch_60.pth is removed 2023/03/02 23:48:33 - mmengine - INFO - The best checkpoint with 0.6848 generic/hmean at 90 epoch is saved to best_generic/hmean_epoch_90.pth. 2023/03/02 23:48:38 - mmengine - INFO - Epoch(train) [91][10/32] lr: 1.0000e-06 eta: 0:25:07 time: 0.4845 data_time: 0.0376 memory: 19552 loss: 0.1021 loss_ce: 0.1021 2023/03/02 23:48:42 - mmengine - INFO - Epoch(train) [91][20/32] lr: 1.0000e-06 eta: 0:25:03 time: 0.4471 data_time: 0.0014 memory: 18779 loss: 0.1084 loss_ce: 0.1084 2023/03/02 23:48:47 - mmengine - INFO - Epoch(train) [91][30/32] lr: 1.0000e-06 eta: 0:24:58 time: 0.4437 data_time: 0.0010 memory: 19958 loss: 0.1008 loss_ce: 0.1008 2023/03/02 23:48:47 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:48:52 - mmengine - INFO - Epoch(train) [92][10/32] lr: 1.0000e-06 eta: 0:24:54 time: 0.5147 data_time: 0.0467 memory: 18597 loss: 0.1186 loss_ce: 0.1186 2023/03/02 23:48:57 - mmengine - INFO - Epoch(train) [92][20/32] lr: 1.0000e-06 eta: 0:24:50 time: 0.4293 data_time: 0.0014 memory: 18083 loss: 0.1215 loss_ce: 0.1215 2023/03/02 23:49:01 - mmengine - INFO - Epoch(train) [92][30/32] lr: 1.0000e-06 eta: 0:24:46 time: 0.4648 data_time: 0.0012 memory: 21255 loss: 0.1126 loss_ce: 0.1126 2023/03/02 23:49:02 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:49:07 - mmengine - INFO - Epoch(train) [93][10/32] lr: 1.0000e-06 eta: 0:24:41 time: 0.5010 data_time: 0.0844 memory: 18413 loss: 0.1018 loss_ce: 0.1018 2023/03/02 23:49:11 - mmengine - INFO - Epoch(train) [93][20/32] lr: 1.0000e-06 eta: 0:24:37 time: 0.4350 data_time: 0.0015 memory: 17936 loss: 0.1313 loss_ce: 0.1313 2023/03/02 23:49:16 - mmengine - INFO - Epoch(train) [93][30/32] lr: 1.0000e-06 eta: 0:24:33 time: 0.4357 data_time: 0.0012 memory: 18083 loss: 0.1142 loss_ce: 0.1142 2023/03/02 23:49:16 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:49:22 - mmengine - INFO - Epoch(train) [94][10/32] lr: 1.0000e-06 eta: 0:24:29 time: 0.5300 data_time: 0.0543 memory: 20172 loss: 0.1062 loss_ce: 0.1062 2023/03/02 23:49:26 - mmengine - INFO - Epoch(train) [94][20/32] lr: 1.0000e-06 eta: 0:24:25 time: 0.4615 data_time: 0.0013 memory: 21932 loss: 0.1053 loss_ce: 0.1053 2023/03/02 23:49:28 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:49:31 - mmengine - INFO - Epoch(train) [94][30/32] lr: 1.0000e-06 eta: 0:24:20 time: 0.4295 data_time: 0.0010 memory: 18083 loss: 0.1198 loss_ce: 0.1198 2023/03/02 23:49:31 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:49:37 - mmengine - INFO - Epoch(train) [95][10/32] lr: 1.0000e-06 eta: 0:24:16 time: 0.5377 data_time: 0.0880 memory: 22984 loss: 0.1240 loss_ce: 0.1240 2023/03/02 23:49:41 - mmengine - INFO - Epoch(train) [95][20/32] lr: 1.0000e-06 eta: 0:24:12 time: 0.4393 data_time: 0.0012 memory: 17919 loss: 0.1140 loss_ce: 0.1140 2023/03/02 23:49:46 - mmengine - INFO - Epoch(train) [95][30/32] lr: 1.0000e-06 eta: 0:24:08 time: 0.4868 data_time: 0.0013 memory: 20820 loss: 0.0975 loss_ce: 0.0975 2023/03/02 23:49:46 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:49:51 - mmengine - INFO - Epoch(train) [96][10/32] lr: 1.0000e-06 eta: 0:24:03 time: 0.4995 data_time: 0.0732 memory: 20689 loss: 0.1112 loss_ce: 0.1112 2023/03/02 23:49:56 - mmengine - INFO - Epoch(train) [96][20/32] lr: 1.0000e-06 eta: 0:23:59 time: 0.4500 data_time: 0.0015 memory: 19958 loss: 0.1139 loss_ce: 0.1139 2023/03/02 23:50:00 - mmengine - INFO - Epoch(train) [96][30/32] lr: 1.0000e-06 eta: 0:23:55 time: 0.4406 data_time: 0.0013 memory: 19958 loss: 0.1327 loss_ce: 0.1327 2023/03/02 23:50:01 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:50:06 - mmengine - INFO - Epoch(train) [97][10/32] lr: 1.0000e-06 eta: 0:23:51 time: 0.5298 data_time: 0.0966 memory: 18083 loss: 0.1129 loss_ce: 0.1129 2023/03/02 23:50:11 - mmengine - INFO - Epoch(train) [97][20/32] lr: 1.0000e-06 eta: 0:23:46 time: 0.4352 data_time: 0.0014 memory: 19738 loss: 0.1189 loss_ce: 0.1189 2023/03/02 23:50:15 - mmengine - INFO - Epoch(train) [97][30/32] lr: 1.0000e-06 eta: 0:23:43 time: 0.4745 data_time: 0.0010 memory: 20204 loss: 0.1117 loss_ce: 0.1117 2023/03/02 23:50:16 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:50:21 - mmengine - INFO - Epoch(train) [98][10/32] lr: 1.0000e-06 eta: 0:23:38 time: 0.5274 data_time: 0.0762 memory: 20172 loss: 0.1055 loss_ce: 0.1055 2023/03/02 23:50:25 - mmengine - INFO - Epoch(train) [98][20/32] lr: 1.0000e-06 eta: 0:23:34 time: 0.4071 data_time: 0.0014 memory: 20889 loss: 0.1092 loss_ce: 0.1092 2023/03/02 23:50:30 - mmengine - INFO - Epoch(train) [98][30/32] lr: 1.0000e-06 eta: 0:23:30 time: 0.4379 data_time: 0.0010 memory: 18246 loss: 0.1161 loss_ce: 0.1161 2023/03/02 23:50:30 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:50:36 - mmengine - INFO - Epoch(train) [99][10/32] lr: 1.0000e-06 eta: 0:23:26 time: 0.5745 data_time: 0.1113 memory: 19061 loss: 0.0961 loss_ce: 0.0961 2023/03/02 23:50:41 - mmengine - INFO - Epoch(train) [99][20/32] lr: 1.0000e-06 eta: 0:23:21 time: 0.4348 data_time: 0.0012 memory: 20172 loss: 0.1072 loss_ce: 0.1072 2023/03/02 23:50:45 - mmengine - INFO - Epoch(train) [99][30/32] lr: 1.0000e-06 eta: 0:23:17 time: 0.4115 data_time: 0.0011 memory: 18083 loss: 0.1037 loss_ce: 0.1037 2023/03/02 23:50:45 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:50:51 - mmengine - INFO - Epoch(train) [100][10/32] lr: 1.0000e-06 eta: 0:23:13 time: 0.5466 data_time: 0.0720 memory: 24319 loss: 0.1105 loss_ce: 0.1105 2023/03/02 23:50:55 - mmengine - INFO - Epoch(train) [100][20/32] lr: 1.0000e-06 eta: 0:23:08 time: 0.4453 data_time: 0.0013 memory: 17137 loss: 0.1060 loss_ce: 0.1060 2023/03/02 23:51:00 - mmengine - INFO - Epoch(train) [100][30/32] lr: 1.0000e-06 eta: 0:23:05 time: 0.4942 data_time: 0.0012 memory: 19346 loss: 0.1038 loss_ce: 0.1038 2023/03/02 23:51:01 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:51:15 - mmengine - INFO - Epoch(val) [100][10/63] eta: 0:01:14 time: 1.4022 data_time: 0.0027 memory: 13878 2023/03/02 23:51:38 - mmengine - INFO - Epoch(val) [100][20/63] eta: 0:01:19 time: 2.2884 data_time: 0.0003 memory: 1075 2023/03/02 23:52:01 - mmengine - INFO - Epoch(val) [100][30/63] eta: 0:01:06 time: 2.3266 data_time: 0.0003 memory: 1075 2023/03/02 23:52:11 - mmengine - INFO - Epoch(val) [100][40/63] eta: 0:00:40 time: 1.0391 data_time: 0.0004 memory: 1075 2023/03/02 23:52:33 - mmengine - INFO - Epoch(val) [100][50/63] eta: 0:00:23 time: 2.1165 data_time: 0.0004 memory: 1075 2023/03/02 23:52:46 - mmengine - INFO - Epoch(val) [100][60/63] eta: 0:00:05 time: 1.3131 data_time: 0.0003 memory: 1075 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.80, recall: 0.6091, precision: 0.7472, hmean: 0.6711 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.81, recall: 0.6076, precision: 0.7534, hmean: 0.6727 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.82, recall: 0.6057, precision: 0.7601, hmean: 0.6742 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.83, recall: 0.6047, precision: 0.7682, hmean: 0.6767 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.84, recall: 0.5994, precision: 0.7747, hmean: 0.6759 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.85, recall: 0.5994, precision: 0.7806, hmean: 0.6781 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.86, recall: 0.5970, precision: 0.7918, hmean: 0.6808 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.87, recall: 0.5912, precision: 0.7969, hmean: 0.6788 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.88, recall: 0.5864, precision: 0.8061, hmean: 0.6789 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.89, recall: 0.5778, precision: 0.8158, hmean: 0.6764 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.90, recall: 0.5753, precision: 0.8270, hmean: 0.6786 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.91, recall: 0.5705, precision: 0.8345, hmean: 0.6777 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.92, recall: 0.5619, precision: 0.8463, hmean: 0.6753 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.93, recall: 0.5532, precision: 0.8505, hmean: 0.6704 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.94, recall: 0.5431, precision: 0.8598, hmean: 0.6657 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.95, recall: 0.5320, precision: 0.8694, hmean: 0.6601 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.96, recall: 0.5195, precision: 0.8765, hmean: 0.6524 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.97, recall: 0.4969, precision: 0.8813, hmean: 0.6355 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.98, recall: 0.4752, precision: 0.8908, hmean: 0.6198 2023/03/02 23:53:58 - mmengine - INFO - text score threshold: 0.99, recall: 0.4324, precision: 0.9098, hmean: 0.5862 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.80, recall: 0.6577, precision: 0.8069, hmean: 0.7247 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.81, recall: 0.6553, precision: 0.8125, hmean: 0.7255 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.82, recall: 0.6533, precision: 0.8199, hmean: 0.7272 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.83, recall: 0.6509, precision: 0.8269, hmean: 0.7284 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.84, recall: 0.6447, precision: 0.8332, hmean: 0.7269 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.85, recall: 0.6437, precision: 0.8382, hmean: 0.7282 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.86, recall: 0.6399, precision: 0.8487, hmean: 0.7296 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.87, recall: 0.6322, precision: 0.8520, hmean: 0.7258 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.88, recall: 0.6249, precision: 0.8590, hmean: 0.7235 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.89, recall: 0.6129, precision: 0.8654, hmean: 0.7176 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.90, recall: 0.6091, precision: 0.8754, hmean: 0.7183 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.91, recall: 0.6023, precision: 0.8810, hmean: 0.7155 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.92, recall: 0.5893, precision: 0.8876, hmean: 0.7083 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.93, recall: 0.5802, precision: 0.8919, hmean: 0.7030 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.94, recall: 0.5672, precision: 0.8979, hmean: 0.6952 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.95, recall: 0.5527, precision: 0.9032, hmean: 0.6858 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.96, recall: 0.5364, precision: 0.9050, hmean: 0.6735 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.97, recall: 0.5113, precision: 0.9069, hmean: 0.6539 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.98, recall: 0.4863, precision: 0.9116, hmean: 0.6342 2023/03/02 23:54:09 - mmengine - INFO - text score threshold: 0.99, recall: 0.4410, precision: 0.9281, hmean: 0.5979 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.80, recall: 0.7135, precision: 0.8754, hmean: 0.7862 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.81, recall: 0.7102, precision: 0.8806, hmean: 0.7862 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.82, recall: 0.7078, precision: 0.8882, hmean: 0.7878 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.83, recall: 0.7034, precision: 0.8936, hmean: 0.7872 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.84, recall: 0.6962, precision: 0.8998, hmean: 0.7850 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.85, recall: 0.6933, precision: 0.9028, hmean: 0.7843 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.86, recall: 0.6866, precision: 0.9106, hmean: 0.7829 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.87, recall: 0.6779, precision: 0.9137, hmean: 0.7783 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.88, recall: 0.6678, precision: 0.9179, hmean: 0.7731 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.89, recall: 0.6529, precision: 0.9218, hmean: 0.7644 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.90, recall: 0.6442, precision: 0.9260, hmean: 0.7598 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.91, recall: 0.6355, precision: 0.9296, hmean: 0.7549 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.92, recall: 0.6201, precision: 0.9340, hmean: 0.7454 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.93, recall: 0.6086, precision: 0.9356, hmean: 0.7375 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.94, recall: 0.5932, precision: 0.9390, hmean: 0.7271 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.95, recall: 0.5768, precision: 0.9426, hmean: 0.7157 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.96, recall: 0.5599, precision: 0.9448, hmean: 0.7031 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.97, recall: 0.5330, precision: 0.9453, hmean: 0.6817 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.98, recall: 0.5060, precision: 0.9486, hmean: 0.6600 2023/03/02 23:54:19 - mmengine - INFO - text score threshold: 0.99, recall: 0.4569, precision: 0.9615, hmean: 0.6195 2023/03/02 23:54:19 - mmengine - INFO - Epoch(val) [100][63/63] generic/precision: 0.7918 generic/recall: 0.5970 generic/hmean: 0.6808 weak/precision: 0.8487 weak/recall: 0.6399 weak/hmean: 0.7296 strong/precision: 0.8882 strong/recall: 0.7078 strong/hmean: 0.7878 2023/03/02 23:54:25 - mmengine - INFO - Epoch(train) [101][10/32] lr: 1.0000e-06 eta: 0:23:00 time: 0.5345 data_time: 0.0562 memory: 19142 loss: 0.1237 loss_ce: 0.1237 2023/03/02 23:54:29 - mmengine - INFO - Epoch(train) [101][20/32] lr: 1.0000e-06 eta: 0:22:56 time: 0.4518 data_time: 0.0012 memory: 19751 loss: 0.0858 loss_ce: 0.0858 2023/03/02 23:54:34 - mmengine - INFO - Epoch(train) [101][30/32] lr: 1.0000e-06 eta: 0:22:52 time: 0.4718 data_time: 0.0011 memory: 22486 loss: 0.0951 loss_ce: 0.0951 2023/03/02 23:54:35 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:54:40 - mmengine - INFO - Epoch(train) [102][10/32] lr: 1.0000e-06 eta: 0:22:48 time: 0.5273 data_time: 0.0292 memory: 23431 loss: 0.1094 loss_ce: 0.1094 2023/03/02 23:54:44 - mmengine - INFO - Epoch(train) [102][20/32] lr: 1.0000e-06 eta: 0:22:43 time: 0.4082 data_time: 0.0012 memory: 19040 loss: 0.1163 loss_ce: 0.1163 2023/03/02 23:54:48 - mmengine - INFO - Epoch(train) [102][30/32] lr: 1.0000e-06 eta: 0:22:39 time: 0.4221 data_time: 0.0010 memory: 18779 loss: 0.1119 loss_ce: 0.1119 2023/03/02 23:54:49 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:54:54 - mmengine - INFO - Epoch(train) [103][10/32] lr: 1.0000e-06 eta: 0:22:34 time: 0.5391 data_time: 0.0804 memory: 24544 loss: 0.1074 loss_ce: 0.1074 2023/03/02 23:54:59 - mmengine - INFO - Epoch(train) [103][20/32] lr: 1.0000e-06 eta: 0:22:30 time: 0.4621 data_time: 0.0014 memory: 21276 loss: 0.0898 loss_ce: 0.0898 2023/03/02 23:55:03 - mmengine - INFO - Epoch(train) [103][30/32] lr: 1.0000e-06 eta: 0:22:26 time: 0.4286 data_time: 0.0013 memory: 18128 loss: 0.0917 loss_ce: 0.0917 2023/03/02 23:55:04 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:55:09 - mmengine - INFO - Epoch(train) [104][10/32] lr: 1.0000e-06 eta: 0:22:22 time: 0.5282 data_time: 0.0771 memory: 19975 loss: 0.1024 loss_ce: 0.1024 2023/03/02 23:55:14 - mmengine - INFO - Epoch(train) [104][20/32] lr: 1.0000e-06 eta: 0:22:18 time: 0.4648 data_time: 0.0014 memory: 25455 loss: 0.1010 loss_ce: 0.1010 2023/03/02 23:55:18 - mmengine - INFO - Epoch(train) [104][30/32] lr: 1.0000e-06 eta: 0:22:13 time: 0.4345 data_time: 0.0010 memory: 18083 loss: 0.1010 loss_ce: 0.1010 2023/03/02 23:55:19 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:55:25 - mmengine - INFO - Epoch(train) [105][10/32] lr: 1.0000e-06 eta: 0:22:09 time: 0.5547 data_time: 0.0489 memory: 18083 loss: 0.1037 loss_ce: 0.1037 2023/03/02 23:55:29 - mmengine - INFO - Epoch(train) [105][20/32] lr: 1.0000e-06 eta: 0:22:05 time: 0.4090 data_time: 0.0015 memory: 19958 loss: 0.0966 loss_ce: 0.0966 2023/03/02 23:55:34 - mmengine - INFO - Epoch(train) [105][30/32] lr: 1.0000e-06 eta: 0:22:01 time: 0.4746 data_time: 0.0011 memory: 20820 loss: 0.1221 loss_ce: 0.1221 2023/03/02 23:55:34 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:55:39 - mmengine - INFO - Epoch(train) [106][10/32] lr: 1.0000e-06 eta: 0:21:56 time: 0.5155 data_time: 0.0633 memory: 18413 loss: 0.1047 loss_ce: 0.1047 2023/03/02 23:55:44 - mmengine - INFO - Epoch(train) [106][20/32] lr: 1.0000e-06 eta: 0:21:52 time: 0.4649 data_time: 0.0013 memory: 19975 loss: 0.1016 loss_ce: 0.1016 2023/03/02 23:55:48 - mmengine - INFO - Epoch(train) [106][30/32] lr: 1.0000e-06 eta: 0:21:47 time: 0.4124 data_time: 0.0010 memory: 18966 loss: 0.1183 loss_ce: 0.1183 2023/03/02 23:55:49 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:55:54 - mmengine - INFO - Epoch(train) [107][10/32] lr: 1.0000e-06 eta: 0:21:43 time: 0.5490 data_time: 0.0301 memory: 18535 loss: 0.1094 loss_ce: 0.1094 2023/03/02 23:55:59 - mmengine - INFO - Epoch(train) [107][20/32] lr: 1.0000e-06 eta: 0:21:39 time: 0.4532 data_time: 0.0014 memory: 19587 loss: 0.1079 loss_ce: 0.1079 2023/03/02 23:56:03 - mmengine - INFO - Epoch(train) [107][30/32] lr: 1.0000e-06 eta: 0:21:34 time: 0.4501 data_time: 0.0011 memory: 19958 loss: 0.1007 loss_ce: 0.1007 2023/03/02 23:56:04 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:56:10 - mmengine - INFO - Epoch(train) [108][10/32] lr: 1.0000e-06 eta: 0:21:30 time: 0.5606 data_time: 0.0337 memory: 24790 loss: 0.0976 loss_ce: 0.0976 2023/03/02 23:56:14 - mmengine - INFO - Epoch(train) [108][20/32] lr: 1.0000e-06 eta: 0:21:26 time: 0.4417 data_time: 0.0013 memory: 20235 loss: 0.1085 loss_ce: 0.1085 2023/03/02 23:56:19 - mmengine - INFO - Epoch(train) [108][30/32] lr: 1.0000e-06 eta: 0:21:22 time: 0.4735 data_time: 0.0012 memory: 19407 loss: 0.1088 loss_ce: 0.1088 2023/03/02 23:56:20 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:56:25 - mmengine - INFO - Epoch(train) [109][10/32] lr: 1.0000e-06 eta: 0:21:18 time: 0.5508 data_time: 0.0650 memory: 20204 loss: 0.1086 loss_ce: 0.1086 2023/03/02 23:56:30 - mmengine - INFO - Epoch(train) [109][20/32] lr: 1.0000e-06 eta: 0:21:13 time: 0.4684 data_time: 0.0012 memory: 23088 loss: 0.1065 loss_ce: 0.1065 2023/03/02 23:56:35 - mmengine - INFO - Epoch(train) [109][30/32] lr: 1.0000e-06 eta: 0:21:10 time: 0.5017 data_time: 0.0011 memory: 21688 loss: 0.0931 loss_ce: 0.0931 2023/03/02 23:56:35 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:56:41 - mmengine - INFO - Epoch(train) [110][10/32] lr: 1.0000e-06 eta: 0:21:05 time: 0.5113 data_time: 0.0788 memory: 18222 loss: 0.1109 loss_ce: 0.1109 2023/03/02 23:56:45 - mmengine - INFO - Epoch(train) [110][20/32] lr: 1.0000e-06 eta: 0:21:01 time: 0.4411 data_time: 0.0013 memory: 17733 loss: 0.1169 loss_ce: 0.1169 2023/03/02 23:56:50 - mmengine - INFO - Epoch(train) [110][30/32] lr: 1.0000e-06 eta: 0:20:57 time: 0.4790 data_time: 0.0011 memory: 25455 loss: 0.0993 loss_ce: 0.0993 2023/03/02 23:56:50 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/02 23:57:04 - mmengine - INFO - Epoch(val) [110][10/63] eta: 0:01:12 time: 1.3660 data_time: 0.0040 memory: 15988 2023/03/02 23:57:28 - mmengine - INFO - Epoch(val) [110][20/63] eta: 0:01:20 time: 2.4009 data_time: 0.0004 memory: 1075 2023/03/02 23:57:49 - mmengine - INFO - Epoch(val) [110][30/63] eta: 0:01:05 time: 2.1459 data_time: 0.0003 memory: 1075 2023/03/02 23:58:00 - mmengine - INFO - Epoch(val) [110][40/63] eta: 0:00:40 time: 1.0481 data_time: 0.0004 memory: 1075 2023/03/02 23:58:22 - mmengine - INFO - Epoch(val) [110][50/63] eta: 0:00:23 time: 2.2122 data_time: 0.0003 memory: 1075 2023/03/02 23:58:35 - mmengine - INFO - Epoch(val) [110][60/63] eta: 0:00:05 time: 1.3121 data_time: 0.0003 memory: 1075 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.80, recall: 0.6100, precision: 0.7533, hmean: 0.6741 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.81, recall: 0.6095, precision: 0.7581, hmean: 0.6757 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.82, recall: 0.6076, precision: 0.7648, hmean: 0.6772 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.83, recall: 0.6066, precision: 0.7735, hmean: 0.6800 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.84, recall: 0.6023, precision: 0.7833, hmean: 0.6810 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.85, recall: 0.5985, precision: 0.7907, hmean: 0.6813 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.86, recall: 0.5941, precision: 0.7977, hmean: 0.6810 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.87, recall: 0.5922, precision: 0.8039, hmean: 0.6820 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.88, recall: 0.5898, precision: 0.8107, hmean: 0.6828 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.89, recall: 0.5831, precision: 0.8199, hmean: 0.6815 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.90, recall: 0.5753, precision: 0.8299, hmean: 0.6796 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.91, recall: 0.5681, precision: 0.8375, hmean: 0.6770 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.92, recall: 0.5585, precision: 0.8436, hmean: 0.6721 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.93, recall: 0.5503, precision: 0.8523, hmean: 0.6688 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.94, recall: 0.5383, precision: 0.8627, hmean: 0.6629 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.95, recall: 0.5267, precision: 0.8703, hmean: 0.6563 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.96, recall: 0.5084, precision: 0.8720, hmean: 0.6423 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.97, recall: 0.4940, precision: 0.8875, hmean: 0.6347 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.98, recall: 0.4718, precision: 0.8983, hmean: 0.6187 2023/03/02 23:59:35 - mmengine - INFO - text score threshold: 0.99, recall: 0.4304, precision: 0.9150, hmean: 0.5855 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.80, recall: 0.6562, precision: 0.8103, hmean: 0.7252 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.81, recall: 0.6543, precision: 0.8138, hmean: 0.7254 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.82, recall: 0.6514, precision: 0.8200, hmean: 0.7261 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.83, recall: 0.6500, precision: 0.8287, hmean: 0.7285 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.84, recall: 0.6432, precision: 0.8366, hmean: 0.7273 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.85, recall: 0.6389, precision: 0.8441, hmean: 0.7273 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.86, recall: 0.6326, precision: 0.8494, hmean: 0.7252 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.87, recall: 0.6293, precision: 0.8542, hmean: 0.7247 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.88, recall: 0.6254, precision: 0.8597, hmean: 0.7241 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.89, recall: 0.6163, precision: 0.8666, hmean: 0.7203 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.90, recall: 0.6062, precision: 0.8743, hmean: 0.7160 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.91, recall: 0.5985, precision: 0.8822, hmean: 0.7131 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.92, recall: 0.5883, precision: 0.8887, hmean: 0.7080 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.93, recall: 0.5768, precision: 0.8934, hmean: 0.7010 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.94, recall: 0.5595, precision: 0.8966, hmean: 0.6890 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.95, recall: 0.5460, precision: 0.9021, hmean: 0.6803 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.96, recall: 0.5277, precision: 0.9050, hmean: 0.6667 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.97, recall: 0.5075, precision: 0.9118, hmean: 0.6520 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.98, recall: 0.4810, precision: 0.9157, hmean: 0.6307 2023/03/02 23:59:44 - mmengine - INFO - text score threshold: 0.99, recall: 0.4372, precision: 0.9294, hmean: 0.5946 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.80, recall: 0.7106, precision: 0.8775, hmean: 0.7853 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.81, recall: 0.7078, precision: 0.8802, hmean: 0.7846 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.82, recall: 0.7039, precision: 0.8861, hmean: 0.7845 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.83, recall: 0.7005, precision: 0.8932, hmean: 0.7852 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.84, recall: 0.6923, precision: 0.9004, hmean: 0.7828 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.85, recall: 0.6870, precision: 0.9078, hmean: 0.7821 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.86, recall: 0.6793, precision: 0.9121, hmean: 0.7787 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.87, recall: 0.6750, precision: 0.9163, hmean: 0.7774 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.88, recall: 0.6683, precision: 0.9186, hmean: 0.7737 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.89, recall: 0.6562, precision: 0.9228, hmean: 0.7670 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.90, recall: 0.6423, precision: 0.9264, hmean: 0.7586 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.91, recall: 0.6331, precision: 0.9333, hmean: 0.7544 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.92, recall: 0.6192, precision: 0.9353, hmean: 0.7451 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.93, recall: 0.6052, precision: 0.9374, hmean: 0.7355 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.94, recall: 0.5874, precision: 0.9414, hmean: 0.7234 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.95, recall: 0.5715, precision: 0.9443, hmean: 0.7121 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.96, recall: 0.5522, precision: 0.9472, hmean: 0.6977 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.97, recall: 0.5291, precision: 0.9507, hmean: 0.6799 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.98, recall: 0.5022, precision: 0.9560, hmean: 0.6585 2023/03/02 23:59:53 - mmengine - INFO - text score threshold: 0.99, recall: 0.4535, precision: 0.9642, hmean: 0.6169 2023/03/02 23:59:53 - mmengine - INFO - Epoch(val) [110][63/63] generic/precision: 0.8107 generic/recall: 0.5898 generic/hmean: 0.6828 weak/precision: 0.8287 weak/recall: 0.6500 weak/hmean: 0.7285 strong/precision: 0.8775 strong/recall: 0.7106 strong/hmean: 0.7853 2023/03/02 23:59:58 - mmengine - INFO - Epoch(train) [111][10/32] lr: 1.0000e-06 eta: 0:20:52 time: 0.5256 data_time: 0.0804 memory: 19174 loss: 0.1283 loss_ce: 0.1283 2023/03/03 00:00:03 - mmengine - INFO - Epoch(train) [111][20/32] lr: 1.0000e-06 eta: 0:20:48 time: 0.5016 data_time: 0.0014 memory: 18741 loss: 0.1051 loss_ce: 0.1051 2023/03/03 00:00:08 - mmengine - INFO - Epoch(train) [111][30/32] lr: 1.0000e-06 eta: 0:20:44 time: 0.4359 data_time: 0.0011 memory: 23779 loss: 0.1052 loss_ce: 0.1052 2023/03/03 00:00:08 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:00:13 - mmengine - INFO - Epoch(train) [112][10/32] lr: 1.0000e-06 eta: 0:20:39 time: 0.5087 data_time: 0.0622 memory: 21985 loss: 0.1055 loss_ce: 0.1055 2023/03/03 00:00:18 - mmengine - INFO - Epoch(train) [112][20/32] lr: 1.0000e-06 eta: 0:20:34 time: 0.4261 data_time: 0.0014 memory: 19975 loss: 0.0967 loss_ce: 0.0967 2023/03/03 00:00:22 - mmengine - INFO - Epoch(train) [112][30/32] lr: 1.0000e-06 eta: 0:20:30 time: 0.4619 data_time: 0.0012 memory: 24319 loss: 0.0927 loss_ce: 0.0927 2023/03/03 00:00:23 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:00:28 - mmengine - INFO - Epoch(train) [113][10/32] lr: 1.0000e-06 eta: 0:20:25 time: 0.5303 data_time: 0.1236 memory: 18083 loss: 0.1111 loss_ce: 0.1111 2023/03/03 00:00:32 - mmengine - INFO - Epoch(train) [113][20/32] lr: 1.0000e-06 eta: 0:20:21 time: 0.4368 data_time: 0.0015 memory: 22984 loss: 0.1022 loss_ce: 0.1022 2023/03/03 00:00:37 - mmengine - INFO - Epoch(train) [113][30/32] lr: 1.0000e-06 eta: 0:20:17 time: 0.4681 data_time: 0.0014 memory: 19751 loss: 0.0984 loss_ce: 0.0984 2023/03/03 00:00:38 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:00:43 - mmengine - INFO - Epoch(train) [114][10/32] lr: 1.0000e-06 eta: 0:20:12 time: 0.5047 data_time: 0.0669 memory: 18413 loss: 0.0999 loss_ce: 0.0999 2023/03/03 00:00:47 - mmengine - INFO - Epoch(train) [114][20/32] lr: 1.0000e-06 eta: 0:20:08 time: 0.4693 data_time: 0.0015 memory: 18325 loss: 0.1066 loss_ce: 0.1066 2023/03/03 00:00:52 - mmengine - INFO - Epoch(train) [114][30/32] lr: 1.0000e-06 eta: 0:20:04 time: 0.4879 data_time: 0.0011 memory: 23431 loss: 0.1030 loss_ce: 0.1030 2023/03/03 00:00:53 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:00:58 - mmengine - INFO - Epoch(train) [115][10/32] lr: 1.0000e-06 eta: 0:19:59 time: 0.5082 data_time: 0.0533 memory: 18940 loss: 0.0993 loss_ce: 0.0993 2023/03/03 00:01:03 - mmengine - INFO - Epoch(train) [115][20/32] lr: 1.0000e-06 eta: 0:19:55 time: 0.4913 data_time: 0.0013 memory: 19958 loss: 0.1004 loss_ce: 0.1004 2023/03/03 00:01:07 - mmengine - INFO - Epoch(train) [115][30/32] lr: 1.0000e-06 eta: 0:19:50 time: 0.4020 data_time: 0.0013 memory: 18348 loss: 0.1111 loss_ce: 0.1111 2023/03/03 00:01:08 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:01:13 - mmengine - INFO - Epoch(train) [116][10/32] lr: 1.0000e-06 eta: 0:19:46 time: 0.5737 data_time: 0.1020 memory: 24319 loss: 0.1187 loss_ce: 0.1187 2023/03/03 00:01:18 - mmengine - INFO - Epoch(train) [116][20/32] lr: 1.0000e-06 eta: 0:19:42 time: 0.4754 data_time: 0.0012 memory: 21276 loss: 0.1055 loss_ce: 0.1055 2023/03/03 00:01:23 - mmengine - INFO - Epoch(train) [116][30/32] lr: 1.0000e-06 eta: 0:19:38 time: 0.4834 data_time: 0.0012 memory: 18526 loss: 0.1052 loss_ce: 0.1052 2023/03/03 00:01:23 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:01:29 - mmengine - INFO - Epoch(train) [117][10/32] lr: 1.0000e-06 eta: 0:19:33 time: 0.5396 data_time: 0.0725 memory: 19174 loss: 0.1016 loss_ce: 0.1016 2023/03/03 00:01:33 - mmengine - INFO - Epoch(train) [117][20/32] lr: 1.0000e-06 eta: 0:19:28 time: 0.4411 data_time: 0.0014 memory: 19751 loss: 0.1114 loss_ce: 0.1114 2023/03/03 00:01:38 - mmengine - INFO - Epoch(train) [117][30/32] lr: 1.0000e-06 eta: 0:19:24 time: 0.4493 data_time: 0.0013 memory: 19751 loss: 0.0992 loss_ce: 0.0992 2023/03/03 00:01:38 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:01:44 - mmengine - INFO - Epoch(train) [118][10/32] lr: 1.0000e-06 eta: 0:19:19 time: 0.5308 data_time: 0.0332 memory: 19150 loss: 0.1070 loss_ce: 0.1070 2023/03/03 00:01:49 - mmengine - INFO - Epoch(train) [118][20/32] lr: 1.0000e-06 eta: 0:19:15 time: 0.4847 data_time: 0.0015 memory: 20600 loss: 0.1037 loss_ce: 0.1037 2023/03/03 00:01:53 - mmengine - INFO - Epoch(train) [118][30/32] lr: 1.0000e-06 eta: 0:19:11 time: 0.4510 data_time: 0.0013 memory: 19958 loss: 0.1098 loss_ce: 0.1098 2023/03/03 00:01:54 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:01:59 - mmengine - INFO - Epoch(train) [119][10/32] lr: 1.0000e-06 eta: 0:19:07 time: 0.5642 data_time: 0.1040 memory: 19751 loss: 0.1119 loss_ce: 0.1119 2023/03/03 00:02:04 - mmengine - INFO - Epoch(train) [119][20/32] lr: 1.0000e-06 eta: 0:19:02 time: 0.4528 data_time: 0.0013 memory: 20820 loss: 0.1233 loss_ce: 0.1233 2023/03/03 00:02:09 - mmengine - INFO - Epoch(train) [119][30/32] lr: 1.0000e-06 eta: 0:18:58 time: 0.4978 data_time: 0.0011 memory: 18597 loss: 0.0848 loss_ce: 0.0848 2023/03/03 00:02:09 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:02:15 - mmengine - INFO - Epoch(train) [120][10/32] lr: 1.0000e-06 eta: 0:18:53 time: 0.5390 data_time: 0.0501 memory: 19751 loss: 0.0892 loss_ce: 0.0892 2023/03/03 00:02:19 - mmengine - INFO - Epoch(train) [120][20/32] lr: 1.0000e-06 eta: 0:18:49 time: 0.4545 data_time: 0.0013 memory: 22351 loss: 0.0985 loss_ce: 0.0985 2023/03/03 00:02:24 - mmengine - INFO - Epoch(train) [120][30/32] lr: 1.0000e-06 eta: 0:18:45 time: 0.4711 data_time: 0.0014 memory: 18597 loss: 0.1066 loss_ce: 0.1066 2023/03/03 00:02:25 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:02:38 - mmengine - INFO - Epoch(val) [120][10/63] eta: 0:01:12 time: 1.3661 data_time: 0.0036 memory: 15198 2023/03/03 00:03:02 - mmengine - INFO - Epoch(val) [120][20/63] eta: 0:01:20 time: 2.3595 data_time: 0.0004 memory: 1075 2023/03/03 00:03:24 - mmengine - INFO - Epoch(val) [120][30/63] eta: 0:01:05 time: 2.2483 data_time: 0.0004 memory: 1075 2023/03/03 00:03:36 - mmengine - INFO - Epoch(val) [120][40/63] eta: 0:00:40 time: 1.1312 data_time: 0.0004 memory: 1075 2023/03/03 00:03:58 - mmengine - INFO - Epoch(val) [120][50/63] eta: 0:00:24 time: 2.2035 data_time: 0.0004 memory: 1075 2023/03/03 00:04:11 - mmengine - INFO - Epoch(val) [120][60/63] eta: 0:00:05 time: 1.2896 data_time: 0.0003 memory: 1075 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.80, recall: 0.6124, precision: 0.7540, hmean: 0.6759 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.81, recall: 0.6115, precision: 0.7600, hmean: 0.6777 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.82, recall: 0.6095, precision: 0.7636, hmean: 0.6779 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.83, recall: 0.6081, precision: 0.7744, hmean: 0.6812 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.84, recall: 0.6042, precision: 0.7824, hmean: 0.6819 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.85, recall: 0.6009, precision: 0.7879, hmean: 0.6818 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.86, recall: 0.5975, precision: 0.7986, hmean: 0.6836 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.87, recall: 0.5941, precision: 0.8029, hmean: 0.6829 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.88, recall: 0.5893, precision: 0.8155, hmean: 0.6842 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.89, recall: 0.5840, precision: 0.8224, hmean: 0.6830 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.90, recall: 0.5778, precision: 0.8304, hmean: 0.6814 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.91, recall: 0.5676, precision: 0.8391, hmean: 0.6772 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.92, recall: 0.5599, precision: 0.8446, hmean: 0.6734 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.93, recall: 0.5518, precision: 0.8533, hmean: 0.6702 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.94, recall: 0.5402, precision: 0.8651, hmean: 0.6651 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.95, recall: 0.5291, precision: 0.8729, hmean: 0.6589 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.96, recall: 0.5147, precision: 0.8806, hmean: 0.6497 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.97, recall: 0.4974, precision: 0.8921, hmean: 0.6386 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.98, recall: 0.4733, precision: 0.8969, hmean: 0.6196 2023/03/03 00:05:11 - mmengine - INFO - text score threshold: 0.99, recall: 0.4328, precision: 0.9127, hmean: 0.5872 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.80, recall: 0.6615, precision: 0.8145, hmean: 0.7301 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.81, recall: 0.6591, precision: 0.8193, hmean: 0.7305 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.82, recall: 0.6567, precision: 0.8227, hmean: 0.7304 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.83, recall: 0.6529, precision: 0.8314, hmean: 0.7314 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.84, recall: 0.6476, precision: 0.8385, hmean: 0.7308 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.85, recall: 0.6423, precision: 0.8422, hmean: 0.7288 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.86, recall: 0.6375, precision: 0.8520, hmean: 0.7293 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.87, recall: 0.6326, precision: 0.8549, hmean: 0.7272 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.88, recall: 0.6249, precision: 0.8648, hmean: 0.7255 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.89, recall: 0.6177, precision: 0.8698, hmean: 0.7224 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.90, recall: 0.6095, precision: 0.8761, hmean: 0.7189 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.91, recall: 0.5961, precision: 0.8811, hmean: 0.7111 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.92, recall: 0.5859, precision: 0.8838, hmean: 0.7047 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.93, recall: 0.5763, precision: 0.8913, hmean: 0.7000 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.94, recall: 0.5623, precision: 0.9005, hmean: 0.6924 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.95, recall: 0.5508, precision: 0.9087, hmean: 0.6859 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.96, recall: 0.5325, precision: 0.9110, hmean: 0.6721 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.97, recall: 0.5099, precision: 0.9145, hmean: 0.6547 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.98, recall: 0.4824, precision: 0.9142, hmean: 0.6316 2023/03/03 00:05:20 - mmengine - INFO - text score threshold: 0.99, recall: 0.4396, precision: 0.9269, hmean: 0.5963 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.80, recall: 0.7155, precision: 0.8809, hmean: 0.7896 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.81, recall: 0.7116, precision: 0.8845, hmean: 0.7887 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.82, recall: 0.7087, precision: 0.8878, hmean: 0.7882 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.83, recall: 0.7029, precision: 0.8952, hmean: 0.7875 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.84, recall: 0.6957, precision: 0.9009, hmean: 0.7851 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.85, recall: 0.6895, precision: 0.9040, hmean: 0.7823 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.86, recall: 0.6827, precision: 0.9125, hmean: 0.7811 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.87, recall: 0.6769, precision: 0.9148, hmean: 0.7781 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.88, recall: 0.6659, precision: 0.9214, hmean: 0.7731 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.89, recall: 0.6567, precision: 0.9247, hmean: 0.7680 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.90, recall: 0.6452, precision: 0.9273, hmean: 0.7609 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.91, recall: 0.6298, precision: 0.9310, hmean: 0.7513 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.92, recall: 0.6182, precision: 0.9325, hmean: 0.7435 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.93, recall: 0.6047, precision: 0.9352, hmean: 0.7345 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.94, recall: 0.5888, precision: 0.9429, hmean: 0.7250 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.95, recall: 0.5753, precision: 0.9492, hmean: 0.7164 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.96, recall: 0.5566, precision: 0.9522, hmean: 0.7025 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.97, recall: 0.5320, precision: 0.9542, hmean: 0.6832 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.98, recall: 0.5041, precision: 0.9553, hmean: 0.6599 2023/03/03 00:05:29 - mmengine - INFO - text score threshold: 0.99, recall: 0.4574, precision: 0.9645, hmean: 0.6205 2023/03/03 00:05:29 - mmengine - INFO - Epoch(val) [120][63/63] generic/precision: 0.8155 generic/recall: 0.5893 generic/hmean: 0.6842 weak/precision: 0.8314 weak/recall: 0.6529 weak/hmean: 0.7314 strong/precision: 0.8809 strong/recall: 0.7155 strong/hmean: 0.7896 2023/03/03 00:05:34 - mmengine - INFO - Epoch(train) [121][10/32] lr: 1.0000e-06 eta: 0:18:40 time: 0.5216 data_time: 0.0539 memory: 18741 loss: 0.1128 loss_ce: 0.1128 2023/03/03 00:05:39 - mmengine - INFO - Epoch(train) [121][20/32] lr: 1.0000e-06 eta: 0:18:36 time: 0.4850 data_time: 0.0016 memory: 18966 loss: 0.0930 loss_ce: 0.0930 2023/03/03 00:05:43 - mmengine - INFO - Epoch(train) [121][30/32] lr: 1.0000e-06 eta: 0:18:31 time: 0.4166 data_time: 0.0013 memory: 17919 loss: 0.1228 loss_ce: 0.1228 2023/03/03 00:05:44 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:05:50 - mmengine - INFO - Epoch(train) [122][10/32] lr: 1.0000e-06 eta: 0:18:27 time: 0.5473 data_time: 0.0618 memory: 24319 loss: 0.1008 loss_ce: 0.1008 2023/03/03 00:05:54 - mmengine - INFO - Epoch(train) [122][20/32] lr: 1.0000e-06 eta: 0:18:22 time: 0.4606 data_time: 0.0015 memory: 17428 loss: 0.0994 loss_ce: 0.0994 2023/03/03 00:05:59 - mmengine - INFO - Epoch(train) [122][30/32] lr: 1.0000e-06 eta: 0:18:18 time: 0.4799 data_time: 0.0015 memory: 21255 loss: 0.1070 loss_ce: 0.1070 2023/03/03 00:06:00 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:06:05 - mmengine - INFO - Epoch(train) [123][10/32] lr: 1.0000e-06 eta: 0:18:13 time: 0.5029 data_time: 0.0402 memory: 18345 loss: 0.1112 loss_ce: 0.1112 2023/03/03 00:06:09 - mmengine - INFO - Epoch(train) [123][20/32] lr: 1.0000e-06 eta: 0:18:09 time: 0.4551 data_time: 0.0015 memory: 25455 loss: 0.1242 loss_ce: 0.1242 2023/03/03 00:06:14 - mmengine - INFO - Epoch(train) [123][30/32] lr: 1.0000e-06 eta: 0:18:05 time: 0.4891 data_time: 0.0013 memory: 22545 loss: 0.0956 loss_ce: 0.0956 2023/03/03 00:06:15 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:06:21 - mmengine - INFO - Epoch(train) [124][10/32] lr: 1.0000e-06 eta: 0:18:01 time: 0.5978 data_time: 0.0917 memory: 25455 loss: 0.0969 loss_ce: 0.0969 2023/03/03 00:06:25 - mmengine - INFO - Epoch(train) [124][20/32] lr: 1.0000e-06 eta: 0:17:56 time: 0.4559 data_time: 0.0013 memory: 21046 loss: 0.0974 loss_ce: 0.0974 2023/03/03 00:06:30 - mmengine - INFO - Epoch(train) [124][30/32] lr: 1.0000e-06 eta: 0:17:52 time: 0.4510 data_time: 0.0011 memory: 20832 loss: 0.1098 loss_ce: 0.1098 2023/03/03 00:06:31 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:06:36 - mmengine - INFO - Epoch(train) [125][10/32] lr: 1.0000e-06 eta: 0:17:47 time: 0.5609 data_time: 0.0784 memory: 19789 loss: 0.1042 loss_ce: 0.1042 2023/03/03 00:06:41 - mmengine - INFO - Epoch(train) [125][20/32] lr: 1.0000e-06 eta: 0:17:43 time: 0.4496 data_time: 0.0013 memory: 22104 loss: 0.1033 loss_ce: 0.1033 2023/03/03 00:06:45 - mmengine - INFO - Epoch(train) [125][30/32] lr: 1.0000e-06 eta: 0:17:38 time: 0.4310 data_time: 0.0011 memory: 18246 loss: 0.1003 loss_ce: 0.1003 2023/03/03 00:06:46 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:06:51 - mmengine - INFO - Epoch(train) [126][10/32] lr: 1.0000e-06 eta: 0:17:34 time: 0.5509 data_time: 0.0463 memory: 19958 loss: 0.1245 loss_ce: 0.1245 2023/03/03 00:06:56 - mmengine - INFO - Epoch(train) [126][20/32] lr: 1.0000e-06 eta: 0:17:29 time: 0.4597 data_time: 0.0013 memory: 19142 loss: 0.0994 loss_ce: 0.0994 2023/03/03 00:07:00 - mmengine - INFO - Epoch(train) [126][30/32] lr: 1.0000e-06 eta: 0:17:25 time: 0.4661 data_time: 0.0013 memory: 18128 loss: 0.1160 loss_ce: 0.1160 2023/03/03 00:07:01 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:07:07 - mmengine - INFO - Epoch(train) [127][10/32] lr: 1.0000e-06 eta: 0:17:20 time: 0.5696 data_time: 0.0996 memory: 18966 loss: 0.1190 loss_ce: 0.1190 2023/03/03 00:07:11 - mmengine - INFO - Epoch(train) [127][20/32] lr: 1.0000e-06 eta: 0:17:16 time: 0.4543 data_time: 0.0013 memory: 19958 loss: 0.1001 loss_ce: 0.1001 2023/03/03 00:07:15 - mmengine - INFO - Epoch(train) [127][30/32] lr: 1.0000e-06 eta: 0:17:11 time: 0.4072 data_time: 0.0012 memory: 17919 loss: 0.1233 loss_ce: 0.1233 2023/03/03 00:07:16 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:07:22 - mmengine - INFO - Epoch(train) [128][10/32] lr: 1.0000e-06 eta: 0:17:07 time: 0.5537 data_time: 0.0816 memory: 18911 loss: 0.1009 loss_ce: 0.1009 2023/03/03 00:07:26 - mmengine - INFO - Epoch(train) [128][20/32] lr: 1.0000e-06 eta: 0:17:02 time: 0.4538 data_time: 0.0015 memory: 19958 loss: 0.1058 loss_ce: 0.1058 2023/03/03 00:07:31 - mmengine - INFO - Epoch(train) [128][30/32] lr: 1.0000e-06 eta: 0:16:58 time: 0.4403 data_time: 0.0013 memory: 20204 loss: 0.1004 loss_ce: 0.1004 2023/03/03 00:07:31 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:07:37 - mmengine - INFO - Epoch(train) [129][10/32] lr: 1.0000e-06 eta: 0:16:53 time: 0.5429 data_time: 0.0797 memory: 19958 loss: 0.1100 loss_ce: 0.1100 2023/03/03 00:07:42 - mmengine - INFO - Epoch(train) [129][20/32] lr: 1.0000e-06 eta: 0:16:49 time: 0.4861 data_time: 0.0014 memory: 25281 loss: 0.0921 loss_ce: 0.0921 2023/03/03 00:07:46 - mmengine - INFO - Epoch(train) [129][30/32] lr: 1.0000e-06 eta: 0:16:44 time: 0.4522 data_time: 0.0012 memory: 19751 loss: 0.1126 loss_ce: 0.1126 2023/03/03 00:07:47 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:07:52 - mmengine - INFO - Epoch(train) [130][10/32] lr: 1.0000e-06 eta: 0:16:39 time: 0.5174 data_time: 0.0336 memory: 19552 loss: 0.0995 loss_ce: 0.0995 2023/03/03 00:07:56 - mmengine - INFO - Epoch(train) [130][20/32] lr: 1.0000e-06 eta: 0:16:35 time: 0.4433 data_time: 0.0013 memory: 20820 loss: 0.0927 loss_ce: 0.0927 2023/03/03 00:08:01 - mmengine - INFO - Epoch(train) [130][30/32] lr: 1.0000e-06 eta: 0:16:31 time: 0.4414 data_time: 0.0011 memory: 18282 loss: 0.0993 loss_ce: 0.0993 2023/03/03 00:08:01 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:08:15 - mmengine - INFO - Epoch(val) [130][10/63] eta: 0:01:14 time: 1.4013 data_time: 0.0035 memory: 16226 2023/03/03 00:08:40 - mmengine - INFO - Epoch(val) [130][20/63] eta: 0:01:23 time: 2.4883 data_time: 0.0004 memory: 1075 2023/03/03 00:09:02 - mmengine - INFO - Epoch(val) [130][30/63] eta: 0:01:06 time: 2.1781 data_time: 0.0006 memory: 1075 2023/03/03 00:09:12 - mmengine - INFO - Epoch(val) [130][40/63] eta: 0:00:40 time: 1.0096 data_time: 0.0004 memory: 1075 2023/03/03 00:09:34 - mmengine - INFO - Epoch(val) [130][50/63] eta: 0:00:24 time: 2.1765 data_time: 0.0004 memory: 1075 2023/03/03 00:09:47 - mmengine - INFO - Epoch(val) [130][60/63] eta: 0:00:05 time: 1.2816 data_time: 0.0004 memory: 1075 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.80, recall: 0.6095, precision: 0.7513, hmean: 0.6730 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.81, recall: 0.6086, precision: 0.7573, hmean: 0.6749 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.82, recall: 0.6062, precision: 0.7626, hmean: 0.6754 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.83, recall: 0.6033, precision: 0.7678, hmean: 0.6757 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.84, recall: 0.6004, precision: 0.7755, hmean: 0.6768 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.85, recall: 0.5989, precision: 0.7824, hmean: 0.6785 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.86, recall: 0.5951, precision: 0.7888, hmean: 0.6784 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.87, recall: 0.5912, precision: 0.7948, hmean: 0.6781 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.88, recall: 0.5893, precision: 0.8090, hmean: 0.6819 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.89, recall: 0.5859, precision: 0.8130, hmean: 0.6810 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.90, recall: 0.5782, precision: 0.8215, hmean: 0.6787 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.91, recall: 0.5715, precision: 0.8330, hmean: 0.6779 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.92, recall: 0.5604, precision: 0.8368, hmean: 0.6713 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.93, recall: 0.5513, precision: 0.8519, hmean: 0.6694 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.94, recall: 0.5392, precision: 0.8582, hmean: 0.6623 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.95, recall: 0.5291, precision: 0.8702, hmean: 0.6581 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.96, recall: 0.5152, precision: 0.8799, hmean: 0.6499 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.97, recall: 0.4998, precision: 0.8879, hmean: 0.6396 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.98, recall: 0.4728, precision: 0.8968, hmean: 0.6192 2023/03/03 00:10:47 - mmengine - INFO - text score threshold: 0.99, recall: 0.4381, precision: 0.9137, hmean: 0.5923 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.80, recall: 0.6572, precision: 0.8101, hmean: 0.7257 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.81, recall: 0.6548, precision: 0.8149, hmean: 0.7261 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.82, recall: 0.6519, precision: 0.8201, hmean: 0.7264 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.83, recall: 0.6471, precision: 0.8235, hmean: 0.7247 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.84, recall: 0.6428, precision: 0.8302, hmean: 0.7246 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.85, recall: 0.6408, precision: 0.8371, hmean: 0.7259 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.86, recall: 0.6351, precision: 0.8417, hmean: 0.7239 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.87, recall: 0.6302, precision: 0.8472, hmean: 0.7228 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.88, recall: 0.6259, precision: 0.8592, hmean: 0.7242 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.89, recall: 0.6225, precision: 0.8637, hmean: 0.7236 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.90, recall: 0.6124, precision: 0.8700, hmean: 0.7188 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.91, recall: 0.6033, precision: 0.8793, hmean: 0.7156 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.92, recall: 0.5917, precision: 0.8835, hmean: 0.7088 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.93, recall: 0.5773, precision: 0.8921, hmean: 0.7010 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.94, recall: 0.5638, precision: 0.8973, hmean: 0.6925 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.95, recall: 0.5503, precision: 0.9050, hmean: 0.6844 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.96, recall: 0.5344, precision: 0.9128, hmean: 0.6742 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.97, recall: 0.5152, precision: 0.9153, hmean: 0.6593 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.98, recall: 0.4848, precision: 0.9196, hmean: 0.6349 2023/03/03 00:10:56 - mmengine - INFO - text score threshold: 0.99, recall: 0.4454, precision: 0.9287, hmean: 0.6020 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.80, recall: 0.7102, precision: 0.8754, hmean: 0.7842 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.81, recall: 0.7078, precision: 0.8808, hmean: 0.7848 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.82, recall: 0.7044, precision: 0.8861, hmean: 0.7849 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.83, recall: 0.6986, precision: 0.8891, hmean: 0.7824 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.84, recall: 0.6933, precision: 0.8955, hmean: 0.7815 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.85, recall: 0.6880, precision: 0.8987, hmean: 0.7794 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.86, recall: 0.6818, precision: 0.9036, hmean: 0.7772 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.87, recall: 0.6745, precision: 0.9068, hmean: 0.7736 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.88, recall: 0.6683, precision: 0.9174, hmean: 0.7733 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.89, recall: 0.6635, precision: 0.9205, hmean: 0.7711 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.90, recall: 0.6509, precision: 0.9248, hmean: 0.7641 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.91, recall: 0.6389, precision: 0.9312, hmean: 0.7579 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.92, recall: 0.6245, precision: 0.9324, hmean: 0.7480 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.93, recall: 0.6062, precision: 0.9368, hmean: 0.7360 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.94, recall: 0.5912, precision: 0.9410, hmean: 0.7262 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.95, recall: 0.5744, precision: 0.9446, hmean: 0.7144 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.96, recall: 0.5551, precision: 0.9482, hmean: 0.7003 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.97, recall: 0.5349, precision: 0.9504, hmean: 0.6845 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.98, recall: 0.5031, precision: 0.9543, hmean: 0.6589 2023/03/03 00:11:05 - mmengine - INFO - text score threshold: 0.99, recall: 0.4608, precision: 0.9608, hmean: 0.6228 2023/03/03 00:11:05 - mmengine - INFO - Epoch(val) [130][63/63] generic/precision: 0.8090 generic/recall: 0.5893 generic/hmean: 0.6819 weak/precision: 0.8201 weak/recall: 0.6519 weak/hmean: 0.7264 strong/precision: 0.8861 strong/recall: 0.7044 strong/hmean: 0.7849 2023/03/03 00:11:11 - mmengine - INFO - Epoch(train) [131][10/32] lr: 1.0000e-06 eta: 0:16:26 time: 0.5529 data_time: 0.0880 memory: 17428 loss: 0.1139 loss_ce: 0.1139 2023/03/03 00:11:15 - mmengine - INFO - Epoch(train) [131][20/32] lr: 1.0000e-06 eta: 0:16:22 time: 0.4935 data_time: 0.0015 memory: 19958 loss: 0.0934 loss_ce: 0.0934 2023/03/03 00:11:20 - mmengine - INFO - Epoch(train) [131][30/32] lr: 1.0000e-06 eta: 0:16:17 time: 0.4627 data_time: 0.0011 memory: 19344 loss: 0.1194 loss_ce: 0.1194 2023/03/03 00:11:21 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:11:26 - mmengine - INFO - Epoch(train) [132][10/32] lr: 1.0000e-06 eta: 0:16:12 time: 0.5334 data_time: 0.0997 memory: 19552 loss: 0.0967 loss_ce: 0.0967 2023/03/03 00:11:31 - mmengine - INFO - Epoch(train) [132][20/32] lr: 1.0000e-06 eta: 0:16:08 time: 0.4904 data_time: 0.0015 memory: 21509 loss: 0.0875 loss_ce: 0.0875 2023/03/03 00:11:35 - mmengine - INFO - Epoch(train) [132][30/32] lr: 1.0000e-06 eta: 0:16:04 time: 0.4387 data_time: 0.0013 memory: 19346 loss: 0.1007 loss_ce: 0.1007 2023/03/03 00:11:36 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:11:41 - mmengine - INFO - Epoch(train) [133][10/32] lr: 1.0000e-06 eta: 0:15:59 time: 0.5422 data_time: 0.1010 memory: 25455 loss: 0.0903 loss_ce: 0.0903 2023/03/03 00:11:46 - mmengine - INFO - Epoch(train) [133][20/32] lr: 1.0000e-06 eta: 0:15:54 time: 0.4734 data_time: 0.0015 memory: 19552 loss: 0.1073 loss_ce: 0.1073 2023/03/03 00:11:51 - mmengine - INFO - Epoch(train) [133][30/32] lr: 1.0000e-06 eta: 0:15:50 time: 0.4666 data_time: 0.0014 memory: 18434 loss: 0.1242 loss_ce: 0.1242 2023/03/03 00:11:52 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:11:57 - mmengine - INFO - Epoch(train) [134][10/32] lr: 1.0000e-06 eta: 0:15:45 time: 0.5803 data_time: 0.1125 memory: 20518 loss: 0.0975 loss_ce: 0.0975 2023/03/03 00:12:02 - mmengine - INFO - Epoch(train) [134][20/32] lr: 1.0000e-06 eta: 0:15:41 time: 0.4454 data_time: 0.0015 memory: 19251 loss: 0.1104 loss_ce: 0.1104 2023/03/03 00:12:06 - mmengine - INFO - Epoch(train) [134][30/32] lr: 1.0000e-06 eta: 0:15:37 time: 0.4440 data_time: 0.0013 memory: 19196 loss: 0.1041 loss_ce: 0.1041 2023/03/03 00:12:07 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:12:12 - mmengine - INFO - Epoch(train) [135][10/32] lr: 1.0000e-06 eta: 0:15:32 time: 0.5412 data_time: 0.0737 memory: 19142 loss: 0.1070 loss_ce: 0.1070 2023/03/03 00:12:17 - mmengine - INFO - Epoch(train) [135][20/32] lr: 1.0000e-06 eta: 0:15:27 time: 0.4414 data_time: 0.0014 memory: 19142 loss: 0.1087 loss_ce: 0.1087 2023/03/03 00:12:21 - mmengine - INFO - Epoch(train) [135][30/32] lr: 1.0000e-06 eta: 0:15:23 time: 0.4419 data_time: 0.0013 memory: 19975 loss: 0.1017 loss_ce: 0.1017 2023/03/03 00:12:22 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:12:27 - mmengine - INFO - Epoch(train) [136][10/32] lr: 1.0000e-06 eta: 0:15:18 time: 0.5286 data_time: 0.0832 memory: 18739 loss: 0.1107 loss_ce: 0.1107 2023/03/03 00:12:32 - mmengine - INFO - Epoch(train) [136][20/32] lr: 1.0000e-06 eta: 0:15:13 time: 0.4558 data_time: 0.0015 memory: 18966 loss: 0.1017 loss_ce: 0.1017 2023/03/03 00:12:36 - mmengine - INFO - Epoch(train) [136][30/32] lr: 1.0000e-06 eta: 0:15:09 time: 0.4762 data_time: 0.0012 memory: 24875 loss: 0.0975 loss_ce: 0.0975 2023/03/03 00:12:37 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:12:42 - mmengine - INFO - Epoch(train) [137][10/32] lr: 1.0000e-06 eta: 0:15:04 time: 0.4999 data_time: 0.0396 memory: 18348 loss: 0.1014 loss_ce: 0.1014 2023/03/03 00:12:46 - mmengine - INFO - Epoch(train) [137][20/32] lr: 1.0000e-06 eta: 0:14:59 time: 0.4596 data_time: 0.0016 memory: 19751 loss: 0.0944 loss_ce: 0.0944 2023/03/03 00:12:51 - mmengine - INFO - Epoch(train) [137][30/32] lr: 1.0000e-06 eta: 0:14:55 time: 0.4084 data_time: 0.0012 memory: 17279 loss: 0.1274 loss_ce: 0.1274 2023/03/03 00:12:51 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:12:57 - mmengine - INFO - Epoch(train) [138][10/32] lr: 1.0000e-06 eta: 0:14:50 time: 0.5536 data_time: 0.1056 memory: 20073 loss: 0.1062 loss_ce: 0.1062 2023/03/03 00:13:01 - mmengine - INFO - Epoch(train) [138][20/32] lr: 1.0000e-06 eta: 0:14:45 time: 0.4376 data_time: 0.0017 memory: 17467 loss: 0.1174 loss_ce: 0.1174 2023/03/03 00:13:05 - mmengine - INFO - Epoch(train) [138][30/32] lr: 1.0000e-06 eta: 0:14:41 time: 0.4477 data_time: 0.0012 memory: 20172 loss: 0.1117 loss_ce: 0.1117 2023/03/03 00:13:06 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:13:12 - mmengine - INFO - Epoch(train) [139][10/32] lr: 1.0000e-06 eta: 0:14:36 time: 0.5521 data_time: 0.0904 memory: 21046 loss: 0.1094 loss_ce: 0.1094 2023/03/03 00:13:16 - mmengine - INFO - Epoch(train) [139][20/32] lr: 1.0000e-06 eta: 0:14:32 time: 0.4569 data_time: 0.0015 memory: 20832 loss: 0.1084 loss_ce: 0.1084 2023/03/03 00:13:21 - mmengine - INFO - Epoch(train) [139][30/32] lr: 1.0000e-06 eta: 0:14:27 time: 0.4418 data_time: 0.0013 memory: 25747 loss: 0.1095 loss_ce: 0.1095 2023/03/03 00:13:21 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:13:27 - mmengine - INFO - Epoch(train) [140][10/32] lr: 1.0000e-06 eta: 0:14:22 time: 0.5978 data_time: 0.0777 memory: 23710 loss: 0.0983 loss_ce: 0.0983 2023/03/03 00:13:32 - mmengine - INFO - Epoch(train) [140][20/32] lr: 1.0000e-06 eta: 0:14:18 time: 0.4135 data_time: 0.0016 memory: 21347 loss: 0.1013 loss_ce: 0.1013 2023/03/03 00:13:36 - mmengine - INFO - Epoch(train) [140][30/32] lr: 1.0000e-06 eta: 0:14:13 time: 0.4478 data_time: 0.0016 memory: 18779 loss: 0.1017 loss_ce: 0.1017 2023/03/03 00:13:37 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:13:50 - mmengine - INFO - Epoch(val) [140][10/63] eta: 0:01:10 time: 1.3223 data_time: 0.0033 memory: 16962 2023/03/03 00:14:14 - mmengine - INFO - Epoch(val) [140][20/63] eta: 0:01:21 time: 2.4507 data_time: 0.0004 memory: 1075 2023/03/03 00:14:37 - mmengine - INFO - Epoch(val) [140][30/63] eta: 0:01:06 time: 2.3006 data_time: 0.0005 memory: 1075 2023/03/03 00:14:49 - mmengine - INFO - Epoch(val) [140][40/63] eta: 0:00:41 time: 1.1279 data_time: 0.0004 memory: 1075 2023/03/03 00:15:10 - mmengine - INFO - Epoch(val) [140][50/63] eta: 0:00:24 time: 2.1450 data_time: 0.0004 memory: 1075 2023/03/03 00:15:23 - mmengine - INFO - Epoch(val) [140][60/63] eta: 0:00:05 time: 1.2850 data_time: 0.0003 memory: 1075 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.80, recall: 0.6129, precision: 0.7546, hmean: 0.6764 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.81, recall: 0.6115, precision: 0.7591, hmean: 0.6773 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.82, recall: 0.6100, precision: 0.7651, hmean: 0.6788 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.83, recall: 0.6081, precision: 0.7715, hmean: 0.6801 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.84, recall: 0.6057, precision: 0.7756, hmean: 0.6802 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.85, recall: 0.6013, precision: 0.7826, hmean: 0.6801 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.86, recall: 0.5975, precision: 0.7909, hmean: 0.6807 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.87, recall: 0.5965, precision: 0.8035, hmean: 0.6847 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.88, recall: 0.5903, precision: 0.8092, hmean: 0.6826 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.89, recall: 0.5859, precision: 0.8157, hmean: 0.6820 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.90, recall: 0.5806, precision: 0.8232, hmean: 0.6810 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.91, recall: 0.5729, precision: 0.8363, hmean: 0.6800 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.92, recall: 0.5633, precision: 0.8411, hmean: 0.6747 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.93, recall: 0.5522, precision: 0.8522, hmean: 0.6702 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.94, recall: 0.5441, precision: 0.8613, hmean: 0.6669 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.95, recall: 0.5325, precision: 0.8722, hmean: 0.6613 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.96, recall: 0.5200, precision: 0.8809, hmean: 0.6540 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.97, recall: 0.5007, precision: 0.8881, hmean: 0.6404 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.98, recall: 0.4786, precision: 0.8963, hmean: 0.6240 2023/03/03 00:16:25 - mmengine - INFO - text score threshold: 0.99, recall: 0.4391, precision: 0.9138, hmean: 0.5932 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.80, recall: 0.6596, precision: 0.8121, hmean: 0.7279 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.81, recall: 0.6572, precision: 0.8159, hmean: 0.7280 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.82, recall: 0.6553, precision: 0.8219, hmean: 0.7292 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.83, recall: 0.6519, precision: 0.8271, hmean: 0.7291 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.84, recall: 0.6490, precision: 0.8311, hmean: 0.7288 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.85, recall: 0.6442, precision: 0.8383, hmean: 0.7286 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.86, recall: 0.6384, precision: 0.8451, hmean: 0.7274 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.87, recall: 0.6351, precision: 0.8554, hmean: 0.7289 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.88, recall: 0.6273, precision: 0.8601, hmean: 0.7255 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.89, recall: 0.6211, precision: 0.8646, hmean: 0.7229 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.90, recall: 0.6139, precision: 0.8703, hmean: 0.7199 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.91, recall: 0.6023, precision: 0.8791, hmean: 0.7149 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.92, recall: 0.5912, precision: 0.8828, hmean: 0.7082 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.93, recall: 0.5778, precision: 0.8915, hmean: 0.7011 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.94, recall: 0.5667, precision: 0.8971, hmean: 0.6946 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.95, recall: 0.5518, precision: 0.9038, hmean: 0.6852 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.96, recall: 0.5364, precision: 0.9086, hmean: 0.6745 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.97, recall: 0.5147, precision: 0.9129, hmean: 0.6583 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.98, recall: 0.4887, precision: 0.9152, hmean: 0.6372 2023/03/03 00:16:36 - mmengine - INFO - text score threshold: 0.99, recall: 0.4449, precision: 0.9259, hmean: 0.6010 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.80, recall: 0.7169, precision: 0.8826, hmean: 0.7912 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.81, recall: 0.7126, precision: 0.8846, hmean: 0.7893 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.82, recall: 0.7092, precision: 0.8895, hmean: 0.7892 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.83, recall: 0.7053, precision: 0.8949, hmean: 0.7889 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.84, recall: 0.7015, precision: 0.8983, hmean: 0.7878 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.85, recall: 0.6948, precision: 0.9041, hmean: 0.7857 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.86, recall: 0.6870, precision: 0.9095, hmean: 0.7828 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.87, recall: 0.6803, precision: 0.9163, hmean: 0.7809 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.88, recall: 0.6707, precision: 0.9195, hmean: 0.7756 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.89, recall: 0.6625, precision: 0.9223, hmean: 0.7711 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.90, recall: 0.6529, precision: 0.9256, hmean: 0.7657 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.91, recall: 0.6370, precision: 0.9297, hmean: 0.7560 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.92, recall: 0.6235, precision: 0.9310, hmean: 0.7468 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.93, recall: 0.6076, precision: 0.9376, hmean: 0.7374 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.94, recall: 0.5941, precision: 0.9405, hmean: 0.7282 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.95, recall: 0.5768, precision: 0.9448, hmean: 0.7163 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.96, recall: 0.5604, precision: 0.9494, hmean: 0.7048 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.97, recall: 0.5359, precision: 0.9505, hmean: 0.6853 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.98, recall: 0.5084, precision: 0.9522, hmean: 0.6629 2023/03/03 00:16:46 - mmengine - INFO - text score threshold: 0.99, recall: 0.4612, precision: 0.9599, hmean: 0.6231 2023/03/03 00:16:46 - mmengine - INFO - Epoch(val) [140][63/63] generic/precision: 0.8035 generic/recall: 0.5965 generic/hmean: 0.6847 weak/precision: 0.8219 weak/recall: 0.6553 weak/hmean: 0.7292 strong/precision: 0.8826 strong/recall: 0.7169 strong/hmean: 0.7912 2023/03/03 00:16:52 - mmengine - INFO - Epoch(train) [141][10/32] lr: 1.0000e-06 eta: 0:14:08 time: 0.5459 data_time: 0.0697 memory: 17733 loss: 0.0968 loss_ce: 0.0968 2023/03/03 00:16:56 - mmengine - INFO - Epoch(train) [141][20/32] lr: 1.0000e-06 eta: 0:14:04 time: 0.4461 data_time: 0.0013 memory: 17137 loss: 0.1168 loss_ce: 0.1168 2023/03/03 00:17:01 - mmengine - INFO - Epoch(train) [141][30/32] lr: 1.0000e-06 eta: 0:14:00 time: 0.4800 data_time: 0.0017 memory: 22104 loss: 0.0930 loss_ce: 0.0930 2023/03/03 00:17:02 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:17:08 - mmengine - INFO - Epoch(train) [142][10/32] lr: 1.0000e-06 eta: 0:13:55 time: 0.5675 data_time: 0.0897 memory: 20297 loss: 0.1210 loss_ce: 0.1210 2023/03/03 00:17:12 - mmengine - INFO - Epoch(train) [142][20/32] lr: 1.0000e-06 eta: 0:13:50 time: 0.4572 data_time: 0.0018 memory: 19958 loss: 0.0993 loss_ce: 0.0993 2023/03/03 00:17:16 - mmengine - INFO - Epoch(train) [142][30/32] lr: 1.0000e-06 eta: 0:13:46 time: 0.4208 data_time: 0.0014 memory: 18083 loss: 0.1084 loss_ce: 0.1084 2023/03/03 00:17:17 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:17:23 - mmengine - INFO - Epoch(train) [143][10/32] lr: 1.0000e-06 eta: 0:13:41 time: 0.5800 data_time: 0.0885 memory: 19175 loss: 0.1082 loss_ce: 0.1082 2023/03/03 00:17:28 - mmengine - INFO - Epoch(train) [143][20/32] lr: 1.0000e-06 eta: 0:13:37 time: 0.4722 data_time: 0.0018 memory: 19344 loss: 0.1092 loss_ce: 0.1092 2023/03/03 00:17:32 - mmengine - INFO - Epoch(train) [143][30/32] lr: 1.0000e-06 eta: 0:13:32 time: 0.4362 data_time: 0.0015 memory: 24319 loss: 0.1015 loss_ce: 0.1015 2023/03/03 00:17:33 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:17:38 - mmengine - INFO - Epoch(train) [144][10/32] lr: 1.0000e-06 eta: 0:13:27 time: 0.5316 data_time: 0.0523 memory: 18597 loss: 0.0928 loss_ce: 0.0928 2023/03/03 00:17:43 - mmengine - INFO - Epoch(train) [144][20/32] lr: 1.0000e-06 eta: 0:13:23 time: 0.4646 data_time: 0.0015 memory: 18165 loss: 0.1083 loss_ce: 0.1083 2023/03/03 00:17:47 - mmengine - INFO - Epoch(train) [144][30/32] lr: 1.0000e-06 eta: 0:13:18 time: 0.4727 data_time: 0.0014 memory: 19522 loss: 0.0946 loss_ce: 0.0946 2023/03/03 00:17:48 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:17:53 - mmengine - INFO - Epoch(train) [145][10/32] lr: 1.0000e-06 eta: 0:13:13 time: 0.5418 data_time: 0.0191 memory: 18083 loss: 0.1007 loss_ce: 0.1007 2023/03/03 00:17:58 - mmengine - INFO - Epoch(train) [145][20/32] lr: 1.0000e-06 eta: 0:13:09 time: 0.4536 data_time: 0.0014 memory: 23623 loss: 0.1088 loss_ce: 0.1088 2023/03/03 00:18:02 - mmengine - INFO - Epoch(train) [145][30/32] lr: 1.0000e-06 eta: 0:13:04 time: 0.4068 data_time: 0.0015 memory: 17919 loss: 0.1049 loss_ce: 0.1049 2023/03/03 00:18:03 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:18:08 - mmengine - INFO - Epoch(train) [146][10/32] lr: 1.0000e-06 eta: 0:12:59 time: 0.5572 data_time: 0.0524 memory: 19958 loss: 0.1031 loss_ce: 0.1031 2023/03/03 00:18:13 - mmengine - INFO - Epoch(train) [146][20/32] lr: 1.0000e-06 eta: 0:12:55 time: 0.4602 data_time: 0.0015 memory: 19730 loss: 0.1032 loss_ce: 0.1032 2023/03/03 00:18:18 - mmengine - INFO - Epoch(train) [146][30/32] lr: 1.0000e-06 eta: 0:12:51 time: 0.4945 data_time: 0.0011 memory: 18940 loss: 0.0921 loss_ce: 0.0921 2023/03/03 00:18:18 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:18:23 - mmengine - INFO - Epoch(train) [147][10/32] lr: 1.0000e-06 eta: 0:12:45 time: 0.4949 data_time: 0.0286 memory: 18966 loss: 0.0968 loss_ce: 0.0968 2023/03/03 00:18:28 - mmengine - INFO - Epoch(train) [147][20/32] lr: 1.0000e-06 eta: 0:12:41 time: 0.4885 data_time: 0.0014 memory: 21276 loss: 0.1017 loss_ce: 0.1017 2023/03/03 00:18:33 - mmengine - INFO - Epoch(train) [147][30/32] lr: 1.0000e-06 eta: 0:12:37 time: 0.4371 data_time: 0.0011 memory: 19958 loss: 0.0978 loss_ce: 0.0978 2023/03/03 00:18:33 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:18:39 - mmengine - INFO - Epoch(train) [148][10/32] lr: 1.0000e-06 eta: 0:12:31 time: 0.5291 data_time: 0.0689 memory: 20331 loss: 0.0861 loss_ce: 0.0861 2023/03/03 00:18:43 - mmengine - INFO - Epoch(train) [148][20/32] lr: 1.0000e-06 eta: 0:12:27 time: 0.4548 data_time: 0.0013 memory: 24544 loss: 0.0953 loss_ce: 0.0953 2023/03/03 00:18:48 - mmengine - INFO - Epoch(train) [148][30/32] lr: 1.0000e-06 eta: 0:12:23 time: 0.4704 data_time: 0.0013 memory: 18165 loss: 0.0930 loss_ce: 0.0930 2023/03/03 00:18:49 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:18:54 - mmengine - INFO - Epoch(train) [149][10/32] lr: 1.0000e-06 eta: 0:12:18 time: 0.5721 data_time: 0.0329 memory: 24319 loss: 0.1010 loss_ce: 0.1010 2023/03/03 00:18:59 - mmengine - INFO - Epoch(train) [149][20/32] lr: 1.0000e-06 eta: 0:12:13 time: 0.4943 data_time: 0.0013 memory: 19053 loss: 0.0959 loss_ce: 0.0959 2023/03/03 00:19:04 - mmengine - INFO - Epoch(train) [149][30/32] lr: 1.0000e-06 eta: 0:12:09 time: 0.4232 data_time: 0.0012 memory: 19552 loss: 0.0986 loss_ce: 0.0986 2023/03/03 00:19:04 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:19:10 - mmengine - INFO - Epoch(train) [150][10/32] lr: 1.0000e-06 eta: 0:12:04 time: 0.5802 data_time: 0.0879 memory: 19318 loss: 0.0852 loss_ce: 0.0852 2023/03/03 00:19:15 - mmengine - INFO - Epoch(train) [150][20/32] lr: 1.0000e-06 eta: 0:11:59 time: 0.4816 data_time: 0.0015 memory: 21046 loss: 0.0911 loss_ce: 0.0911 2023/03/03 00:19:20 - mmengine - INFO - Epoch(train) [150][30/32] lr: 1.0000e-06 eta: 0:11:55 time: 0.4724 data_time: 0.0013 memory: 19751 loss: 0.0942 loss_ce: 0.0942 2023/03/03 00:19:20 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:19:34 - mmengine - INFO - Epoch(val) [150][10/63] eta: 0:01:11 time: 1.3560 data_time: 0.0039 memory: 15775 2023/03/03 00:19:57 - mmengine - INFO - Epoch(val) [150][20/63] eta: 0:01:20 time: 2.3728 data_time: 0.0005 memory: 1075 2023/03/03 00:20:20 - mmengine - INFO - Epoch(val) [150][30/63] eta: 0:01:05 time: 2.2318 data_time: 0.0004 memory: 1075 2023/03/03 00:20:32 - mmengine - INFO - Epoch(val) [150][40/63] eta: 0:00:41 time: 1.1955 data_time: 0.0003 memory: 1075 2023/03/03 00:20:54 - mmengine - INFO - Epoch(val) [150][50/63] eta: 0:00:24 time: 2.2299 data_time: 0.0003 memory: 1075 2023/03/03 00:21:07 - mmengine - INFO - Epoch(val) [150][60/63] eta: 0:00:05 time: 1.3243 data_time: 0.0005 memory: 1075 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.80, recall: 0.6086, precision: 0.7457, hmean: 0.6702 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.81, recall: 0.6086, precision: 0.7524, hmean: 0.6729 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.82, recall: 0.6076, precision: 0.7584, hmean: 0.6747 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.83, recall: 0.6057, precision: 0.7638, hmean: 0.6756 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.84, recall: 0.6018, precision: 0.7716, hmean: 0.6762 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.85, recall: 0.5999, precision: 0.7778, hmean: 0.6774 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.86, recall: 0.5985, precision: 0.7872, hmean: 0.6800 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.87, recall: 0.5936, precision: 0.7929, hmean: 0.6790 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.88, recall: 0.5893, precision: 0.7984, hmean: 0.6781 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.89, recall: 0.5864, precision: 0.8098, hmean: 0.6803 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.90, recall: 0.5787, precision: 0.8177, hmean: 0.6778 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.91, recall: 0.5725, precision: 0.8245, hmean: 0.6758 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.92, recall: 0.5652, precision: 0.8362, hmean: 0.6745 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.93, recall: 0.5527, precision: 0.8472, hmean: 0.6690 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.94, recall: 0.5416, precision: 0.8568, hmean: 0.6637 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.95, recall: 0.5282, precision: 0.8693, hmean: 0.6571 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.96, recall: 0.5161, precision: 0.8773, hmean: 0.6499 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.97, recall: 0.5017, precision: 0.8861, hmean: 0.6406 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.98, recall: 0.4786, precision: 0.8987, hmean: 0.6246 2023/03/03 00:22:08 - mmengine - INFO - text score threshold: 0.99, recall: 0.4434, precision: 0.9164, hmean: 0.5977 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.80, recall: 0.6558, precision: 0.8035, hmean: 0.7222 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.81, recall: 0.6543, precision: 0.8089, hmean: 0.7234 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.82, recall: 0.6524, precision: 0.8143, hmean: 0.7244 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.83, recall: 0.6495, precision: 0.8191, hmean: 0.7245 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.84, recall: 0.6447, precision: 0.8265, hmean: 0.7244 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.85, recall: 0.6423, precision: 0.8327, hmean: 0.7252 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.86, recall: 0.6399, precision: 0.8417, hmean: 0.7270 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.87, recall: 0.6346, precision: 0.8476, hmean: 0.7258 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.88, recall: 0.6283, precision: 0.8513, hmean: 0.7230 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.89, recall: 0.6235, precision: 0.8610, hmean: 0.7233 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.90, recall: 0.6134, precision: 0.8667, hmean: 0.7184 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.91, recall: 0.6042, precision: 0.8703, hmean: 0.7133 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.92, recall: 0.5932, precision: 0.8775, hmean: 0.7078 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.93, recall: 0.5778, precision: 0.8856, hmean: 0.6993 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.94, recall: 0.5638, precision: 0.8919, hmean: 0.6909 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.95, recall: 0.5474, precision: 0.9010, hmean: 0.6810 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.96, recall: 0.5344, precision: 0.9083, hmean: 0.6729 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.97, recall: 0.5161, precision: 0.9116, hmean: 0.6591 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.98, recall: 0.4877, precision: 0.9159, hmean: 0.6365 2023/03/03 00:22:17 - mmengine - INFO - text score threshold: 0.99, recall: 0.4473, precision: 0.9244, hmean: 0.6029 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.80, recall: 0.7150, precision: 0.8761, hmean: 0.7874 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.81, recall: 0.7126, precision: 0.8810, hmean: 0.7879 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.82, recall: 0.7078, precision: 0.8834, hmean: 0.7859 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.83, recall: 0.7039, precision: 0.8877, hmean: 0.7852 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.84, recall: 0.6976, precision: 0.8944, hmean: 0.7839 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.85, recall: 0.6948, precision: 0.9007, hmean: 0.7845 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.86, recall: 0.6890, precision: 0.9063, hmean: 0.7828 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.87, recall: 0.6822, precision: 0.9113, hmean: 0.7803 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.88, recall: 0.6745, precision: 0.9139, hmean: 0.7762 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.89, recall: 0.6663, precision: 0.9202, hmean: 0.7730 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.90, recall: 0.6543, precision: 0.9245, hmean: 0.7663 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.91, recall: 0.6428, precision: 0.9258, hmean: 0.7587 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.92, recall: 0.6288, precision: 0.9302, hmean: 0.7504 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.93, recall: 0.6091, precision: 0.9336, hmean: 0.7372 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.94, recall: 0.5936, precision: 0.9391, hmean: 0.7274 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.95, recall: 0.5720, precision: 0.9414, hmean: 0.7116 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.96, recall: 0.5575, precision: 0.9476, hmean: 0.7020 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.97, recall: 0.5383, precision: 0.9507, hmean: 0.6874 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.98, recall: 0.5060, precision: 0.9503, hmean: 0.6604 2023/03/03 00:22:26 - mmengine - INFO - text score threshold: 0.99, recall: 0.4651, precision: 0.9612, hmean: 0.6269 2023/03/03 00:22:26 - mmengine - INFO - Epoch(val) [150][63/63] generic/precision: 0.8098 generic/recall: 0.5864 generic/hmean: 0.6803 weak/precision: 0.8417 weak/recall: 0.6399 weak/hmean: 0.7270 strong/precision: 0.8810 strong/recall: 0.7126 strong/hmean: 0.7879 2023/03/03 00:22:32 - mmengine - INFO - Epoch(train) [151][10/32] lr: 1.0000e-06 eta: 0:11:50 time: 0.5540 data_time: 0.0714 memory: 20766 loss: 0.1169 loss_ce: 0.1169 2023/03/03 00:22:37 - mmengine - INFO - Epoch(train) [151][20/32] lr: 1.0000e-06 eta: 0:11:46 time: 0.4598 data_time: 0.0013 memory: 18511 loss: 0.0912 loss_ce: 0.0912 2023/03/03 00:22:41 - mmengine - INFO - Epoch(train) [151][30/32] lr: 1.0000e-06 eta: 0:11:41 time: 0.4466 data_time: 0.0012 memory: 23984 loss: 0.0939 loss_ce: 0.0939 2023/03/03 00:22:42 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:22:47 - mmengine - INFO - Epoch(train) [152][10/32] lr: 1.0000e-06 eta: 0:11:36 time: 0.5186 data_time: 0.0891 memory: 18325 loss: 0.1045 loss_ce: 0.1045 2023/03/03 00:22:52 - mmengine - INFO - Epoch(train) [152][20/32] lr: 1.0000e-06 eta: 0:11:31 time: 0.4756 data_time: 0.0014 memory: 19558 loss: 0.0907 loss_ce: 0.0907 2023/03/03 00:22:56 - mmengine - INFO - Epoch(train) [152][30/32] lr: 1.0000e-06 eta: 0:11:27 time: 0.4407 data_time: 0.0013 memory: 25455 loss: 0.1099 loss_ce: 0.1099 2023/03/03 00:22:56 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:23:02 - mmengine - INFO - Epoch(train) [153][10/32] lr: 1.0000e-06 eta: 0:11:22 time: 0.5448 data_time: 0.0499 memory: 20172 loss: 0.1023 loss_ce: 0.1023 2023/03/03 00:23:07 - mmengine - INFO - Epoch(train) [153][20/32] lr: 1.0000e-06 eta: 0:11:17 time: 0.4623 data_time: 0.0014 memory: 18830 loss: 0.1114 loss_ce: 0.1114 2023/03/03 00:23:11 - mmengine - INFO - Epoch(train) [153][30/32] lr: 1.0000e-06 eta: 0:11:13 time: 0.4195 data_time: 0.0012 memory: 20655 loss: 0.0966 loss_ce: 0.0966 2023/03/03 00:23:11 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:23:17 - mmengine - INFO - Epoch(train) [154][10/32] lr: 1.0000e-06 eta: 0:11:08 time: 0.5661 data_time: 0.0814 memory: 24544 loss: 0.1054 loss_ce: 0.1054 2023/03/03 00:23:22 - mmengine - INFO - Epoch(train) [154][20/32] lr: 1.0000e-06 eta: 0:11:03 time: 0.4519 data_time: 0.0015 memory: 22104 loss: 0.1143 loss_ce: 0.1143 2023/03/03 00:23:26 - mmengine - INFO - Epoch(train) [154][30/32] lr: 1.0000e-06 eta: 0:10:59 time: 0.4619 data_time: 0.0013 memory: 24319 loss: 0.1015 loss_ce: 0.1015 2023/03/03 00:23:27 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:23:32 - mmengine - INFO - Epoch(train) [155][10/32] lr: 1.0000e-06 eta: 0:10:54 time: 0.5428 data_time: 0.0756 memory: 20172 loss: 0.0959 loss_ce: 0.0959 2023/03/03 00:23:37 - mmengine - INFO - Epoch(train) [155][20/32] lr: 1.0000e-06 eta: 0:10:49 time: 0.4553 data_time: 0.0014 memory: 18884 loss: 0.1099 loss_ce: 0.1099 2023/03/03 00:23:41 - mmengine - INFO - Epoch(train) [155][30/32] lr: 1.0000e-06 eta: 0:10:45 time: 0.4568 data_time: 0.0011 memory: 17919 loss: 0.1119 loss_ce: 0.1119 2023/03/03 00:23:42 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:23:47 - mmengine - INFO - Epoch(train) [156][10/32] lr: 1.0000e-06 eta: 0:10:40 time: 0.5420 data_time: 0.0726 memory: 19751 loss: 0.1089 loss_ce: 0.1089 2023/03/03 00:23:52 - mmengine - INFO - Epoch(train) [156][20/32] lr: 1.0000e-06 eta: 0:10:35 time: 0.5040 data_time: 0.0012 memory: 17733 loss: 0.1101 loss_ce: 0.1101 2023/03/03 00:23:57 - mmengine - INFO - Epoch(train) [156][30/32] lr: 1.0000e-06 eta: 0:10:31 time: 0.4829 data_time: 0.0011 memory: 22984 loss: 0.1012 loss_ce: 0.1012 2023/03/03 00:23:58 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:24:03 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:24:03 - mmengine - INFO - Epoch(train) [157][10/32] lr: 1.0000e-06 eta: 0:10:26 time: 0.5445 data_time: 0.0604 memory: 19958 loss: 0.1128 loss_ce: 0.1128 2023/03/03 00:24:08 - mmengine - INFO - Epoch(train) [157][20/32] lr: 1.0000e-06 eta: 0:10:21 time: 0.4858 data_time: 0.0014 memory: 24319 loss: 0.0982 loss_ce: 0.0982 2023/03/03 00:24:13 - mmengine - INFO - Epoch(train) [157][30/32] lr: 1.0000e-06 eta: 0:10:17 time: 0.4476 data_time: 0.0013 memory: 23984 loss: 0.0888 loss_ce: 0.0888 2023/03/03 00:24:13 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:24:19 - mmengine - INFO - Epoch(train) [158][10/32] lr: 1.0000e-06 eta: 0:10:12 time: 0.5748 data_time: 0.1340 memory: 20820 loss: 0.0930 loss_ce: 0.0930 2023/03/03 00:24:24 - mmengine - INFO - Epoch(train) [158][20/32] lr: 1.0000e-06 eta: 0:10:07 time: 0.4532 data_time: 0.0014 memory: 18660 loss: 0.0920 loss_ce: 0.0920 2023/03/03 00:24:28 - mmengine - INFO - Epoch(train) [158][30/32] lr: 1.0000e-06 eta: 0:10:03 time: 0.4340 data_time: 0.0011 memory: 19751 loss: 0.0902 loss_ce: 0.0902 2023/03/03 00:24:29 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:24:34 - mmengine - INFO - Epoch(train) [159][10/32] lr: 1.0000e-06 eta: 0:09:58 time: 0.5353 data_time: 0.0850 memory: 18413 loss: 0.1078 loss_ce: 0.1078 2023/03/03 00:24:39 - mmengine - INFO - Epoch(train) [159][20/32] lr: 1.0000e-06 eta: 0:09:53 time: 0.4937 data_time: 0.0014 memory: 22486 loss: 0.0913 loss_ce: 0.0913 2023/03/03 00:24:43 - mmengine - INFO - Epoch(train) [159][30/32] lr: 1.0000e-06 eta: 0:09:49 time: 0.4493 data_time: 0.0012 memory: 19958 loss: 0.1028 loss_ce: 0.1028 2023/03/03 00:24:44 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:24:50 - mmengine - INFO - Epoch(train) [160][10/32] lr: 1.0000e-06 eta: 0:09:44 time: 0.5698 data_time: 0.0647 memory: 20204 loss: 0.1024 loss_ce: 0.1024 2023/03/03 00:24:54 - mmengine - INFO - Epoch(train) [160][20/32] lr: 1.0000e-06 eta: 0:09:39 time: 0.4580 data_time: 0.0015 memory: 19958 loss: 0.1087 loss_ce: 0.1087 2023/03/03 00:24:59 - mmengine - INFO - Epoch(train) [160][30/32] lr: 1.0000e-06 eta: 0:09:35 time: 0.4304 data_time: 0.0014 memory: 19552 loss: 0.1020 loss_ce: 0.1020 2023/03/03 00:24:59 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:25:13 - mmengine - INFO - Epoch(val) [160][10/63] eta: 0:01:12 time: 1.3728 data_time: 0.0034 memory: 17302 2023/03/03 00:25:38 - mmengine - INFO - Epoch(val) [160][20/63] eta: 0:01:22 time: 2.4566 data_time: 0.0004 memory: 1075 2023/03/03 00:26:01 - mmengine - INFO - Epoch(val) [160][30/63] eta: 0:01:07 time: 2.2851 data_time: 0.0004 memory: 1075 2023/03/03 00:26:13 - mmengine - INFO - Epoch(val) [160][40/63] eta: 0:00:42 time: 1.2136 data_time: 0.0004 memory: 1075 2023/03/03 00:26:34 - mmengine - INFO - Epoch(val) [160][50/63] eta: 0:00:24 time: 2.1449 data_time: 0.0003 memory: 1075 2023/03/03 00:26:47 - mmengine - INFO - Epoch(val) [160][60/63] eta: 0:00:05 time: 1.3025 data_time: 0.0004 memory: 1075 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.80, recall: 0.6139, precision: 0.7478, hmean: 0.6742 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.81, recall: 0.6119, precision: 0.7521, hmean: 0.6748 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.82, recall: 0.6110, precision: 0.7554, hmean: 0.6755 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.83, recall: 0.6091, precision: 0.7616, hmean: 0.6768 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.84, recall: 0.6066, precision: 0.7688, hmean: 0.6781 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.85, recall: 0.6052, precision: 0.7798, hmean: 0.6815 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.86, recall: 0.6013, precision: 0.7920, hmean: 0.6836 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.87, recall: 0.5970, precision: 0.7990, hmean: 0.6834 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.88, recall: 0.5917, precision: 0.8086, hmean: 0.6833 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.89, recall: 0.5879, precision: 0.8156, hmean: 0.6833 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.90, recall: 0.5792, precision: 0.8234, hmean: 0.6800 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.91, recall: 0.5744, precision: 0.8302, hmean: 0.6790 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.92, recall: 0.5662, precision: 0.8394, hmean: 0.6763 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.93, recall: 0.5571, precision: 0.8501, hmean: 0.6731 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.94, recall: 0.5460, precision: 0.8610, hmean: 0.6682 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.95, recall: 0.5344, precision: 0.8692, hmean: 0.6619 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.96, recall: 0.5224, precision: 0.8764, hmean: 0.6546 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.97, recall: 0.5070, precision: 0.8886, hmean: 0.6456 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.98, recall: 0.4800, precision: 0.8950, hmean: 0.6249 2023/03/03 00:27:48 - mmengine - INFO - text score threshold: 0.99, recall: 0.4473, precision: 0.9144, hmean: 0.6007 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.80, recall: 0.6591, precision: 0.8029, hmean: 0.7240 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.81, recall: 0.6572, precision: 0.8077, hmean: 0.7247 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.82, recall: 0.6562, precision: 0.8113, hmean: 0.7256 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.83, recall: 0.6529, precision: 0.8164, hmean: 0.7255 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.84, recall: 0.6505, precision: 0.8243, hmean: 0.7271 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.85, recall: 0.6471, precision: 0.8337, hmean: 0.7287 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.86, recall: 0.6408, precision: 0.8440, hmean: 0.7285 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.87, recall: 0.6355, precision: 0.8505, hmean: 0.7275 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.88, recall: 0.6283, precision: 0.8586, hmean: 0.7256 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.89, recall: 0.6230, precision: 0.8644, hmean: 0.7241 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.90, recall: 0.6129, precision: 0.8713, hmean: 0.7196 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.91, recall: 0.6066, precision: 0.8768, hmean: 0.7171 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.92, recall: 0.5951, precision: 0.8822, hmean: 0.7108 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.93, recall: 0.5826, precision: 0.8891, hmean: 0.7039 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.94, recall: 0.5672, precision: 0.8945, hmean: 0.6942 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.95, recall: 0.5537, precision: 0.9005, hmean: 0.6857 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.96, recall: 0.5392, precision: 0.9047, hmean: 0.6757 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.97, recall: 0.5205, precision: 0.9122, hmean: 0.6628 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.98, recall: 0.4906, precision: 0.9147, hmean: 0.6387 2023/03/03 00:27:58 - mmengine - INFO - text score threshold: 0.99, recall: 0.4545, precision: 0.9291, hmean: 0.6104 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.80, recall: 0.7183, precision: 0.8751, hmean: 0.7890 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.81, recall: 0.7155, precision: 0.8793, hmean: 0.7890 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.82, recall: 0.7126, precision: 0.8810, hmean: 0.7879 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.83, recall: 0.7082, precision: 0.8856, hmean: 0.7871 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.84, recall: 0.7039, precision: 0.8920, hmean: 0.7869 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.85, recall: 0.6981, precision: 0.8995, hmean: 0.7861 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.86, recall: 0.6899, precision: 0.9087, hmean: 0.7843 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.87, recall: 0.6822, precision: 0.9130, hmean: 0.7809 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.88, recall: 0.6726, precision: 0.9191, hmean: 0.7768 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.89, recall: 0.6644, precision: 0.9218, hmean: 0.7722 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.90, recall: 0.6529, precision: 0.9281, hmean: 0.7665 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.91, recall: 0.6442, precision: 0.9311, hmean: 0.7615 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.92, recall: 0.6298, precision: 0.9336, hmean: 0.7522 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.93, recall: 0.6148, precision: 0.9383, hmean: 0.7429 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.94, recall: 0.5956, precision: 0.9393, hmean: 0.7289 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.95, recall: 0.5797, precision: 0.9428, hmean: 0.7179 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.96, recall: 0.5643, precision: 0.9467, hmean: 0.7071 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.97, recall: 0.5431, precision: 0.9519, hmean: 0.6916 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.98, recall: 0.5104, precision: 0.9515, hmean: 0.6644 2023/03/03 00:28:07 - mmengine - INFO - text score threshold: 0.99, recall: 0.4714, precision: 0.9636, hmean: 0.6330 2023/03/03 00:28:07 - mmengine - INFO - Epoch(val) [160][63/63] generic/precision: 0.7920 generic/recall: 0.6013 generic/hmean: 0.6836 weak/precision: 0.8337 weak/recall: 0.6471 weak/hmean: 0.7287 strong/precision: 0.8751 strong/recall: 0.7183 strong/hmean: 0.7890 2023/03/03 00:28:13 - mmengine - INFO - Epoch(train) [161][10/32] lr: 1.0000e-06 eta: 0:09:29 time: 0.5604 data_time: 0.0431 memory: 19958 loss: 0.1018 loss_ce: 0.1018 2023/03/03 00:28:17 - mmengine - INFO - Epoch(train) [161][20/32] lr: 1.0000e-06 eta: 0:09:25 time: 0.4795 data_time: 0.0013 memory: 19266 loss: 0.0872 loss_ce: 0.0872 2023/03/03 00:28:22 - mmengine - INFO - Epoch(train) [161][30/32] lr: 1.0000e-06 eta: 0:09:21 time: 0.4382 data_time: 0.0013 memory: 18184 loss: 0.0891 loss_ce: 0.0891 2023/03/03 00:28:22 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:28:28 - mmengine - INFO - Epoch(train) [162][10/32] lr: 1.0000e-06 eta: 0:09:15 time: 0.5758 data_time: 0.0775 memory: 19344 loss: 0.0893 loss_ce: 0.0893 2023/03/03 00:28:32 - mmengine - INFO - Epoch(train) [162][20/32] lr: 1.0000e-06 eta: 0:09:11 time: 0.4475 data_time: 0.0013 memory: 24319 loss: 0.1062 loss_ce: 0.1062 2023/03/03 00:28:37 - mmengine - INFO - Epoch(train) [162][30/32] lr: 1.0000e-06 eta: 0:09:06 time: 0.4259 data_time: 0.0013 memory: 18128 loss: 0.0932 loss_ce: 0.0932 2023/03/03 00:28:37 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:28:43 - mmengine - INFO - Epoch(train) [163][10/32] lr: 1.0000e-06 eta: 0:09:01 time: 0.5291 data_time: 0.0754 memory: 18413 loss: 0.0911 loss_ce: 0.0911 2023/03/03 00:28:47 - mmengine - INFO - Epoch(train) [163][20/32] lr: 1.0000e-06 eta: 0:08:57 time: 0.4385 data_time: 0.0013 memory: 17810 loss: 0.0970 loss_ce: 0.0970 2023/03/03 00:28:51 - mmengine - INFO - Epoch(train) [163][30/32] lr: 1.0000e-06 eta: 0:08:52 time: 0.4419 data_time: 0.0013 memory: 19958 loss: 0.1078 loss_ce: 0.1078 2023/03/03 00:28:52 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:28:58 - mmengine - INFO - Epoch(train) [164][10/32] lr: 1.0000e-06 eta: 0:08:47 time: 0.6105 data_time: 0.0600 memory: 22545 loss: 0.0907 loss_ce: 0.0907 2023/03/03 00:29:03 - mmengine - INFO - Epoch(train) [164][20/32] lr: 1.0000e-06 eta: 0:08:42 time: 0.4745 data_time: 0.0014 memory: 19958 loss: 0.1034 loss_ce: 0.1034 2023/03/03 00:29:07 - mmengine - INFO - Epoch(train) [164][30/32] lr: 1.0000e-06 eta: 0:08:38 time: 0.4634 data_time: 0.0013 memory: 21985 loss: 0.1028 loss_ce: 0.1028 2023/03/03 00:29:08 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:29:13 - mmengine - INFO - Epoch(train) [165][10/32] lr: 1.0000e-06 eta: 0:08:33 time: 0.5563 data_time: 0.0787 memory: 20832 loss: 0.1152 loss_ce: 0.1152 2023/03/03 00:29:18 - mmengine - INFO - Epoch(train) [165][20/32] lr: 1.0000e-06 eta: 0:08:28 time: 0.4530 data_time: 0.0015 memory: 18850 loss: 0.1038 loss_ce: 0.1038 2023/03/03 00:29:22 - mmengine - INFO - Epoch(train) [165][30/32] lr: 1.0000e-06 eta: 0:08:24 time: 0.4287 data_time: 0.0013 memory: 19958 loss: 0.1103 loss_ce: 0.1103 2023/03/03 00:29:23 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:29:28 - mmengine - INFO - Epoch(train) [166][10/32] lr: 1.0000e-06 eta: 0:08:19 time: 0.5526 data_time: 0.0511 memory: 17961 loss: 0.1016 loss_ce: 0.1016 2023/03/03 00:29:33 - mmengine - INFO - Epoch(train) [166][20/32] lr: 1.0000e-06 eta: 0:08:14 time: 0.4695 data_time: 0.0015 memory: 18834 loss: 0.0958 loss_ce: 0.0958 2023/03/03 00:29:37 - mmengine - INFO - Epoch(train) [166][30/32] lr: 1.0000e-06 eta: 0:08:10 time: 0.4276 data_time: 0.0013 memory: 19346 loss: 0.0927 loss_ce: 0.0927 2023/03/03 00:29:38 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:29:44 - mmengine - INFO - Epoch(train) [167][10/32] lr: 1.0000e-06 eta: 0:08:04 time: 0.5628 data_time: 0.1040 memory: 20820 loss: 0.1025 loss_ce: 0.1025 2023/03/03 00:29:48 - mmengine - INFO - Epoch(train) [167][20/32] lr: 1.0000e-06 eta: 0:08:00 time: 0.4558 data_time: 0.0013 memory: 19751 loss: 0.0957 loss_ce: 0.0957 2023/03/03 00:29:53 - mmengine - INFO - Epoch(train) [167][30/32] lr: 1.0000e-06 eta: 0:07:55 time: 0.4675 data_time: 0.0013 memory: 19958 loss: 0.1139 loss_ce: 0.1139 2023/03/03 00:29:54 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:29:59 - mmengine - INFO - Epoch(train) [168][10/32] lr: 1.0000e-06 eta: 0:07:50 time: 0.5735 data_time: 0.0869 memory: 23088 loss: 0.0886 loss_ce: 0.0886 2023/03/03 00:30:04 - mmengine - INFO - Epoch(train) [168][20/32] lr: 1.0000e-06 eta: 0:07:46 time: 0.4688 data_time: 0.0015 memory: 19958 loss: 0.0957 loss_ce: 0.0957 2023/03/03 00:30:09 - mmengine - INFO - Epoch(train) [168][30/32] lr: 1.0000e-06 eta: 0:07:41 time: 0.4685 data_time: 0.0014 memory: 19142 loss: 0.0991 loss_ce: 0.0991 2023/03/03 00:30:09 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:30:15 - mmengine - INFO - Epoch(train) [169][10/32] lr: 1.0000e-06 eta: 0:07:36 time: 0.5422 data_time: 0.0608 memory: 22984 loss: 0.0840 loss_ce: 0.0840 2023/03/03 00:30:20 - mmengine - INFO - Epoch(train) [169][20/32] lr: 1.0000e-06 eta: 0:07:32 time: 0.4688 data_time: 0.0019 memory: 24875 loss: 0.0991 loss_ce: 0.0991 2023/03/03 00:30:24 - mmengine - INFO - Epoch(train) [169][30/32] lr: 1.0000e-06 eta: 0:07:27 time: 0.4875 data_time: 0.0019 memory: 19142 loss: 0.0958 loss_ce: 0.0958 2023/03/03 00:30:25 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:30:30 - mmengine - INFO - Epoch(train) [170][10/32] lr: 1.0000e-06 eta: 0:07:22 time: 0.5436 data_time: 0.0578 memory: 19552 loss: 0.0906 loss_ce: 0.0906 2023/03/03 00:30:35 - mmengine - INFO - Epoch(train) [170][20/32] lr: 1.0000e-06 eta: 0:07:17 time: 0.4839 data_time: 0.0013 memory: 21688 loss: 0.0916 loss_ce: 0.0916 2023/03/03 00:30:40 - mmengine - INFO - Epoch(train) [170][30/32] lr: 1.0000e-06 eta: 0:07:13 time: 0.4451 data_time: 0.0013 memory: 24229 loss: 0.1144 loss_ce: 0.1144 2023/03/03 00:30:40 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:30:55 - mmengine - INFO - Epoch(val) [170][10/63] eta: 0:01:16 time: 1.4517 data_time: 0.0034 memory: 18413 2023/03/03 00:31:19 - mmengine - INFO - Epoch(val) [170][20/63] eta: 0:01:22 time: 2.3979 data_time: 0.0004 memory: 1075 2023/03/03 00:31:41 - mmengine - INFO - Epoch(val) [170][30/63] eta: 0:01:07 time: 2.2666 data_time: 0.0005 memory: 1075 2023/03/03 00:31:52 - mmengine - INFO - Epoch(val) [170][40/63] eta: 0:00:41 time: 1.0340 data_time: 0.0003 memory: 1075 2023/03/03 00:32:13 - mmengine - INFO - Epoch(val) [170][50/63] eta: 0:00:24 time: 2.0886 data_time: 0.0004 memory: 1075 2023/03/03 00:32:25 - mmengine - INFO - Epoch(val) [170][60/63] eta: 0:00:05 time: 1.2512 data_time: 0.0004 memory: 1075 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.80, recall: 0.6110, precision: 0.7430, hmean: 0.6705 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.81, recall: 0.6105, precision: 0.7494, hmean: 0.6729 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.82, recall: 0.6095, precision: 0.7585, hmean: 0.6759 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.83, recall: 0.6071, precision: 0.7661, hmean: 0.6774 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.84, recall: 0.6062, precision: 0.7729, hmean: 0.6794 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.85, recall: 0.6023, precision: 0.7804, hmean: 0.6799 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.86, recall: 0.6013, precision: 0.7920, hmean: 0.6836 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.87, recall: 0.5980, precision: 0.8013, hmean: 0.6849 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.88, recall: 0.5961, precision: 0.8102, hmean: 0.6868 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.89, recall: 0.5912, precision: 0.8187, hmean: 0.6866 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.90, recall: 0.5845, precision: 0.8236, hmean: 0.6838 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.91, recall: 0.5782, precision: 0.8329, hmean: 0.6826 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.92, recall: 0.5657, precision: 0.8417, hmean: 0.6766 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.93, recall: 0.5561, precision: 0.8499, hmean: 0.6723 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.94, recall: 0.5436, precision: 0.8592, hmean: 0.6659 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.95, recall: 0.5301, precision: 0.8717, hmean: 0.6593 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.96, recall: 0.5195, precision: 0.8758, hmean: 0.6522 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.97, recall: 0.5026, precision: 0.8825, hmean: 0.6405 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.98, recall: 0.4810, precision: 0.8936, hmean: 0.6254 2023/03/03 00:33:27 - mmengine - INFO - text score threshold: 0.99, recall: 0.4444, precision: 0.9130, hmean: 0.5978 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.80, recall: 0.6591, precision: 0.8015, hmean: 0.7234 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.81, recall: 0.6577, precision: 0.8073, hmean: 0.7249 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.82, recall: 0.6558, precision: 0.8161, hmean: 0.7272 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.83, recall: 0.6529, precision: 0.8238, hmean: 0.7284 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.84, recall: 0.6514, precision: 0.8306, hmean: 0.7302 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.85, recall: 0.6456, precision: 0.8366, hmean: 0.7288 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.86, recall: 0.6432, precision: 0.8472, hmean: 0.7313 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.87, recall: 0.6389, precision: 0.8561, hmean: 0.7317 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.88, recall: 0.6346, precision: 0.8626, hmean: 0.7312 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.89, recall: 0.6283, precision: 0.8700, hmean: 0.7297 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.90, recall: 0.6201, precision: 0.8738, hmean: 0.7254 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.91, recall: 0.6110, precision: 0.8800, hmean: 0.7212 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.92, recall: 0.5965, precision: 0.8875, hmean: 0.7135 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.93, recall: 0.5835, precision: 0.8918, hmean: 0.7055 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.94, recall: 0.5667, precision: 0.8957, hmean: 0.6942 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.95, recall: 0.5498, precision: 0.9042, hmean: 0.6838 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.96, recall: 0.5383, precision: 0.9075, hmean: 0.6757 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.97, recall: 0.5190, precision: 0.9112, hmean: 0.6613 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.98, recall: 0.4921, precision: 0.9141, hmean: 0.6397 2023/03/03 00:33:37 - mmengine - INFO - text score threshold: 0.99, recall: 0.4506, precision: 0.9258, hmean: 0.6062 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.80, recall: 0.7155, precision: 0.8700, hmean: 0.7852 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.81, recall: 0.7135, precision: 0.8759, hmean: 0.7864 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.82, recall: 0.7097, precision: 0.8832, hmean: 0.7870 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.83, recall: 0.7044, precision: 0.8888, hmean: 0.7859 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.84, recall: 0.7015, precision: 0.8944, hmean: 0.7863 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.85, recall: 0.6943, precision: 0.8996, hmean: 0.7837 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.86, recall: 0.6890, precision: 0.9074, hmean: 0.7833 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.87, recall: 0.6832, precision: 0.9155, hmean: 0.7825 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.88, recall: 0.6774, precision: 0.9208, hmean: 0.7806 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.89, recall: 0.6683, precision: 0.9253, hmean: 0.7761 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.90, recall: 0.6591, precision: 0.9288, hmean: 0.7711 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.91, recall: 0.6476, precision: 0.9327, hmean: 0.7644 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.92, recall: 0.6302, precision: 0.9377, hmean: 0.7538 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.93, recall: 0.6148, precision: 0.9397, hmean: 0.7433 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.94, recall: 0.5970, precision: 0.9437, hmean: 0.7313 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.95, recall: 0.5763, precision: 0.9477, hmean: 0.7168 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.96, recall: 0.5623, precision: 0.9481, hmean: 0.7060 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.97, recall: 0.5421, precision: 0.9518, hmean: 0.6908 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.98, recall: 0.5128, precision: 0.9526, hmean: 0.6667 2023/03/03 00:33:47 - mmengine - INFO - text score threshold: 0.99, recall: 0.4680, precision: 0.9614, hmean: 0.6295 2023/03/03 00:33:47 - mmengine - INFO - Epoch(val) [170][63/63] generic/precision: 0.8102 generic/recall: 0.5961 generic/hmean: 0.6868 weak/precision: 0.8561 weak/recall: 0.6389 weak/hmean: 0.7317 strong/precision: 0.8832 strong/recall: 0.7097 strong/hmean: 0.7870 2023/03/03 00:33:47 - mmengine - INFO - The previous best checkpoint mmocr/projects/SPTS/work_dirs/spts_resnet50_350e_icdar2015/best_generic/hmean_epoch_90.pth is removed 2023/03/03 00:33:49 - mmengine - INFO - The best checkpoint with 0.6868 generic/hmean at 170 epoch is saved to best_generic/hmean_epoch_170.pth. 2023/03/03 00:33:54 - mmengine - INFO - Epoch(train) [171][10/32] lr: 1.0000e-06 eta: 0:07:07 time: 0.5044 data_time: 0.0612 memory: 18940 loss: 0.1121 loss_ce: 0.1121 2023/03/03 00:33:59 - mmengine - INFO - Epoch(train) [171][20/32] lr: 1.0000e-06 eta: 0:07:03 time: 0.4593 data_time: 0.0014 memory: 21046 loss: 0.1054 loss_ce: 0.1054 2023/03/03 00:34:03 - mmengine - INFO - Epoch(train) [171][30/32] lr: 1.0000e-06 eta: 0:06:58 time: 0.4594 data_time: 0.0014 memory: 19373 loss: 0.1123 loss_ce: 0.1123 2023/03/03 00:34:04 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:34:10 - mmengine - INFO - Epoch(train) [172][10/32] lr: 1.0000e-06 eta: 0:06:53 time: 0.5950 data_time: 0.1211 memory: 22984 loss: 0.0921 loss_ce: 0.0921 2023/03/03 00:34:14 - mmengine - INFO - Epoch(train) [172][20/32] lr: 1.0000e-06 eta: 0:06:49 time: 0.4219 data_time: 0.0016 memory: 17919 loss: 0.1089 loss_ce: 0.1089 2023/03/03 00:34:18 - mmengine - INFO - Epoch(train) [172][30/32] lr: 1.0000e-06 eta: 0:06:44 time: 0.4438 data_time: 0.0015 memory: 19971 loss: 0.1001 loss_ce: 0.1001 2023/03/03 00:34:19 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:34:25 - mmengine - INFO - Epoch(train) [173][10/32] lr: 1.0000e-06 eta: 0:06:39 time: 0.5298 data_time: 0.0781 memory: 17428 loss: 0.1174 loss_ce: 0.1174 2023/03/03 00:34:29 - mmengine - INFO - Epoch(train) [173][20/32] lr: 1.0000e-06 eta: 0:06:34 time: 0.4645 data_time: 0.0013 memory: 24875 loss: 0.1014 loss_ce: 0.1014 2023/03/03 00:34:34 - mmengine - INFO - Epoch(train) [173][30/32] lr: 1.0000e-06 eta: 0:06:30 time: 0.5098 data_time: 0.0011 memory: 20021 loss: 0.0907 loss_ce: 0.0907 2023/03/03 00:34:35 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:34:40 - mmengine - INFO - Epoch(train) [174][10/32] lr: 1.0000e-06 eta: 0:06:25 time: 0.5252 data_time: 0.0641 memory: 19958 loss: 0.1040 loss_ce: 0.1040 2023/03/03 00:34:45 - mmengine - INFO - Epoch(train) [174][20/32] lr: 1.0000e-06 eta: 0:06:20 time: 0.4527 data_time: 0.0015 memory: 24319 loss: 0.0878 loss_ce: 0.0878 2023/03/03 00:34:49 - mmengine - INFO - Epoch(train) [174][30/32] lr: 1.0000e-06 eta: 0:06:16 time: 0.4103 data_time: 0.0011 memory: 18246 loss: 0.1054 loss_ce: 0.1054 2023/03/03 00:34:50 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:34:55 - mmengine - INFO - Epoch(train) [175][10/32] lr: 1.0000e-06 eta: 0:06:10 time: 0.5736 data_time: 0.1156 memory: 17936 loss: 0.0955 loss_ce: 0.0955 2023/03/03 00:35:00 - mmengine - INFO - Epoch(train) [175][20/32] lr: 1.0000e-06 eta: 0:06:06 time: 0.4314 data_time: 0.0015 memory: 17733 loss: 0.0957 loss_ce: 0.0957 2023/03/03 00:35:04 - mmengine - INFO - Epoch(train) [175][30/32] lr: 1.0000e-06 eta: 0:06:01 time: 0.4568 data_time: 0.0012 memory: 21046 loss: 0.1137 loss_ce: 0.1137 2023/03/03 00:35:05 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:35:10 - mmengine - INFO - Epoch(train) [176][10/32] lr: 1.0000e-06 eta: 0:05:56 time: 0.5209 data_time: 0.0763 memory: 24319 loss: 0.1030 loss_ce: 0.1030 2023/03/03 00:35:15 - mmengine - INFO - Epoch(train) [176][20/32] lr: 1.0000e-06 eta: 0:05:52 time: 0.4973 data_time: 0.0015 memory: 17728 loss: 0.1116 loss_ce: 0.1116 2023/03/03 00:35:19 - mmengine - INFO - Epoch(train) [176][30/32] lr: 1.0000e-06 eta: 0:05:47 time: 0.4185 data_time: 0.0014 memory: 18597 loss: 0.1066 loss_ce: 0.1066 2023/03/03 00:35:20 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:35:26 - mmengine - INFO - Epoch(train) [177][10/32] lr: 1.0000e-06 eta: 0:05:42 time: 0.5487 data_time: 0.0566 memory: 22214 loss: 0.0898 loss_ce: 0.0898 2023/03/03 00:35:30 - mmengine - INFO - Epoch(train) [177][20/32] lr: 1.0000e-06 eta: 0:05:37 time: 0.4637 data_time: 0.0013 memory: 21255 loss: 0.0949 loss_ce: 0.0949 2023/03/03 00:35:35 - mmengine - INFO - Epoch(train) [177][30/32] lr: 1.0000e-06 eta: 0:05:33 time: 0.4446 data_time: 0.0011 memory: 17919 loss: 0.0939 loss_ce: 0.0939 2023/03/03 00:35:35 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:35:41 - mmengine - INFO - Epoch(train) [178][10/32] lr: 1.0000e-06 eta: 0:05:27 time: 0.6114 data_time: 0.1145 memory: 18325 loss: 0.1086 loss_ce: 0.1086 2023/03/03 00:35:46 - mmengine - INFO - Epoch(train) [178][20/32] lr: 1.0000e-06 eta: 0:05:23 time: 0.4391 data_time: 0.0013 memory: 24788 loss: 0.0991 loss_ce: 0.0991 2023/03/03 00:35:51 - mmengine - INFO - Epoch(train) [178][30/32] lr: 1.0000e-06 eta: 0:05:18 time: 0.4701 data_time: 0.0013 memory: 23640 loss: 0.1039 loss_ce: 0.1039 2023/03/03 00:35:51 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:35:56 - mmengine - INFO - Epoch(train) [179][10/32] lr: 1.0000e-06 eta: 0:05:13 time: 0.5238 data_time: 0.0877 memory: 19364 loss: 0.1047 loss_ce: 0.1047 2023/03/03 00:36:01 - mmengine - INFO - Epoch(train) [179][20/32] lr: 1.0000e-06 eta: 0:05:08 time: 0.4750 data_time: 0.0014 memory: 22486 loss: 0.0914 loss_ce: 0.0914 2023/03/03 00:36:06 - mmengine - INFO - Epoch(train) [179][30/32] lr: 1.0000e-06 eta: 0:05:04 time: 0.4456 data_time: 0.0011 memory: 18416 loss: 0.1224 loss_ce: 0.1224 2023/03/03 00:36:06 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:36:12 - mmengine - INFO - Epoch(train) [180][10/32] lr: 1.0000e-06 eta: 0:04:59 time: 0.5495 data_time: 0.0519 memory: 19463 loss: 0.0892 loss_ce: 0.0892 2023/03/03 00:36:16 - mmengine - INFO - Epoch(train) [180][20/32] lr: 1.0000e-06 eta: 0:04:54 time: 0.4352 data_time: 0.0014 memory: 19958 loss: 0.1121 loss_ce: 0.1121 2023/03/03 00:36:21 - mmengine - INFO - Epoch(train) [180][30/32] lr: 1.0000e-06 eta: 0:04:50 time: 0.4442 data_time: 0.0013 memory: 18597 loss: 0.1045 loss_ce: 0.1045 2023/03/03 00:36:21 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:36:35 - mmengine - INFO - Epoch(val) [180][10/63] eta: 0:01:12 time: 1.3770 data_time: 0.0042 memory: 17607 2023/03/03 00:36:59 - mmengine - INFO - Epoch(val) [180][20/63] eta: 0:01:20 time: 2.3481 data_time: 0.0004 memory: 1075 2023/03/03 00:37:21 - mmengine - INFO - Epoch(val) [180][30/63] eta: 0:01:05 time: 2.2400 data_time: 0.0004 memory: 1075 2023/03/03 00:37:32 - mmengine - INFO - Epoch(val) [180][40/63] eta: 0:00:40 time: 1.1112 data_time: 0.0004 memory: 1075 2023/03/03 00:37:54 - mmengine - INFO - Epoch(val) [180][50/63] eta: 0:00:23 time: 2.1526 data_time: 0.0003 memory: 1075 2023/03/03 00:38:06 - mmengine - INFO - Epoch(val) [180][60/63] eta: 0:00:05 time: 1.2383 data_time: 0.0004 memory: 1075 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.80, recall: 0.6172, precision: 0.7537, hmean: 0.6787 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.81, recall: 0.6172, precision: 0.7645, hmean: 0.6830 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.82, recall: 0.6172, precision: 0.7704, hmean: 0.6854 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.83, recall: 0.6168, precision: 0.7778, hmean: 0.6880 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.84, recall: 0.6143, precision: 0.7838, hmean: 0.6888 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.85, recall: 0.6115, precision: 0.7903, hmean: 0.6895 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.86, recall: 0.6076, precision: 0.7972, hmean: 0.6896 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.87, recall: 0.6042, precision: 0.8066, hmean: 0.6909 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.88, recall: 0.5999, precision: 0.8128, hmean: 0.6903 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.89, recall: 0.5932, precision: 0.8208, hmean: 0.6887 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.90, recall: 0.5888, precision: 0.8292, hmean: 0.6886 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.91, recall: 0.5821, precision: 0.8355, hmean: 0.6862 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.92, recall: 0.5758, precision: 0.8423, hmean: 0.6840 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.93, recall: 0.5619, precision: 0.8506, hmean: 0.6767 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.94, recall: 0.5445, precision: 0.8542, hmean: 0.6651 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.95, recall: 0.5364, precision: 0.8656, hmean: 0.6623 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.96, recall: 0.5286, precision: 0.8812, hmean: 0.6608 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.97, recall: 0.5104, precision: 0.8848, hmean: 0.6473 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.98, recall: 0.4858, precision: 0.8961, hmean: 0.6300 2023/03/03 00:39:08 - mmengine - INFO - text score threshold: 0.99, recall: 0.4497, precision: 0.9130, hmean: 0.6026 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.80, recall: 0.6596, precision: 0.8054, hmean: 0.7253 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.81, recall: 0.6586, precision: 0.8157, hmean: 0.7288 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.82, recall: 0.6582, precision: 0.8215, hmean: 0.7308 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.83, recall: 0.6572, precision: 0.8288, hmean: 0.7331 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.84, recall: 0.6543, precision: 0.8348, hmean: 0.7336 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.85, recall: 0.6514, precision: 0.8419, hmean: 0.7345 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.86, recall: 0.6466, precision: 0.8484, hmean: 0.7339 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.87, recall: 0.6413, precision: 0.8560, hmean: 0.7333 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.88, recall: 0.6360, precision: 0.8617, hmean: 0.7319 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.89, recall: 0.6264, precision: 0.8668, hmean: 0.7272 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.90, recall: 0.6182, precision: 0.8705, hmean: 0.7230 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.91, recall: 0.6105, precision: 0.8763, hmean: 0.7196 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.92, recall: 0.6038, precision: 0.8831, hmean: 0.7172 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.93, recall: 0.5874, precision: 0.8892, hmean: 0.7075 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.94, recall: 0.5676, precision: 0.8905, hmean: 0.6933 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.95, recall: 0.5561, precision: 0.8974, hmean: 0.6867 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.96, recall: 0.5416, precision: 0.9029, hmean: 0.6771 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.97, recall: 0.5243, precision: 0.9090, hmean: 0.6650 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.98, recall: 0.4945, precision: 0.9121, hmean: 0.6413 2023/03/03 00:39:17 - mmengine - INFO - text score threshold: 0.99, recall: 0.4559, precision: 0.9257, hmean: 0.6110 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.80, recall: 0.7169, precision: 0.8754, hmean: 0.7882 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.81, recall: 0.7135, precision: 0.8837, hmean: 0.7896 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.82, recall: 0.7121, precision: 0.8888, hmean: 0.7907 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.83, recall: 0.7097, precision: 0.8950, hmean: 0.7916 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.84, recall: 0.7063, precision: 0.9011, hmean: 0.7919 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.85, recall: 0.7010, precision: 0.9060, hmean: 0.7904 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.86, recall: 0.6943, precision: 0.9109, hmean: 0.7880 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.87, recall: 0.6866, precision: 0.9165, hmean: 0.7850 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.88, recall: 0.6793, precision: 0.9204, hmean: 0.7817 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.89, recall: 0.6683, precision: 0.9247, hmean: 0.7759 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.90, recall: 0.6586, precision: 0.9275, hmean: 0.7703 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.91, recall: 0.6490, precision: 0.9316, hmean: 0.7650 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.92, recall: 0.6389, precision: 0.9345, hmean: 0.7589 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.93, recall: 0.6196, precision: 0.9380, hmean: 0.7463 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.94, recall: 0.5989, precision: 0.9396, hmean: 0.7315 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.95, recall: 0.5840, precision: 0.9425, hmean: 0.7212 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.96, recall: 0.5681, precision: 0.9470, hmean: 0.7102 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.97, recall: 0.5474, precision: 0.9491, hmean: 0.6944 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.98, recall: 0.5161, precision: 0.9520, hmean: 0.6694 2023/03/03 00:39:27 - mmengine - INFO - text score threshold: 0.99, recall: 0.4733, precision: 0.9609, hmean: 0.6342 2023/03/03 00:39:27 - mmengine - INFO - Epoch(val) [180][63/63] generic/precision: 0.8066 generic/recall: 0.6042 generic/hmean: 0.6909 weak/precision: 0.8419 weak/recall: 0.6514 weak/hmean: 0.7345 strong/precision: 0.9011 strong/recall: 0.7063 strong/hmean: 0.7919 2023/03/03 00:39:28 - mmengine - INFO - The previous best checkpoint mmocr/projects/SPTS/work_dirs/spts_resnet50_350e_icdar2015/best_generic/hmean_epoch_170.pth is removed 2023/03/03 00:39:30 - mmengine - INFO - The best checkpoint with 0.6909 generic/hmean at 180 epoch is saved to best_generic/hmean_epoch_180.pth. 2023/03/03 00:39:35 - mmengine - INFO - Epoch(train) [181][10/32] lr: 1.0000e-06 eta: 0:04:44 time: 0.5101 data_time: 0.0302 memory: 20651 loss: 0.0914 loss_ce: 0.0914 2023/03/03 00:39:40 - mmengine - INFO - Epoch(train) [181][20/32] lr: 1.0000e-06 eta: 0:04:40 time: 0.4517 data_time: 0.0013 memory: 23640 loss: 0.1087 loss_ce: 0.1087 2023/03/03 00:39:45 - mmengine - INFO - Epoch(train) [181][30/32] lr: 1.0000e-06 eta: 0:04:35 time: 0.5103 data_time: 0.0011 memory: 23923 loss: 0.1001 loss_ce: 0.1001 2023/03/03 00:39:45 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:39:51 - mmengine - INFO - Epoch(train) [182][10/32] lr: 1.0000e-06 eta: 0:04:30 time: 0.5649 data_time: 0.0790 memory: 21651 loss: 0.0992 loss_ce: 0.0992 2023/03/03 00:39:56 - mmengine - INFO - Epoch(train) [182][20/32] lr: 1.0000e-06 eta: 0:04:25 time: 0.4550 data_time: 0.0018 memory: 19751 loss: 0.0959 loss_ce: 0.0959 2023/03/03 00:40:00 - mmengine - INFO - Epoch(train) [182][30/32] lr: 1.0000e-06 eta: 0:04:21 time: 0.4277 data_time: 0.0014 memory: 17644 loss: 0.1020 loss_ce: 0.1020 2023/03/03 00:40:00 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:40:06 - mmengine - INFO - Epoch(train) [183][10/32] lr: 1.0000e-06 eta: 0:04:15 time: 0.5377 data_time: 0.0640 memory: 24875 loss: 0.1000 loss_ce: 0.1000 2023/03/03 00:40:10 - mmengine - INFO - Epoch(train) [183][20/32] lr: 1.0000e-06 eta: 0:04:11 time: 0.4507 data_time: 0.0016 memory: 17733 loss: 0.0940 loss_ce: 0.0940 2023/03/03 00:40:15 - mmengine - INFO - Epoch(train) [183][30/32] lr: 1.0000e-06 eta: 0:04:06 time: 0.4265 data_time: 0.0014 memory: 20382 loss: 0.0980 loss_ce: 0.0980 2023/03/03 00:40:15 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:40:21 - mmengine - INFO - Epoch(train) [184][10/32] lr: 1.0000e-06 eta: 0:04:01 time: 0.5650 data_time: 0.1037 memory: 19119 loss: 0.0944 loss_ce: 0.0944 2023/03/03 00:40:26 - mmengine - INFO - Epoch(train) [184][20/32] lr: 1.0000e-06 eta: 0:03:57 time: 0.4766 data_time: 0.0015 memory: 20790 loss: 0.0871 loss_ce: 0.0871 2023/03/03 00:40:30 - mmengine - INFO - Epoch(train) [184][30/32] lr: 1.0000e-06 eta: 0:03:52 time: 0.4403 data_time: 0.0013 memory: 22870 loss: 0.1052 loss_ce: 0.1052 2023/03/03 00:40:31 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:40:37 - mmengine - INFO - Epoch(train) [185][10/32] lr: 1.0000e-06 eta: 0:03:47 time: 0.5407 data_time: 0.0295 memory: 18966 loss: 0.1082 loss_ce: 0.1082 2023/03/03 00:40:41 - mmengine - INFO - Epoch(train) [185][20/32] lr: 1.0000e-06 eta: 0:03:42 time: 0.4818 data_time: 0.0015 memory: 19751 loss: 0.0956 loss_ce: 0.0956 2023/03/03 00:40:46 - mmengine - INFO - Epoch(train) [185][30/32] lr: 1.0000e-06 eta: 0:03:38 time: 0.4588 data_time: 0.0012 memory: 18083 loss: 0.1070 loss_ce: 0.1070 2023/03/03 00:40:47 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:40:52 - mmengine - INFO - Epoch(train) [186][10/32] lr: 1.0000e-06 eta: 0:03:32 time: 0.5635 data_time: 0.0739 memory: 18597 loss: 0.1189 loss_ce: 0.1189 2023/03/03 00:40:57 - mmengine - INFO - Epoch(train) [186][20/32] lr: 1.0000e-06 eta: 0:03:28 time: 0.4924 data_time: 0.0015 memory: 19789 loss: 0.0918 loss_ce: 0.0918 2023/03/03 00:41:02 - mmengine - INFO - Epoch(train) [186][30/32] lr: 1.0000e-06 eta: 0:03:23 time: 0.4404 data_time: 0.0014 memory: 17919 loss: 0.1020 loss_ce: 0.1020 2023/03/03 00:41:02 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:41:08 - mmengine - INFO - Epoch(train) [187][10/32] lr: 1.0000e-06 eta: 0:03:18 time: 0.5971 data_time: 0.1189 memory: 19958 loss: 0.1156 loss_ce: 0.1156 2023/03/03 00:41:13 - mmengine - INFO - Epoch(train) [187][20/32] lr: 1.0000e-06 eta: 0:03:13 time: 0.4537 data_time: 0.0014 memory: 17583 loss: 0.1040 loss_ce: 0.1040 2023/03/03 00:41:17 - mmengine - INFO - Epoch(train) [187][30/32] lr: 1.0000e-06 eta: 0:03:09 time: 0.4711 data_time: 0.0014 memory: 19751 loss: 0.0955 loss_ce: 0.0955 2023/03/03 00:41:18 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:41:24 - mmengine - INFO - Epoch(train) [188][10/32] lr: 1.0000e-06 eta: 0:03:03 time: 0.5493 data_time: 0.0248 memory: 18966 loss: 0.0999 loss_ce: 0.0999 2023/03/03 00:41:27 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:41:28 - mmengine - INFO - Epoch(train) [188][20/32] lr: 1.0000e-06 eta: 0:02:59 time: 0.4621 data_time: 0.0017 memory: 20382 loss: 0.1050 loss_ce: 0.1050 2023/03/03 00:41:33 - mmengine - INFO - Epoch(train) [188][30/32] lr: 1.0000e-06 eta: 0:02:54 time: 0.4249 data_time: 0.0011 memory: 18083 loss: 0.1056 loss_ce: 0.1056 2023/03/03 00:41:33 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:41:39 - mmengine - INFO - Epoch(train) [189][10/32] lr: 1.0000e-06 eta: 0:02:49 time: 0.5783 data_time: 0.0496 memory: 24319 loss: 0.1015 loss_ce: 0.1015 2023/03/03 00:41:44 - mmengine - INFO - Epoch(train) [189][20/32] lr: 1.0000e-06 eta: 0:02:44 time: 0.4667 data_time: 0.0016 memory: 19751 loss: 0.0793 loss_ce: 0.0793 2023/03/03 00:41:48 - mmengine - INFO - Epoch(train) [189][30/32] lr: 1.0000e-06 eta: 0:02:40 time: 0.4249 data_time: 0.0013 memory: 17919 loss: 0.0992 loss_ce: 0.0992 2023/03/03 00:41:49 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:41:55 - mmengine - INFO - Epoch(train) [190][10/32] lr: 1.0000e-06 eta: 0:02:35 time: 0.6323 data_time: 0.1224 memory: 24875 loss: 0.0702 loss_ce: 0.0702 2023/03/03 00:42:00 - mmengine - INFO - Epoch(train) [190][20/32] lr: 1.0000e-06 eta: 0:02:30 time: 0.4717 data_time: 0.0013 memory: 22984 loss: 0.0997 loss_ce: 0.0997 2023/03/03 00:42:04 - mmengine - INFO - Epoch(train) [190][30/32] lr: 1.0000e-06 eta: 0:02:26 time: 0.4343 data_time: 0.0013 memory: 17877 loss: 0.1125 loss_ce: 0.1125 2023/03/03 00:42:05 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:42:18 - mmengine - INFO - Epoch(val) [190][10/63] eta: 0:01:10 time: 1.3236 data_time: 0.0032 memory: 19463 2023/03/03 00:42:41 - mmengine - INFO - Epoch(val) [190][20/63] eta: 0:01:18 time: 2.3136 data_time: 0.0007 memory: 1075 2023/03/03 00:43:03 - mmengine - INFO - Epoch(val) [190][30/63] eta: 0:01:04 time: 2.2189 data_time: 0.0005 memory: 1075 2023/03/03 00:43:14 - mmengine - INFO - Epoch(val) [190][40/63] eta: 0:00:39 time: 1.0794 data_time: 0.0005 memory: 1075 2023/03/03 00:43:37 - mmengine - INFO - Epoch(val) [190][50/63] eta: 0:00:23 time: 2.2466 data_time: 0.0005 memory: 1075 2023/03/03 00:43:49 - mmengine - INFO - Epoch(val) [190][60/63] eta: 0:00:05 time: 1.2328 data_time: 0.0003 memory: 1075 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.80, recall: 0.6139, precision: 0.7469, hmean: 0.6739 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.81, recall: 0.6129, precision: 0.7524, hmean: 0.6755 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.82, recall: 0.6124, precision: 0.7594, hmean: 0.6780 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.83, recall: 0.6095, precision: 0.7668, hmean: 0.6792 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.84, recall: 0.6076, precision: 0.7733, hmean: 0.6805 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.85, recall: 0.6057, precision: 0.7814, hmean: 0.6824 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.86, recall: 0.6028, precision: 0.7869, hmean: 0.6827 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.87, recall: 0.6013, precision: 0.7966, hmean: 0.6853 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.88, recall: 0.5965, precision: 0.8051, hmean: 0.6853 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.89, recall: 0.5898, precision: 0.8145, hmean: 0.6842 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.90, recall: 0.5874, precision: 0.8215, hmean: 0.6850 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.91, recall: 0.5792, precision: 0.8297, hmean: 0.6822 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.92, recall: 0.5662, precision: 0.8346, hmean: 0.6747 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.93, recall: 0.5599, precision: 0.8421, hmean: 0.6726 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.94, recall: 0.5484, precision: 0.8525, hmean: 0.6674 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.95, recall: 0.5368, precision: 0.8643, hmean: 0.6623 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.96, recall: 0.5238, precision: 0.8746, hmean: 0.6552 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.97, recall: 0.5079, precision: 0.8828, hmean: 0.6449 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.98, recall: 0.4829, precision: 0.8923, hmean: 0.6267 2023/03/03 00:44:58 - mmengine - INFO - text score threshold: 0.99, recall: 0.4511, precision: 0.9088, hmean: 0.6030 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.80, recall: 0.6591, precision: 0.8020, hmean: 0.7236 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.81, recall: 0.6577, precision: 0.8073, hmean: 0.7249 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.82, recall: 0.6567, precision: 0.8143, hmean: 0.7271 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.83, recall: 0.6533, precision: 0.8219, hmean: 0.7280 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.84, recall: 0.6509, precision: 0.8284, hmean: 0.7290 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.85, recall: 0.6466, precision: 0.8342, hmean: 0.7285 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.86, recall: 0.6428, precision: 0.8391, hmean: 0.7279 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.87, recall: 0.6399, precision: 0.8476, hmean: 0.7292 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.88, recall: 0.6326, precision: 0.8538, hmean: 0.7268 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.89, recall: 0.6235, precision: 0.8610, hmean: 0.7233 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.90, recall: 0.6196, precision: 0.8667, hmean: 0.7226 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.91, recall: 0.6086, precision: 0.8717, hmean: 0.7168 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.92, recall: 0.5941, precision: 0.8758, hmean: 0.7080 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.93, recall: 0.5869, precision: 0.8827, hmean: 0.7050 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.94, recall: 0.5720, precision: 0.8892, hmean: 0.6962 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.95, recall: 0.5571, precision: 0.8969, hmean: 0.6873 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.96, recall: 0.5407, precision: 0.9027, hmean: 0.6763 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.97, recall: 0.5234, precision: 0.9096, hmean: 0.6644 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.98, recall: 0.4940, precision: 0.9128, hmean: 0.6410 2023/03/03 00:45:09 - mmengine - INFO - text score threshold: 0.99, recall: 0.4588, precision: 0.9243, hmean: 0.6133 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.80, recall: 0.7198, precision: 0.8758, hmean: 0.7902 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.81, recall: 0.7174, precision: 0.8806, hmean: 0.7907 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.82, recall: 0.7140, precision: 0.8854, hmean: 0.7905 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.83, recall: 0.7097, precision: 0.8928, hmean: 0.7908 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.84, recall: 0.7053, precision: 0.8977, hmean: 0.7900 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.85, recall: 0.6996, precision: 0.9025, hmean: 0.7882 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.86, recall: 0.6928, precision: 0.9045, hmean: 0.7846 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.87, recall: 0.6885, precision: 0.9120, hmean: 0.7846 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.88, recall: 0.6793, precision: 0.9168, hmean: 0.7804 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.89, recall: 0.6668, precision: 0.9209, hmean: 0.7735 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.90, recall: 0.6625, precision: 0.9266, hmean: 0.7726 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.91, recall: 0.6490, precision: 0.9297, hmean: 0.7644 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.92, recall: 0.6331, precision: 0.9333, hmean: 0.7544 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.93, recall: 0.6225, precision: 0.9363, hmean: 0.7478 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.94, recall: 0.6038, precision: 0.9386, hmean: 0.7348 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.95, recall: 0.5859, precision: 0.9434, hmean: 0.7229 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.96, recall: 0.5676, precision: 0.9477, hmean: 0.7100 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.97, recall: 0.5460, precision: 0.9490, hmean: 0.6932 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.98, recall: 0.5161, precision: 0.9537, hmean: 0.6698 2023/03/03 00:45:18 - mmengine - INFO - text score threshold: 0.99, recall: 0.4781, precision: 0.9631, hmean: 0.6390 2023/03/03 00:45:18 - mmengine - INFO - Epoch(val) [190][63/63] generic/precision: 0.7966 generic/recall: 0.6013 generic/hmean: 0.6853 weak/precision: 0.8476 weak/recall: 0.6399 weak/hmean: 0.7292 strong/precision: 0.8928 strong/recall: 0.7097 strong/hmean: 0.7908 2023/03/03 00:45:23 - mmengine - INFO - Epoch(train) [191][10/32] lr: 1.0000e-06 eta: 0:02:20 time: 0.5302 data_time: 0.0666 memory: 19958 loss: 0.0965 loss_ce: 0.0965 2023/03/03 00:45:28 - mmengine - INFO - Epoch(train) [191][20/32] lr: 1.0000e-06 eta: 0:02:16 time: 0.4835 data_time: 0.0019 memory: 19958 loss: 0.0887 loss_ce: 0.0887 2023/03/03 00:45:32 - mmengine - INFO - Epoch(train) [191][30/32] lr: 1.0000e-06 eta: 0:02:11 time: 0.4481 data_time: 0.0013 memory: 17733 loss: 0.0975 loss_ce: 0.0975 2023/03/03 00:45:33 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:45:39 - mmengine - INFO - Epoch(train) [192][10/32] lr: 1.0000e-06 eta: 0:02:06 time: 0.5526 data_time: 0.0648 memory: 21046 loss: 0.1126 loss_ce: 0.1126 2023/03/03 00:45:43 - mmengine - INFO - Epoch(train) [192][20/32] lr: 1.0000e-06 eta: 0:02:01 time: 0.4503 data_time: 0.0015 memory: 21876 loss: 0.0934 loss_ce: 0.0934 2023/03/03 00:45:47 - mmengine - INFO - Epoch(train) [192][30/32] lr: 1.0000e-06 eta: 0:01:57 time: 0.4323 data_time: 0.0016 memory: 17940 loss: 0.0833 loss_ce: 0.0833 2023/03/03 00:45:48 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:45:53 - mmengine - INFO - Epoch(train) [193][10/32] lr: 1.0000e-06 eta: 0:01:51 time: 0.5324 data_time: 0.0444 memory: 18966 loss: 0.1048 loss_ce: 0.1048 2023/03/03 00:45:58 - mmengine - INFO - Epoch(train) [193][20/32] lr: 1.0000e-06 eta: 0:01:47 time: 0.4601 data_time: 0.0016 memory: 25455 loss: 0.0858 loss_ce: 0.0858 2023/03/03 00:46:03 - mmengine - INFO - Epoch(train) [193][30/32] lr: 1.0000e-06 eta: 0:01:42 time: 0.4949 data_time: 0.0012 memory: 23742 loss: 0.0934 loss_ce: 0.0934 2023/03/03 00:46:03 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:46:09 - mmengine - INFO - Epoch(train) [194][10/32] lr: 1.0000e-06 eta: 0:01:37 time: 0.5602 data_time: 0.1043 memory: 19958 loss: 0.1011 loss_ce: 0.1011 2023/03/03 00:46:14 - mmengine - INFO - Epoch(train) [194][20/32] lr: 1.0000e-06 eta: 0:01:32 time: 0.4500 data_time: 0.0015 memory: 19142 loss: 0.0957 loss_ce: 0.0957 2023/03/03 00:46:18 - mmengine - INFO - Epoch(train) [194][30/32] lr: 1.0000e-06 eta: 0:01:28 time: 0.4541 data_time: 0.0014 memory: 19587 loss: 0.0935 loss_ce: 0.0935 2023/03/03 00:46:19 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:46:24 - mmengine - INFO - Epoch(train) [195][10/32] lr: 1.0000e-06 eta: 0:01:22 time: 0.5619 data_time: 0.0678 memory: 18779 loss: 0.0949 loss_ce: 0.0949 2023/03/03 00:46:29 - mmengine - INFO - Epoch(train) [195][20/32] lr: 1.0000e-06 eta: 0:01:18 time: 0.4981 data_time: 0.0015 memory: 18325 loss: 0.1053 loss_ce: 0.1053 2023/03/03 00:46:34 - mmengine - INFO - Epoch(train) [195][30/32] lr: 1.0000e-06 eta: 0:01:13 time: 0.4532 data_time: 0.0017 memory: 24875 loss: 0.1003 loss_ce: 0.1003 2023/03/03 00:46:35 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:46:40 - mmengine - INFO - Epoch(train) [196][10/32] lr: 1.0000e-06 eta: 0:01:08 time: 0.5475 data_time: 0.0875 memory: 18246 loss: 0.1290 loss_ce: 0.1290 2023/03/03 00:46:44 - mmengine - INFO - Epoch(train) [196][20/32] lr: 1.0000e-06 eta: 0:01:03 time: 0.4324 data_time: 0.0015 memory: 18353 loss: 0.1085 loss_ce: 0.1085 2023/03/03 00:46:49 - mmengine - INFO - Epoch(train) [196][30/32] lr: 1.0000e-06 eta: 0:00:59 time: 0.4994 data_time: 0.0014 memory: 20820 loss: 0.0941 loss_ce: 0.0941 2023/03/03 00:46:50 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:46:55 - mmengine - INFO - Epoch(train) [197][10/32] lr: 1.0000e-06 eta: 0:00:53 time: 0.5503 data_time: 0.1033 memory: 20172 loss: 0.1009 loss_ce: 0.1009 2023/03/03 00:47:00 - mmengine - INFO - Epoch(train) [197][20/32] lr: 1.0000e-06 eta: 0:00:49 time: 0.4532 data_time: 0.0015 memory: 19142 loss: 0.1072 loss_ce: 0.1072 2023/03/03 00:47:04 - mmengine - INFO - Epoch(train) [197][30/32] lr: 1.0000e-06 eta: 0:00:44 time: 0.4538 data_time: 0.0014 memory: 18940 loss: 0.1080 loss_ce: 0.1080 2023/03/03 00:47:05 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:47:10 - mmengine - INFO - Epoch(train) [198][10/32] lr: 1.0000e-06 eta: 0:00:39 time: 0.5130 data_time: 0.0713 memory: 18779 loss: 0.0895 loss_ce: 0.0895 2023/03/03 00:47:15 - mmengine - INFO - Epoch(train) [198][20/32] lr: 1.0000e-06 eta: 0:00:34 time: 0.4400 data_time: 0.0015 memory: 17850 loss: 0.0985 loss_ce: 0.0985 2023/03/03 00:47:19 - mmengine - INFO - Epoch(train) [198][30/32] lr: 1.0000e-06 eta: 0:00:29 time: 0.4543 data_time: 0.0013 memory: 22545 loss: 0.0917 loss_ce: 0.0917 2023/03/03 00:47:20 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:47:25 - mmengine - INFO - Epoch(train) [199][10/32] lr: 1.0000e-06 eta: 0:00:24 time: 0.5212 data_time: 0.0839 memory: 17892 loss: 0.1133 loss_ce: 0.1133 2023/03/03 00:47:29 - mmengine - INFO - Epoch(train) [199][20/32] lr: 1.0000e-06 eta: 0:00:19 time: 0.4373 data_time: 0.0016 memory: 18246 loss: 0.0909 loss_ce: 0.0909 2023/03/03 00:47:34 - mmengine - INFO - Epoch(train) [199][30/32] lr: 1.0000e-06 eta: 0:00:15 time: 0.4406 data_time: 0.0014 memory: 20600 loss: 0.0939 loss_ce: 0.0939 2023/03/03 00:47:34 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:47:39 - mmengine - INFO - Epoch(train) [200][10/32] lr: 1.0000e-06 eta: 0:00:09 time: 0.5116 data_time: 0.0436 memory: 19846 loss: 0.1096 loss_ce: 0.1096 2023/03/03 00:47:44 - mmengine - INFO - Epoch(train) [200][20/32] lr: 1.0000e-06 eta: 0:00:05 time: 0.4272 data_time: 0.0019 memory: 19958 loss: 0.1134 loss_ce: 0.1134 2023/03/03 00:47:48 - mmengine - INFO - Epoch(train) [200][30/32] lr: 1.0000e-06 eta: 0:00:00 time: 0.4554 data_time: 0.0015 memory: 20407 loss: 0.0875 loss_ce: 0.0875 2023/03/03 00:47:49 - mmengine - INFO - Exp name: spts_resnet50_350e_icdar2015_20230302_230026 2023/03/03 00:47:49 - mmengine - INFO - Saving checkpoint at 200 epochs 2023/03/03 00:47:51 - mmengine - WARNING - `save_param_scheduler` is True but `self.param_schedulers` is None, so skip saving parameter schedulers 2023/03/03 00:48:04 - mmengine - INFO - Epoch(val) [200][10/63] eta: 0:01:06 time: 1.2557 data_time: 0.0034 memory: 17862 2023/03/03 00:48:27 - mmengine - INFO - Epoch(val) [200][20/63] eta: 0:01:16 time: 2.3096 data_time: 0.0005 memory: 1075 2023/03/03 00:48:49 - mmengine - INFO - Epoch(val) [200][30/63] eta: 0:01:03 time: 2.1684 data_time: 0.0005 memory: 1075 2023/03/03 00:49:00 - mmengine - INFO - Epoch(val) [200][40/63] eta: 0:00:39 time: 1.1120 data_time: 0.0003 memory: 1075 2023/03/03 00:49:22 - mmengine - INFO - Epoch(val) [200][50/63] eta: 0:00:23 time: 2.2072 data_time: 0.0004 memory: 1075 2023/03/03 00:49:35 - mmengine - INFO - Epoch(val) [200][60/63] eta: 0:00:05 time: 1.2918 data_time: 0.0004 memory: 1075 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.80, recall: 0.6163, precision: 0.7472, hmean: 0.6755 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.81, recall: 0.6158, precision: 0.7532, hmean: 0.6776 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.82, recall: 0.6139, precision: 0.7603, hmean: 0.6793 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.83, recall: 0.6119, precision: 0.7666, hmean: 0.6806 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.84, recall: 0.6110, precision: 0.7738, hmean: 0.6828 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.85, recall: 0.6091, precision: 0.7833, hmean: 0.6853 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.86, recall: 0.6071, precision: 0.7896, hmean: 0.6864 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.87, recall: 0.6042, precision: 0.7994, hmean: 0.6882 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.88, recall: 0.5970, precision: 0.8057, hmean: 0.6858 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.89, recall: 0.5946, precision: 0.8146, hmean: 0.6874 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.90, recall: 0.5855, precision: 0.8189, hmean: 0.6828 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.91, recall: 0.5802, precision: 0.8282, hmean: 0.6823 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.92, recall: 0.5691, precision: 0.8347, hmean: 0.6768 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.93, recall: 0.5609, precision: 0.8442, hmean: 0.6740 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.94, recall: 0.5508, precision: 0.8512, hmean: 0.6688 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.95, recall: 0.5383, precision: 0.8580, hmean: 0.6615 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.96, recall: 0.5267, precision: 0.8676, hmean: 0.6555 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.97, recall: 0.5084, precision: 0.8763, hmean: 0.6435 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.98, recall: 0.4819, precision: 0.8890, hmean: 0.6250 2023/03/03 00:50:37 - mmengine - INFO - text score threshold: 0.99, recall: 0.4516, precision: 0.9080, hmean: 0.6032 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.80, recall: 0.6620, precision: 0.8027, hmean: 0.7256 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.81, recall: 0.6601, precision: 0.8074, hmean: 0.7264 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.82, recall: 0.6582, precision: 0.8151, hmean: 0.7283 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.83, recall: 0.6558, precision: 0.8215, hmean: 0.7293 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.84, recall: 0.6533, precision: 0.8274, hmean: 0.7302 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.85, recall: 0.6500, precision: 0.8359, hmean: 0.7313 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.86, recall: 0.6471, precision: 0.8416, hmean: 0.7316 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.87, recall: 0.6413, precision: 0.8484, hmean: 0.7305 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.88, recall: 0.6331, precision: 0.8545, hmean: 0.7273 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.89, recall: 0.6283, precision: 0.8608, hmean: 0.7264 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.90, recall: 0.6187, precision: 0.8653, hmean: 0.7215 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.91, recall: 0.6115, precision: 0.8729, hmean: 0.7191 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.92, recall: 0.5989, precision: 0.8785, hmean: 0.7123 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.93, recall: 0.5883, precision: 0.8855, hmean: 0.7070 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.94, recall: 0.5753, precision: 0.8891, hmean: 0.6986 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.95, recall: 0.5609, precision: 0.8941, hmean: 0.6893 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.96, recall: 0.5460, precision: 0.8993, hmean: 0.6794 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.97, recall: 0.5258, precision: 0.9062, hmean: 0.6654 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.98, recall: 0.4940, precision: 0.9112, hmean: 0.6406 2023/03/03 00:50:46 - mmengine - INFO - text score threshold: 0.99, recall: 0.4588, precision: 0.9226, hmean: 0.6129 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.80, recall: 0.7203, precision: 0.8733, hmean: 0.7894 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.81, recall: 0.7179, precision: 0.8781, hmean: 0.7899 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.82, recall: 0.7135, precision: 0.8837, hmean: 0.7896 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.83, recall: 0.7087, precision: 0.8878, hmean: 0.7882 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.84, recall: 0.7049, precision: 0.8927, hmean: 0.7877 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.85, recall: 0.6991, precision: 0.8991, hmean: 0.7866 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.86, recall: 0.6952, precision: 0.9042, hmean: 0.7861 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.87, recall: 0.6880, precision: 0.9102, hmean: 0.7837 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.88, recall: 0.6774, precision: 0.9142, hmean: 0.7782 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.89, recall: 0.6716, precision: 0.9202, hmean: 0.7765 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.90, recall: 0.6596, precision: 0.9226, hmean: 0.7692 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.91, recall: 0.6495, precision: 0.9271, hmean: 0.7639 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.92, recall: 0.6351, precision: 0.9315, hmean: 0.7552 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.93, recall: 0.6221, precision: 0.9362, hmean: 0.7475 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.94, recall: 0.6071, precision: 0.9382, hmean: 0.7372 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.95, recall: 0.5903, precision: 0.9409, hmean: 0.7254 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.96, recall: 0.5734, precision: 0.9445, hmean: 0.7136 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.97, recall: 0.5503, precision: 0.9485, hmean: 0.6965 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.98, recall: 0.5152, precision: 0.9503, hmean: 0.6681 2023/03/03 00:50:55 - mmengine - INFO - text score threshold: 0.99, recall: 0.4771, precision: 0.9593, hmean: 0.6373 2023/03/03 00:50:55 - mmengine - INFO - Epoch(val) [200][63/63] generic/precision: 0.7994 generic/recall: 0.6042 generic/hmean: 0.6882 weak/precision: 0.8416 weak/recall: 0.6471 weak/hmean: 0.7316 strong/precision: 0.8781 strong/recall: 0.7179 strong/hmean: 0.7899