2022/08/25 15:16:49 - mmengine - INFO - ------------------------------------------------------------ System environment: sys.platform: linux Python: 3.7.13 (default, Mar 29 2022, 02:18:16) [GCC 7.5.0] CUDA available: True numpy_random_seed: 1927889406 GPU 0: NVIDIA A100-SXM4-80GB CUDA_HOME: /mnt/cache/share/cuda-11.1 NVCC: Cuda compilation tools, release 11.1, V11.1.74 GCC: gcc (GCC) 5.4.0 PyTorch: 1.9.0+cu111 PyTorch compiling details: PyTorch built with: - GCC 7.3 - C++ Version: 201402 - Intel(R) Math Kernel Library Version 2020.0.0 Product Build 20191122 for Intel(R) 64 architecture applications - Intel(R) MKL-DNN v2.1.2 (Git Hash 98be7e8afa711dc9b66c8ff3504129cb82013cdb) - OpenMP 201511 (a.k.a. OpenMP 4.5) - NNPACK is enabled - CPU capability usage: AVX2 - CUDA Runtime 11.1 - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86 - CuDNN 8.0.5 - Magma 2.5.2 - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.1, CUDNN_VERSION=8.0.5, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-variable -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.9.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, TorchVision: 0.10.0+cu111 OpenCV: 4.5.4 MMEngine: 0.0.1 Runtime environment: cudnn_benchmark: True mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} dist_cfg: {'backend': 'nccl'} seed: None Distributed launcher: none Distributed training: False GPU number: 1 ------------------------------------------------------------ 2022/08/25 15:16:50 - mmengine - INFO - Config: default_scope = 'mmocr' default_hooks = dict( timer=dict(type='IterTimerHook'), logger=dict(type='LoggerHook', interval=100), param_scheduler=dict(type='ParamSchedulerHook'), checkpoint=dict(type='CheckpointHook', interval=1), 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)) env_cfg = dict( cudnn_benchmark=True, mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), dist_cfg=dict(backend='nccl')) log_level = 'INFO' load_from = None resume = False wildreceipt_data_root = 'data/kie/wildreceipt/' wildreceipt_train = dict( type='WildReceiptDataset', data_root='data/kie/wildreceipt/', metainfo='data/kie/wildreceipt/class_list.txt', ann_file='train.txt', pipeline=[ dict(type='LoadImageFromFile'), dict(type='LoadKIEAnnotations'), dict(type='Resize', scale=(1024, 512), keep_ratio=True), dict(type='PackKIEInputs') ]) wildreceipt_test = dict( type='WildReceiptDataset', data_root='data/kie/wildreceipt/', metainfo='data/kie/wildreceipt/class_list.txt', ann_file='test.txt', test_mode=True, pipeline=[ dict(type='LoadImageFromFile'), dict(type='LoadKIEAnnotations'), dict(type='Resize', scale=(1024, 512), keep_ratio=True), dict(type='PackKIEInputs') ]) optim_wrapper = dict( type='OptimWrapper', optimizer=dict(type='Adam', weight_decay=0.0001)) train_cfg = dict(type='EpochBasedTrainLoop', max_epochs=60, val_interval=1) val_cfg = dict(type='ValLoop') test_cfg = dict(type='TestLoop') param_scheduler = [dict(type='MultiStepLR', milestones=[40, 50], end=60)] num_classes = 26 model = dict( type='SDMGR', kie_head=dict( type='SDMGRHead', visual_dim=16, num_classes=26, module_loss=dict(type='SDMGRModuleLoss'), postprocessor=dict(type='SDMGRPostProcessor')), dictionary=dict( type='Dictionary', dict_file='data/kie/wildreceipt/dict.txt', with_padding=True, with_unknown=True, unknown_token=None), backbone=dict(type='UNet', base_channels=16), roi_extractor=dict( type='mmdet.SingleRoIExtractor', roi_layer=dict(type='RoIAlign', output_size=7), featmap_strides=[1]), data_preprocessor=dict( type='ImgDataPreprocessor', mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], bgr_to_rgb=True, pad_size_divisor=32)) train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadKIEAnnotations'), dict(type='Resize', scale=(1024, 512), keep_ratio=True), dict(type='PackKIEInputs') ] test_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadKIEAnnotations'), dict(type='Resize', scale=(1024, 512), keep_ratio=True), dict(type='PackKIEInputs') ] val_evaluator = dict( type='F1Metric', mode='macro', num_classes=26, ignored_classes=[0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 25]) test_evaluator = dict( type='F1Metric', mode='macro', num_classes=26, ignored_classes=[0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 25]) train_dataloader = dict( batch_size=4, num_workers=4, persistent_workers=True, sampler=dict(type='DefaultSampler', shuffle=True), dataset=dict( type='WildReceiptDataset', data_root='data/kie/wildreceipt/', metainfo='data/kie/wildreceipt/class_list.txt', ann_file='train.txt', pipeline=[ dict(type='LoadImageFromFile'), dict(type='LoadKIEAnnotations'), dict(type='Resize', scale=(1024, 512), keep_ratio=True), dict(type='PackKIEInputs') ])) val_dataloader = dict( batch_size=1, num_workers=1, persistent_workers=True, sampler=dict(type='DefaultSampler', shuffle=False), dataset=dict( type='WildReceiptDataset', data_root='data/kie/wildreceipt/', metainfo='data/kie/wildreceipt/class_list.txt', ann_file='test.txt', test_mode=True, pipeline=[ dict(type='LoadImageFromFile'), dict(type='LoadKIEAnnotations'), dict(type='Resize', scale=(1024, 512), keep_ratio=True), dict(type='PackKIEInputs') ])) test_dataloader = dict( batch_size=1, num_workers=1, persistent_workers=True, sampler=dict(type='DefaultSampler', shuffle=False), dataset=dict( type='WildReceiptDataset', data_root='data/kie/wildreceipt/', metainfo='data/kie/wildreceipt/class_list.txt', ann_file='test.txt', test_mode=True, pipeline=[ dict(type='LoadImageFromFile'), dict(type='LoadKIEAnnotations'), dict(type='Resize', scale=(1024, 512), keep_ratio=True), dict(type='PackKIEInputs') ])) launcher = 'none' work_dir = './work_dirs/sdmgr_unet16_60e_wildreceipt' 2022/08/25 15:17:06 - mmengine - INFO - Distributed training is not used, all SyncBatchNorm (SyncBN) layers in the model will be automatically reverted to BatchNormXd layers if they are used. Name of parameter - Initialization information backbone.encoder.0.0.convs.0.conv.weight - torch.Size([16, 3, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.encoder.0.0.convs.0.bn.weight - torch.Size([16]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.0.0.convs.0.bn.bias - torch.Size([16]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.0.0.convs.1.conv.weight - torch.Size([16, 16, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.encoder.0.0.convs.1.bn.weight - torch.Size([16]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.0.0.convs.1.bn.bias - torch.Size([16]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.1.1.convs.0.conv.weight - torch.Size([32, 16, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.encoder.1.1.convs.0.bn.weight - torch.Size([32]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.1.1.convs.0.bn.bias - torch.Size([32]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.1.1.convs.1.conv.weight - torch.Size([32, 32, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.encoder.1.1.convs.1.bn.weight - torch.Size([32]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.1.1.convs.1.bn.bias - torch.Size([32]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.2.1.convs.0.conv.weight - torch.Size([64, 32, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.encoder.2.1.convs.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.2.1.convs.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.2.1.convs.1.conv.weight - torch.Size([64, 64, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.encoder.2.1.convs.1.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.2.1.convs.1.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.3.1.convs.0.conv.weight - torch.Size([128, 64, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.encoder.3.1.convs.0.bn.weight - torch.Size([128]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.3.1.convs.0.bn.bias - torch.Size([128]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.3.1.convs.1.conv.weight - torch.Size([128, 128, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.encoder.3.1.convs.1.bn.weight - torch.Size([128]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.3.1.convs.1.bn.bias - torch.Size([128]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.4.1.convs.0.conv.weight - torch.Size([256, 128, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.encoder.4.1.convs.0.bn.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.4.1.convs.0.bn.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.4.1.convs.1.conv.weight - torch.Size([256, 256, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.encoder.4.1.convs.1.bn.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of SDMGR backbone.encoder.4.1.convs.1.bn.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.0.conv_block.convs.0.conv.weight - torch.Size([16, 32, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.decoder.0.conv_block.convs.0.bn.weight - torch.Size([16]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.0.conv_block.convs.0.bn.bias - torch.Size([16]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.0.conv_block.convs.1.conv.weight - torch.Size([16, 16, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.decoder.0.conv_block.convs.1.bn.weight - torch.Size([16]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.0.conv_block.convs.1.bn.bias - torch.Size([16]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.0.upsample.interp_upsample.1.conv.weight - torch.Size([16, 32, 1, 1]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.decoder.0.upsample.interp_upsample.1.bn.weight - torch.Size([16]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.0.upsample.interp_upsample.1.bn.bias - torch.Size([16]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.1.conv_block.convs.0.conv.weight - torch.Size([32, 64, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.decoder.1.conv_block.convs.0.bn.weight - torch.Size([32]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.1.conv_block.convs.0.bn.bias - torch.Size([32]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.1.conv_block.convs.1.conv.weight - torch.Size([32, 32, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.decoder.1.conv_block.convs.1.bn.weight - torch.Size([32]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.1.conv_block.convs.1.bn.bias - torch.Size([32]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.1.upsample.interp_upsample.1.conv.weight - torch.Size([32, 64, 1, 1]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.decoder.1.upsample.interp_upsample.1.bn.weight - torch.Size([32]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.1.upsample.interp_upsample.1.bn.bias - torch.Size([32]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.2.conv_block.convs.0.conv.weight - torch.Size([64, 128, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.decoder.2.conv_block.convs.0.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.2.conv_block.convs.0.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.2.conv_block.convs.1.conv.weight - torch.Size([64, 64, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.decoder.2.conv_block.convs.1.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.2.conv_block.convs.1.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.2.upsample.interp_upsample.1.conv.weight - torch.Size([64, 128, 1, 1]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.decoder.2.upsample.interp_upsample.1.bn.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.2.upsample.interp_upsample.1.bn.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.3.conv_block.convs.0.conv.weight - torch.Size([128, 256, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.decoder.3.conv_block.convs.0.bn.weight - torch.Size([128]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.3.conv_block.convs.0.bn.bias - torch.Size([128]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.3.conv_block.convs.1.conv.weight - torch.Size([128, 128, 3, 3]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.decoder.3.conv_block.convs.1.bn.weight - torch.Size([128]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.3.conv_block.convs.1.bn.bias - torch.Size([128]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.3.upsample.interp_upsample.1.conv.weight - torch.Size([128, 256, 1, 1]): KaimingInit: a=0, mode=fan_out, nonlinearity=relu, distribution =normal, bias=0 backbone.decoder.3.upsample.interp_upsample.1.bn.weight - torch.Size([128]): The value is the same before and after calling `init_weights` of SDMGR backbone.decoder.3.upsample.interp_upsample.1.bn.bias - torch.Size([128]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.linear0.weight - torch.Size([1024, 16]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.linear0.bias - torch.Size([1024]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.linear1.weight - torch.Size([1024, 256]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.linear1.bias - torch.Size([1024]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.0.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.0.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.1.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.1.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.2.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.2.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.3.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.3.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.4.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.4.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.5.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.5.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.6.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.6.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.7.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.7.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.8.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.8.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.9.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.9.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.10.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.10.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.11.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.11.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.12.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.12.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.13.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.13.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.14.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.14.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.15.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.15.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.16.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.16.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.17.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.17.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.18.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.18.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.19.weight - torch.Size([540, 36]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears0.19.bias - torch.Size([540]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.0.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.0.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.1.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.1.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.2.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.2.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.3.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.3.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.4.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.4.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.5.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.5.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.6.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.6.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.7.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.7.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.8.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.8.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.9.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.9.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.10.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.10.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.11.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.11.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.12.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.12.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.13.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.13.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.14.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.14.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.15.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.15.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.16.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.16.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.17.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.17.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.18.weight - torch.Size([780, 52]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.18.bias - torch.Size([780]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.19.weight - torch.Size([540, 36]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.merge_linears1.19.bias - torch.Size([540]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.linear_out.weight - torch.Size([256, 1024]): The value is the same before and after calling `init_weights` of SDMGR kie_head.fusion.linear_out.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SDMGR kie_head.node_embed.weight - torch.Size([92, 32]): The value is the same before and after calling `init_weights` of SDMGR kie_head.rnn.weight_ih_l0 - torch.Size([1024, 32]): The value is the same before and after calling `init_weights` of SDMGR kie_head.rnn.weight_hh_l0 - torch.Size([1024, 256]): The value is the same before and after calling `init_weights` of SDMGR kie_head.rnn.bias_ih_l0 - torch.Size([1024]): The value is the same before and after calling `init_weights` of SDMGR kie_head.rnn.bias_hh_l0 - torch.Size([1024]): The value is the same before and after calling `init_weights` of SDMGR kie_head.edge_embed.weight - torch.Size([256, 5]): NormalInit: mean=0, std=0.01, bias=0 kie_head.edge_embed.bias - torch.Size([256]): NormalInit: mean=0, std=0.01, bias=0 kie_head.gnn_layers.0.in_fc.weight - torch.Size([256, 768]): The value is the same before and after calling `init_weights` of SDMGR kie_head.gnn_layers.0.in_fc.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SDMGR kie_head.gnn_layers.0.coef_fc.weight - torch.Size([1, 256]): The value is the same before and after calling `init_weights` of SDMGR kie_head.gnn_layers.0.coef_fc.bias - torch.Size([1]): The value is the same before and after calling `init_weights` of SDMGR kie_head.gnn_layers.0.out_fc.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SDMGR kie_head.gnn_layers.0.out_fc.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SDMGR kie_head.gnn_layers.1.in_fc.weight - torch.Size([256, 768]): The value is the same before and after calling `init_weights` of SDMGR kie_head.gnn_layers.1.in_fc.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SDMGR kie_head.gnn_layers.1.coef_fc.weight - torch.Size([1, 256]): The value is the same before and after calling `init_weights` of SDMGR kie_head.gnn_layers.1.coef_fc.bias - torch.Size([1]): The value is the same before and after calling `init_weights` of SDMGR kie_head.gnn_layers.1.out_fc.weight - torch.Size([256, 256]): The value is the same before and after calling `init_weights` of SDMGR kie_head.gnn_layers.1.out_fc.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of SDMGR kie_head.node_cls.weight - torch.Size([26, 256]): The value is the same before and after calling `init_weights` of SDMGR kie_head.node_cls.bias - torch.Size([26]): The value is the same before and after calling `init_weights` of SDMGR kie_head.edge_cls.weight - torch.Size([2, 256]): The value is the same before and after calling `init_weights` of SDMGR kie_head.edge_cls.bias - torch.Size([2]): The value is the same before and after calling `init_weights` of SDMGR 2022/08/25 15:18:15 - mmengine - INFO - Epoch(train) [1][100/317] lr: 1.0000e-03 eta: 3:16:37 time: 0.4245 data_time: 0.0153 memory: 20087 loss_node: 0.9884 loss_edge: 0.0000 acc_node: 68.2051 acc_edge: 100.0000 loss: 0.9884 2022/08/25 15:18:48 - mmengine - INFO - Epoch(train) [1][200/317] lr: 1.0000e-03 eta: 2:28:29 time: 0.1618 data_time: 0.0054 memory: 19528 loss_node: 0.6793 loss_edge: 0.0001 acc_node: 87.6289 acc_edge: 100.0000 loss: 0.6793 2022/08/25 15:19:21 - mmengine - INFO - Epoch(train) [1][300/317] lr: 1.0000e-03 eta: 2:12:55 time: 0.2228 data_time: 0.0044 memory: 19640 loss_node: 0.6782 loss_edge: 0.0000 acc_node: 81.8841 acc_edge: 100.0000 loss: 0.6782 2022/08/25 15:19:29 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:19:29 - mmengine - INFO - Saving checkpoint at 1 epochs 2022/08/25 15:19:45 - mmengine - INFO - Epoch(val) [1][100/472] eta: 0:02:05 time: 0.3379 data_time: 0.1579 memory: 19081 2022/08/25 15:19:55 - mmengine - INFO - Epoch(val) [1][200/472] eta: 0:00:39 time: 0.1437 data_time: 0.1006 memory: 16791 2022/08/25 15:20:05 - mmengine - INFO - Epoch(val) [1][300/472] eta: 0:00:13 time: 0.0775 data_time: 0.0564 memory: 4351 2022/08/25 15:20:14 - mmengine - INFO - Epoch(val) [1][400/472] eta: 0:00:18 time: 0.2613 data_time: 0.1039 memory: 4361 2022/08/25 15:20:21 - mmengine - INFO - Epoch(val) [1][472/472] kie/macro_f1: 0.6412 2022/08/25 15:20:47 - mmengine - INFO - Epoch(train) [2][100/317] lr: 1.0000e-03 eta: 1:56:40 time: 0.1589 data_time: 0.0053 memory: 19864 loss_node: 0.5441 loss_edge: 0.0001 acc_node: 88.8199 acc_edge: 100.0000 loss: 0.5441 2022/08/25 15:21:11 - mmengine - INFO - Epoch(train) [2][200/317] lr: 1.0000e-03 eta: 1:48:14 time: 0.1546 data_time: 0.0056 memory: 18858 loss_node: 0.5287 loss_edge: 0.0001 acc_node: 84.2730 acc_edge: 100.0000 loss: 0.5288 2022/08/25 15:21:47 - mmengine - INFO - Epoch(train) [2][300/317] lr: 1.0000e-03 eta: 1:48:12 time: 0.2892 data_time: 0.0052 memory: 20200 loss_node: 0.4881 loss_edge: 0.0000 acc_node: 83.8028 acc_edge: 100.0000 loss: 0.4881 2022/08/25 15:21:51 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:21:51 - mmengine - INFO - Saving checkpoint at 2 epochs 2022/08/25 15:22:00 - mmengine - INFO - Epoch(val) [2][100/472] eta: 0:00:17 time: 0.0457 data_time: 0.0238 memory: 4875 2022/08/25 15:22:09 - mmengine - INFO - Epoch(val) [2][200/472] eta: 0:00:13 time: 0.0494 data_time: 0.0242 memory: 353 2022/08/25 15:22:19 - mmengine - INFO - Epoch(val) [2][300/472] eta: 0:00:07 time: 0.0454 data_time: 0.0218 memory: 388 2022/08/25 15:22:25 - mmengine - INFO - Epoch(val) [2][400/472] eta: 0:00:01 time: 0.0260 data_time: 0.0042 memory: 353 2022/08/25 15:22:29 - mmengine - INFO - Epoch(val) [2][472/472] kie/macro_f1: 0.7411 2022/08/25 15:22:57 - mmengine - INFO - Epoch(train) [3][100/317] lr: 1.0000e-03 eta: 1:42:29 time: 0.1457 data_time: 0.0045 memory: 9879 loss_node: 0.4939 loss_edge: 0.0001 acc_node: 92.0863 acc_edge: 100.0000 loss: 0.4940 2022/08/25 15:23:26 - mmengine - INFO - Epoch(train) [3][200/317] lr: 1.0000e-03 eta: 1:40:24 time: 0.1568 data_time: 0.0049 memory: 11840 loss_node: 0.4676 loss_edge: 0.0001 acc_node: 78.6765 acc_edge: 100.0000 loss: 0.4676 2022/08/25 15:23:51 - mmengine - INFO - Epoch(train) [3][300/317] lr: 1.0000e-03 eta: 1:37:13 time: 0.3146 data_time: 0.0143 memory: 10271 loss_node: 0.4062 loss_edge: 0.0000 acc_node: 87.8788 acc_edge: 100.0000 loss: 0.4062 2022/08/25 15:23:54 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:23:54 - mmengine - INFO - Saving checkpoint at 3 epochs 2022/08/25 15:24:07 - mmengine - INFO - Epoch(val) [3][100/472] eta: 0:00:20 time: 0.0562 data_time: 0.0352 memory: 5966 2022/08/25 15:24:15 - mmengine - INFO - Epoch(val) [3][200/472] eta: 0:00:17 time: 0.0641 data_time: 0.0415 memory: 353 2022/08/25 15:24:21 - mmengine - INFO - Epoch(val) [3][300/472] eta: 0:00:18 time: 0.1049 data_time: 0.0815 memory: 388 2022/08/25 15:24:27 - mmengine - INFO - Epoch(val) [3][400/472] eta: 0:00:03 time: 0.0434 data_time: 0.0187 memory: 353 2022/08/25 15:24:31 - mmengine - INFO - Epoch(val) [3][472/472] kie/macro_f1: 0.7796 2022/08/25 15:24:48 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:24:58 - mmengine - INFO - Epoch(train) [4][100/317] lr: 1.0000e-03 eta: 1:34:18 time: 0.1615 data_time: 0.0049 memory: 18970 loss_node: 0.3480 loss_edge: 0.0000 acc_node: 84.3023 acc_edge: 100.0000 loss: 0.3480 2022/08/25 15:25:21 - mmengine - INFO - Epoch(train) [4][200/317] lr: 1.0000e-03 eta: 1:31:28 time: 0.3773 data_time: 0.0048 memory: 20423 loss_node: 0.4841 loss_edge: 0.0001 acc_node: 81.2500 acc_edge: 100.0000 loss: 0.4841 2022/08/25 15:25:43 - mmengine - INFO - Epoch(train) [4][300/317] lr: 1.0000e-03 eta: 1:28:52 time: 0.3034 data_time: 0.0044 memory: 8837 loss_node: 0.3910 loss_edge: 0.0001 acc_node: 91.1765 acc_edge: 100.0000 loss: 0.3911 2022/08/25 15:25:47 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:25:47 - mmengine - INFO - Saving checkpoint at 4 epochs 2022/08/25 15:25:52 - mmengine - INFO - Epoch(val) [4][100/472] eta: 0:00:10 time: 0.0294 data_time: 0.0072 memory: 7057 2022/08/25 15:25:55 - mmengine - INFO - Epoch(val) [4][200/472] eta: 0:00:09 time: 0.0354 data_time: 0.0118 memory: 353 2022/08/25 15:25:59 - mmengine - INFO - Epoch(val) [4][300/472] eta: 0:00:10 time: 0.0596 data_time: 0.0361 memory: 388 2022/08/25 15:26:03 - mmengine - INFO - Epoch(val) [4][400/472] eta: 0:00:05 time: 0.0818 data_time: 0.0173 memory: 353 2022/08/25 15:26:06 - mmengine - INFO - Epoch(val) [4][472/472] kie/macro_f1: 0.7837 2022/08/25 15:26:30 - mmengine - INFO - Epoch(train) [5][100/317] lr: 1.0000e-03 eta: 1:26:24 time: 0.1627 data_time: 0.0051 memory: 11056 loss_node: 0.2866 loss_edge: 0.0000 acc_node: 89.1026 acc_edge: 100.0000 loss: 0.2866 2022/08/25 15:26:52 - mmengine - INFO - Epoch(train) [5][200/317] lr: 1.0000e-03 eta: 1:24:31 time: 0.1572 data_time: 0.0043 memory: 18859 loss_node: 0.3405 loss_edge: 0.0000 acc_node: 91.6667 acc_edge: 100.0000 loss: 0.3405 2022/08/25 15:27:11 - mmengine - INFO - Epoch(train) [5][300/317] lr: 1.0000e-03 eta: 1:22:13 time: 0.1573 data_time: 0.0044 memory: 7057 loss_node: 0.4039 loss_edge: 0.0000 acc_node: 85.0877 acc_edge: 100.0000 loss: 0.4039 2022/08/25 15:27:15 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:27:15 - mmengine - INFO - Saving checkpoint at 5 epochs 2022/08/25 15:27:20 - mmengine - INFO - Epoch(val) [5][100/472] eta: 0:00:09 time: 0.0243 data_time: 0.0045 memory: 7057 2022/08/25 15:27:23 - mmengine - INFO - Epoch(val) [5][200/472] eta: 0:00:13 time: 0.0506 data_time: 0.0284 memory: 353 2022/08/25 15:27:26 - mmengine - INFO - Epoch(val) [5][300/472] eta: 0:00:07 time: 0.0438 data_time: 0.0215 memory: 388 2022/08/25 15:27:29 - mmengine - INFO - Epoch(val) [5][400/472] eta: 0:00:01 time: 0.0248 data_time: 0.0022 memory: 353 2022/08/25 15:27:31 - mmengine - INFO - Epoch(val) [5][472/472] kie/macro_f1: 0.7947 2022/08/25 15:27:51 - mmengine - INFO - Epoch(train) [6][100/317] lr: 1.0000e-03 eta: 1:19:43 time: 0.1607 data_time: 0.0051 memory: 7057 loss_node: 0.2930 loss_edge: 0.0000 acc_node: 96.5116 acc_edge: 100.0000 loss: 0.2930 2022/08/25 15:28:11 - mmengine - INFO - Epoch(train) [6][200/317] lr: 1.0000e-03 eta: 1:18:00 time: 0.1669 data_time: 0.0046 memory: 9203 loss_node: 0.3568 loss_edge: 0.0000 acc_node: 88.9952 acc_edge: 100.0000 loss: 0.3569 2022/08/25 15:28:30 - mmengine - INFO - Epoch(train) [6][300/317] lr: 1.0000e-03 eta: 1:16:25 time: 0.1610 data_time: 0.0047 memory: 7780 loss_node: 0.3574 loss_edge: 0.0000 acc_node: 94.2408 acc_edge: 100.0000 loss: 0.3574 2022/08/25 15:28:36 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:28:36 - mmengine - INFO - Saving checkpoint at 6 epochs 2022/08/25 15:28:42 - mmengine - INFO - Epoch(val) [6][100/472] eta: 0:00:14 time: 0.0392 data_time: 0.0166 memory: 7107 2022/08/25 15:28:45 - mmengine - INFO - Epoch(val) [6][200/472] eta: 0:00:17 time: 0.0634 data_time: 0.0413 memory: 353 2022/08/25 15:28:49 - mmengine - INFO - Epoch(val) [6][300/472] eta: 0:00:09 time: 0.0535 data_time: 0.0300 memory: 388 2022/08/25 15:28:52 - mmengine - INFO - Epoch(val) [6][400/472] eta: 0:00:01 time: 0.0255 data_time: 0.0025 memory: 353 2022/08/25 15:28:57 - mmengine - INFO - Epoch(val) [6][472/472] kie/macro_f1: 0.8008 2022/08/25 15:29:22 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:29:23 - mmengine - INFO - Epoch(train) [7][100/317] lr: 1.0000e-03 eta: 1:15:33 time: 0.2755 data_time: 0.0066 memory: 19417 loss_node: 0.2469 loss_edge: 0.0000 acc_node: 96.2264 acc_edge: 100.0000 loss: 0.2470 2022/08/25 15:29:41 - mmengine - INFO - Epoch(train) [7][200/317] lr: 1.0000e-03 eta: 1:13:52 time: 0.2066 data_time: 0.0104 memory: 7100 loss_node: 0.3483 loss_edge: 0.0000 acc_node: 92.8144 acc_edge: 100.0000 loss: 0.3483 2022/08/25 15:30:09 - mmengine - INFO - Epoch(train) [7][300/317] lr: 1.0000e-03 eta: 1:13:36 time: 0.3199 data_time: 0.0707 memory: 7057 loss_node: 0.3031 loss_edge: 0.0000 acc_node: 93.7500 acc_edge: 100.0000 loss: 0.3032 2022/08/25 15:30:15 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:30:15 - mmengine - INFO - Saving checkpoint at 7 epochs 2022/08/25 15:30:25 - mmengine - INFO - Epoch(val) [7][100/472] eta: 0:01:20 time: 0.2155 data_time: 0.1993 memory: 5679 2022/08/25 15:30:32 - mmengine - INFO - Epoch(val) [7][200/472] eta: 0:00:24 time: 0.0917 data_time: 0.0659 memory: 353 2022/08/25 15:30:41 - mmengine - INFO - Epoch(val) [7][300/472] eta: 0:00:08 time: 0.0480 data_time: 0.0237 memory: 388 2022/08/25 15:30:46 - mmengine - INFO - Epoch(val) [7][400/472] eta: 0:00:02 time: 0.0335 data_time: 0.0133 memory: 353 2022/08/25 15:30:50 - mmengine - INFO - Epoch(val) [7][472/472] kie/macro_f1: 0.7880 2022/08/25 15:31:20 - mmengine - INFO - Epoch(train) [8][100/317] lr: 1.0000e-03 eta: 1:13:25 time: 0.2752 data_time: 0.1120 memory: 8865 loss_node: 0.3234 loss_edge: 0.0001 acc_node: 93.4132 acc_edge: 100.0000 loss: 0.3235 2022/08/25 15:31:47 - mmengine - INFO - Epoch(train) [8][200/317] lr: 1.0000e-03 eta: 1:13:04 time: 0.2748 data_time: 0.0059 memory: 7057 loss_node: 0.2455 loss_edge: 0.0000 acc_node: 91.8129 acc_edge: 100.0000 loss: 0.2456 2022/08/25 15:32:12 - mmengine - INFO - Epoch(train) [8][300/317] lr: 1.0000e-03 eta: 1:12:28 time: 0.2390 data_time: 0.0051 memory: 8513 loss_node: 0.2795 loss_edge: 0.0000 acc_node: 90.2256 acc_edge: 100.0000 loss: 0.2796 2022/08/25 15:32:16 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:32:16 - mmengine - INFO - Saving checkpoint at 8 epochs 2022/08/25 15:32:24 - mmengine - INFO - Epoch(val) [8][100/472] eta: 0:00:57 time: 0.1535 data_time: 0.1289 memory: 8784 2022/08/25 15:32:28 - mmengine - INFO - Epoch(val) [8][200/472] eta: 0:00:11 time: 0.0424 data_time: 0.0170 memory: 353 2022/08/25 15:32:32 - mmengine - INFO - Epoch(val) [8][300/472] eta: 0:00:07 time: 0.0434 data_time: 0.0223 memory: 388 2022/08/25 15:32:37 - mmengine - INFO - Epoch(val) [8][400/472] eta: 0:00:02 time: 0.0409 data_time: 0.0195 memory: 353 2022/08/25 15:32:40 - mmengine - INFO - Epoch(val) [8][472/472] kie/macro_f1: 0.7923 2022/08/25 15:33:01 - mmengine - INFO - Epoch(train) [9][100/317] lr: 1.0000e-03 eta: 1:11:08 time: 0.1715 data_time: 0.0060 memory: 7057 loss_node: 0.3244 loss_edge: 0.0000 acc_node: 95.3192 acc_edge: 100.0000 loss: 0.3245 2022/08/25 15:33:20 - mmengine - INFO - Epoch(train) [9][200/317] lr: 1.0000e-03 eta: 1:09:58 time: 0.1544 data_time: 0.0056 memory: 7057 loss_node: 0.2701 loss_edge: 0.0000 acc_node: 88.4259 acc_edge: 100.0000 loss: 0.2701 2022/08/25 15:33:41 - mmengine - INFO - Epoch(train) [9][300/317] lr: 1.0000e-03 eta: 1:09:06 time: 0.1534 data_time: 0.0041 memory: 6837 loss_node: 0.2962 loss_edge: 0.0000 acc_node: 92.8105 acc_edge: 100.0000 loss: 0.2962 2022/08/25 15:33:45 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:33:45 - mmengine - INFO - Saving checkpoint at 9 epochs 2022/08/25 15:33:51 - mmengine - INFO - Epoch(val) [9][100/472] eta: 0:00:10 time: 0.0290 data_time: 0.0074 memory: 18123 2022/08/25 15:33:54 - mmengine - INFO - Epoch(val) [9][200/472] eta: 0:00:08 time: 0.0295 data_time: 0.0106 memory: 353 2022/08/25 15:33:57 - mmengine - INFO - Epoch(val) [9][300/472] eta: 0:00:06 time: 0.0385 data_time: 0.0175 memory: 388 2022/08/25 15:34:00 - mmengine - INFO - Epoch(val) [9][400/472] eta: 0:00:02 time: 0.0280 data_time: 0.0076 memory: 353 2022/08/25 15:34:02 - mmengine - INFO - Epoch(val) [9][472/472] kie/macro_f1: 0.8133 2022/08/25 15:34:20 - mmengine - INFO - Epoch(train) [10][100/317] lr: 1.0000e-03 eta: 1:07:38 time: 0.1565 data_time: 0.0055 memory: 7146 loss_node: 0.3248 loss_edge: 0.0000 acc_node: 87.6923 acc_edge: 100.0000 loss: 0.3248 2022/08/25 15:34:30 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:34:40 - mmengine - INFO - Epoch(train) [10][200/317] lr: 1.0000e-03 eta: 1:06:45 time: 0.1731 data_time: 0.0055 memory: 6621 loss_node: 0.2337 loss_edge: 0.0000 acc_node: 90.9548 acc_edge: 100.0000 loss: 0.2337 2022/08/25 15:35:00 - mmengine - INFO - Epoch(train) [10][300/317] lr: 1.0000e-03 eta: 1:05:51 time: 0.1591 data_time: 0.0042 memory: 6403 loss_node: 0.2875 loss_edge: 0.0000 acc_node: 92.4051 acc_edge: 100.0000 loss: 0.2876 2022/08/25 15:35:05 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:35:05 - mmengine - INFO - Saving checkpoint at 10 epochs 2022/08/25 15:35:10 - mmengine - INFO - Epoch(val) [10][100/472] eta: 0:00:09 time: 0.0246 data_time: 0.0038 memory: 9568 2022/08/25 15:35:13 - mmengine - INFO - Epoch(val) [10][200/472] eta: 0:00:09 time: 0.0355 data_time: 0.0117 memory: 353 2022/08/25 15:35:17 - mmengine - INFO - Epoch(val) [10][300/472] eta: 0:00:08 time: 0.0502 data_time: 0.0288 memory: 388 2022/08/25 15:35:20 - mmengine - INFO - Epoch(val) [10][400/472] eta: 0:00:02 time: 0.0325 data_time: 0.0106 memory: 353 2022/08/25 15:35:24 - mmengine - INFO - Epoch(val) [10][472/472] kie/macro_f1: 0.8360 2022/08/25 15:35:45 - mmengine - INFO - Epoch(train) [11][100/317] lr: 1.0000e-03 eta: 1:04:49 time: 0.2961 data_time: 0.0045 memory: 8028 loss_node: 0.2671 loss_edge: 0.0000 acc_node: 92.9730 acc_edge: 100.0000 loss: 0.2671 2022/08/25 15:36:06 - mmengine - INFO - Epoch(train) [11][200/317] lr: 1.0000e-03 eta: 1:04:10 time: 0.4642 data_time: 0.0047 memory: 8473 loss_node: 0.3445 loss_edge: 0.0000 acc_node: 92.4138 acc_edge: 100.0000 loss: 0.3445 2022/08/25 15:36:26 - mmengine - INFO - Epoch(train) [11][300/317] lr: 1.0000e-03 eta: 1:03:23 time: 0.4033 data_time: 0.0233 memory: 6403 loss_node: 0.2543 loss_edge: 0.0000 acc_node: 83.7209 acc_edge: 100.0000 loss: 0.2544 2022/08/25 15:36:30 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:36:30 - mmengine - INFO - Saving checkpoint at 11 epochs 2022/08/25 15:36:36 - mmengine - INFO - Epoch(val) [11][100/472] eta: 0:00:39 time: 0.1057 data_time: 0.0823 memory: 8144 2022/08/25 15:36:39 - mmengine - INFO - Epoch(val) [11][200/472] eta: 0:00:10 time: 0.0402 data_time: 0.0153 memory: 353 2022/08/25 15:36:42 - mmengine - INFO - Epoch(val) [11][300/472] eta: 0:00:06 time: 0.0399 data_time: 0.0168 memory: 388 2022/08/25 15:36:48 - mmengine - INFO - Epoch(val) [11][400/472] eta: 0:00:03 time: 0.0447 data_time: 0.0196 memory: 353 2022/08/25 15:36:52 - mmengine - INFO - Epoch(val) [11][472/472] kie/macro_f1: 0.8168 2022/08/25 15:37:11 - mmengine - INFO - Epoch(train) [12][100/317] lr: 1.0000e-03 eta: 1:02:25 time: 0.1654 data_time: 0.0049 memory: 7057 loss_node: 0.2518 loss_edge: 0.0000 acc_node: 92.1875 acc_edge: 100.0000 loss: 0.2518 2022/08/25 15:37:38 - mmengine - INFO - Epoch(train) [12][200/317] lr: 1.0000e-03 eta: 1:02:09 time: 0.2518 data_time: 0.0048 memory: 20311 loss_node: 0.2742 loss_edge: 0.0000 acc_node: 92.5000 acc_edge: 100.0000 loss: 0.2742 2022/08/25 15:37:59 - mmengine - INFO - Epoch(train) [12][300/317] lr: 1.0000e-03 eta: 1:01:31 time: 0.1530 data_time: 0.0069 memory: 8189 loss_node: 0.2304 loss_edge: 0.0000 acc_node: 93.2886 acc_edge: 100.0000 loss: 0.2304 2022/08/25 15:38:01 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:38:01 - mmengine - INFO - Saving checkpoint at 12 epochs 2022/08/25 15:38:08 - mmengine - INFO - Epoch(val) [12][100/472] eta: 0:00:15 time: 0.0405 data_time: 0.0193 memory: 6577 2022/08/25 15:38:11 - mmengine - INFO - Epoch(val) [12][200/472] eta: 0:00:12 time: 0.0471 data_time: 0.0244 memory: 353 2022/08/25 15:38:17 - mmengine - INFO - Epoch(val) [12][300/472] eta: 0:00:06 time: 0.0349 data_time: 0.0132 memory: 388 2022/08/25 15:38:20 - mmengine - INFO - Epoch(val) [12][400/472] eta: 0:00:01 time: 0.0252 data_time: 0.0023 memory: 353 2022/08/25 15:38:22 - mmengine - INFO - Epoch(val) [12][472/472] kie/macro_f1: 0.8386 2022/08/25 15:38:43 - mmengine - INFO - Epoch(train) [13][100/317] lr: 1.0000e-03 eta: 1:00:39 time: 0.1710 data_time: 0.0052 memory: 7526 loss_node: 0.2237 loss_edge: 0.0000 acc_node: 93.1034 acc_edge: 100.0000 loss: 0.2237 2022/08/25 15:39:05 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:39:06 - mmengine - INFO - Epoch(train) [13][200/317] lr: 1.0000e-03 eta: 1:00:11 time: 0.1795 data_time: 0.0109 memory: 6624 loss_node: 0.2249 loss_edge: 0.0000 acc_node: 92.4107 acc_edge: 100.0000 loss: 0.2250 2022/08/25 15:39:28 - mmengine - INFO - Epoch(train) [13][300/317] lr: 1.0000e-03 eta: 0:59:39 time: 0.1457 data_time: 0.0048 memory: 7269 loss_node: 0.2441 loss_edge: 0.0000 acc_node: 95.2569 acc_edge: 100.0000 loss: 0.2441 2022/08/25 15:39:35 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:39:35 - mmengine - INFO - Saving checkpoint at 13 epochs 2022/08/25 15:39:46 - mmengine - INFO - Epoch(val) [13][100/472] eta: 0:00:41 time: 0.1127 data_time: 0.0907 memory: 5854 2022/08/25 15:39:55 - mmengine - INFO - Epoch(val) [13][200/472] eta: 0:01:35 time: 0.3523 data_time: 0.0295 memory: 353 2022/08/25 15:40:01 - mmengine - INFO - Epoch(val) [13][300/472] eta: 0:00:15 time: 0.0914 data_time: 0.0674 memory: 388 2022/08/25 15:40:07 - mmengine - INFO - Epoch(val) [13][400/472] eta: 0:00:03 time: 0.0529 data_time: 0.0279 memory: 353 2022/08/25 15:40:10 - mmengine - INFO - Epoch(val) [13][472/472] kie/macro_f1: 0.8589 2022/08/25 15:40:34 - mmengine - INFO - Epoch(train) [14][100/317] lr: 1.0000e-03 eta: 0:59:12 time: 0.1686 data_time: 0.0049 memory: 9933 loss_node: 0.2425 loss_edge: 0.0000 acc_node: 94.0298 acc_edge: 100.0000 loss: 0.2425 2022/08/25 15:40:56 - mmengine - INFO - Epoch(train) [14][200/317] lr: 1.0000e-03 eta: 0:58:41 time: 0.1486 data_time: 0.0052 memory: 10217 loss_node: 0.2237 loss_edge: 0.0000 acc_node: 95.1807 acc_edge: 100.0000 loss: 0.2237 2022/08/25 15:41:18 - mmengine - INFO - Epoch(train) [14][300/317] lr: 1.0000e-03 eta: 0:58:11 time: 0.1605 data_time: 0.0051 memory: 7228 loss_node: 0.2341 loss_edge: 0.0000 acc_node: 97.2973 acc_edge: 100.0000 loss: 0.2342 2022/08/25 15:41:22 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:41:22 - mmengine - INFO - Saving checkpoint at 14 epochs 2022/08/25 15:41:31 - mmengine - INFO - Epoch(val) [14][100/472] eta: 0:00:12 time: 0.0348 data_time: 0.0124 memory: 5753 2022/08/25 15:41:38 - mmengine - INFO - Epoch(val) [14][200/472] eta: 0:01:24 time: 0.3111 data_time: 0.0645 memory: 353 2022/08/25 15:41:42 - mmengine - INFO - Epoch(val) [14][300/472] eta: 0:00:07 time: 0.0462 data_time: 0.0236 memory: 388 2022/08/25 15:41:46 - mmengine - INFO - Epoch(val) [14][400/472] eta: 0:00:01 time: 0.0248 data_time: 0.0017 memory: 353 2022/08/25 15:41:49 - mmengine - INFO - Epoch(val) [14][472/472] kie/macro_f1: 0.8512 2022/08/25 15:42:10 - mmengine - INFO - Epoch(train) [15][100/317] lr: 1.0000e-03 eta: 0:57:27 time: 0.1493 data_time: 0.0048 memory: 7057 loss_node: 0.2262 loss_edge: 0.0001 acc_node: 97.9167 acc_edge: 100.0000 loss: 0.2263 2022/08/25 15:42:28 - mmengine - INFO - Epoch(train) [15][200/317] lr: 1.0000e-03 eta: 0:56:44 time: 0.1464 data_time: 0.0048 memory: 5498 loss_node: 0.1954 loss_edge: 0.0000 acc_node: 93.2990 acc_edge: 100.0000 loss: 0.1954 2022/08/25 15:42:50 - mmengine - INFO - Epoch(train) [15][300/317] lr: 1.0000e-03 eta: 0:56:16 time: 0.1858 data_time: 0.0053 memory: 9542 loss_node: 0.2516 loss_edge: 0.0000 acc_node: 92.7184 acc_edge: 100.0000 loss: 0.2516 2022/08/25 15:42:54 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:42:54 - mmengine - INFO - Saving checkpoint at 15 epochs 2022/08/25 15:42:59 - mmengine - INFO - Epoch(val) [15][100/472] eta: 0:00:09 time: 0.0258 data_time: 0.0053 memory: 6593 2022/08/25 15:43:02 - mmengine - INFO - Epoch(val) [15][200/472] eta: 0:00:12 time: 0.0455 data_time: 0.0212 memory: 353 2022/08/25 15:43:05 - mmengine - INFO - Epoch(val) [15][300/472] eta: 0:00:06 time: 0.0374 data_time: 0.0194 memory: 388 2022/08/25 15:43:09 - mmengine - INFO - Epoch(val) [15][400/472] eta: 0:00:02 time: 0.0291 data_time: 0.0071 memory: 353 2022/08/25 15:43:12 - mmengine - INFO - Epoch(val) [15][472/472] kie/macro_f1: 0.8509 2022/08/25 15:43:33 - mmengine - INFO - Epoch(train) [16][100/317] lr: 1.0000e-03 eta: 0:55:34 time: 0.1616 data_time: 0.0061 memory: 6403 loss_node: 0.1971 loss_edge: 0.0000 acc_node: 97.0370 acc_edge: 100.0000 loss: 0.1971 2022/08/25 15:43:51 - mmengine - INFO - Epoch(train) [16][200/317] lr: 1.0000e-03 eta: 0:54:54 time: 0.1711 data_time: 0.0049 memory: 6621 loss_node: 0.2910 loss_edge: 0.0000 acc_node: 90.2985 acc_edge: 100.0000 loss: 0.2911 2022/08/25 15:43:59 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:44:10 - mmengine - INFO - Epoch(train) [16][300/317] lr: 1.0000e-03 eta: 0:54:20 time: 0.1734 data_time: 0.0048 memory: 7057 loss_node: 0.2180 loss_edge: 0.0000 acc_node: 91.3295 acc_edge: 100.0000 loss: 0.2180 2022/08/25 15:44:15 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:44:15 - mmengine - INFO - Saving checkpoint at 16 epochs 2022/08/25 15:44:20 - mmengine - INFO - Epoch(val) [16][100/472] eta: 0:00:09 time: 0.0244 data_time: 0.0047 memory: 7622 2022/08/25 15:44:23 - mmengine - INFO - Epoch(val) [16][200/472] eta: 0:00:11 time: 0.0432 data_time: 0.0180 memory: 353 2022/08/25 15:44:27 - mmengine - INFO - Epoch(val) [16][300/472] eta: 0:00:06 time: 0.0368 data_time: 0.0135 memory: 388 2022/08/25 15:44:33 - mmengine - INFO - Epoch(val) [16][400/472] eta: 0:00:01 time: 0.0269 data_time: 0.0032 memory: 353 2022/08/25 15:44:37 - mmengine - INFO - Epoch(val) [16][472/472] kie/macro_f1: 0.8554 2022/08/25 15:45:02 - mmengine - INFO - Epoch(train) [17][100/317] lr: 1.0000e-03 eta: 0:53:52 time: 0.1623 data_time: 0.0050 memory: 18635 loss_node: 0.2116 loss_edge: 0.0000 acc_node: 93.6170 acc_edge: 100.0000 loss: 0.2117 2022/08/25 15:45:20 - mmengine - INFO - Epoch(train) [17][200/317] lr: 1.0000e-03 eta: 0:53:15 time: 0.1579 data_time: 0.0046 memory: 5748 loss_node: 0.2014 loss_edge: 0.0000 acc_node: 93.6709 acc_edge: 100.0000 loss: 0.2014 2022/08/25 15:45:40 - mmengine - INFO - Epoch(train) [17][300/317] lr: 1.0000e-03 eta: 0:52:45 time: 0.1586 data_time: 0.0046 memory: 7852 loss_node: 0.2303 loss_edge: 0.0000 acc_node: 97.6190 acc_edge: 100.0000 loss: 0.2303 2022/08/25 15:45:43 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:45:43 - mmengine - INFO - Saving checkpoint at 17 epochs 2022/08/25 15:45:48 - mmengine - INFO - Epoch(val) [17][100/472] eta: 0:00:11 time: 0.0303 data_time: 0.0096 memory: 5966 2022/08/25 15:45:51 - mmengine - INFO - Epoch(val) [17][200/472] eta: 0:00:10 time: 0.0397 data_time: 0.0170 memory: 353 2022/08/25 15:45:57 - mmengine - INFO - Epoch(val) [17][300/472] eta: 0:00:47 time: 0.2790 data_time: 0.0207 memory: 388 2022/08/25 15:46:00 - mmengine - INFO - Epoch(val) [17][400/472] eta: 0:00:01 time: 0.0245 data_time: 0.0014 memory: 353 2022/08/25 15:46:03 - mmengine - INFO - Epoch(val) [17][472/472] kie/macro_f1: 0.8595 2022/08/25 15:46:23 - mmengine - INFO - Epoch(train) [18][100/317] lr: 1.0000e-03 eta: 0:52:04 time: 0.1818 data_time: 0.0099 memory: 7057 loss_node: 0.1929 loss_edge: 0.0000 acc_node: 92.7632 acc_edge: 100.0000 loss: 0.1929 2022/08/25 15:46:45 - mmengine - INFO - Epoch(train) [18][200/317] lr: 1.0000e-03 eta: 0:51:39 time: 0.2239 data_time: 0.0050 memory: 8594 loss_node: 0.2205 loss_edge: 0.0000 acc_node: 95.0000 acc_edge: 100.0000 loss: 0.2206 2022/08/25 15:47:06 - mmengine - INFO - Epoch(train) [18][300/317] lr: 1.0000e-03 eta: 0:51:11 time: 0.2519 data_time: 0.0100 memory: 7057 loss_node: 0.1959 loss_edge: 0.0000 acc_node: 88.2979 acc_edge: 100.0000 loss: 0.1960 2022/08/25 15:47:09 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:47:09 - mmengine - INFO - Saving checkpoint at 18 epochs 2022/08/25 15:47:21 - mmengine - INFO - Epoch(val) [18][100/472] eta: 0:00:26 time: 0.0709 data_time: 0.0495 memory: 5372 2022/08/25 15:47:35 - mmengine - INFO - Epoch(val) [18][200/472] eta: 0:00:37 time: 0.1383 data_time: 0.0918 memory: 353 2022/08/25 15:47:42 - mmengine - INFO - Epoch(val) [18][300/472] eta: 0:00:07 time: 0.0438 data_time: 0.0206 memory: 388 2022/08/25 15:47:54 - mmengine - INFO - Epoch(val) [18][400/472] eta: 0:00:15 time: 0.2084 data_time: 0.1850 memory: 353 2022/08/25 15:48:03 - mmengine - INFO - Epoch(val) [18][472/472] kie/macro_f1: 0.8672 2022/08/25 15:48:32 - mmengine - INFO - Epoch(train) [19][100/317] lr: 1.0000e-03 eta: 0:50:51 time: 0.1732 data_time: 0.0053 memory: 6839 loss_node: 0.2269 loss_edge: 0.0000 acc_node: 92.1569 acc_edge: 100.0000 loss: 0.2269 2022/08/25 15:49:01 - mmengine - INFO - Epoch(train) [19][200/317] lr: 1.0000e-03 eta: 0:50:42 time: 0.2031 data_time: 0.0056 memory: 6403 loss_node: 0.1557 loss_edge: 0.0000 acc_node: 94.3396 acc_edge: 100.0000 loss: 0.1558 2022/08/25 15:49:23 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:49:24 - mmengine - INFO - Epoch(train) [19][300/317] lr: 1.0000e-03 eta: 0:50:18 time: 0.1600 data_time: 0.0056 memory: 8040 loss_node: 0.2245 loss_edge: 0.0000 acc_node: 90.6475 acc_edge: 100.0000 loss: 0.2245 2022/08/25 15:49:30 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:49:30 - mmengine - INFO - Saving checkpoint at 19 epochs 2022/08/25 15:49:37 - mmengine - INFO - Epoch(val) [19][100/472] eta: 0:00:18 time: 0.0494 data_time: 0.0266 memory: 7884 2022/08/25 15:49:41 - mmengine - INFO - Epoch(val) [19][200/472] eta: 0:00:10 time: 0.0393 data_time: 0.0147 memory: 353 2022/08/25 15:49:45 - mmengine - INFO - Epoch(val) [19][300/472] eta: 0:00:17 time: 0.1017 data_time: 0.0635 memory: 388 2022/08/25 15:49:50 - mmengine - INFO - Epoch(val) [19][400/472] eta: 0:00:01 time: 0.0259 data_time: 0.0025 memory: 353 2022/08/25 15:49:53 - mmengine - INFO - Epoch(val) [19][472/472] kie/macro_f1: 0.8463 2022/08/25 15:50:16 - mmengine - INFO - Epoch(train) [20][100/317] lr: 1.0000e-03 eta: 0:49:49 time: 0.2955 data_time: 0.0127 memory: 19864 loss_node: 0.1811 loss_edge: 0.0000 acc_node: 93.6306 acc_edge: 100.0000 loss: 0.1811 2022/08/25 15:50:33 - mmengine - INFO - Epoch(train) [20][200/317] lr: 1.0000e-03 eta: 0:49:14 time: 0.1504 data_time: 0.0046 memory: 7057 loss_node: 0.1968 loss_edge: 0.0000 acc_node: 90.2913 acc_edge: 100.0000 loss: 0.1968 2022/08/25 15:50:53 - mmengine - INFO - Epoch(train) [20][300/317] lr: 1.0000e-03 eta: 0:48:45 time: 0.1681 data_time: 0.0046 memory: 6403 loss_node: 0.1954 loss_edge: 0.0000 acc_node: 96.3768 acc_edge: 100.0000 loss: 0.1954 2022/08/25 15:50:58 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:50:58 - mmengine - INFO - Saving checkpoint at 20 epochs 2022/08/25 15:51:04 - mmengine - INFO - Epoch(val) [20][100/472] eta: 0:00:10 time: 0.0294 data_time: 0.0082 memory: 5530 2022/08/25 15:51:08 - mmengine - INFO - Epoch(val) [20][200/472] eta: 0:00:11 time: 0.0439 data_time: 0.0213 memory: 353 2022/08/25 15:51:12 - mmengine - INFO - Epoch(val) [20][300/472] eta: 0:00:07 time: 0.0432 data_time: 0.0202 memory: 388 2022/08/25 15:51:19 - mmengine - INFO - Epoch(val) [20][400/472] eta: 0:00:02 time: 0.0288 data_time: 0.0048 memory: 353 2022/08/25 15:51:22 - mmengine - INFO - Epoch(val) [20][472/472] kie/macro_f1: 0.8637 2022/08/25 15:51:43 - mmengine - INFO - Epoch(train) [21][100/317] lr: 1.0000e-03 eta: 0:48:13 time: 0.1656 data_time: 0.0051 memory: 6839 loss_node: 0.1945 loss_edge: 0.0000 acc_node: 94.9275 acc_edge: 100.0000 loss: 0.1945 2022/08/25 15:52:04 - mmengine - INFO - Epoch(train) [21][200/317] lr: 1.0000e-03 eta: 0:47:48 time: 0.1793 data_time: 0.0057 memory: 19976 loss_node: 0.2120 loss_edge: 0.0001 acc_node: 96.3504 acc_edge: 100.0000 loss: 0.2121 2022/08/25 15:52:22 - mmengine - INFO - Epoch(train) [21][300/317] lr: 1.0000e-03 eta: 0:47:15 time: 0.1721 data_time: 0.0047 memory: 7057 loss_node: 0.1891 loss_edge: 0.0000 acc_node: 97.1223 acc_edge: 100.0000 loss: 0.1891 2022/08/25 15:52:27 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:52:27 - mmengine - INFO - Saving checkpoint at 21 epochs 2022/08/25 15:52:32 - mmengine - INFO - Epoch(val) [21][100/472] eta: 0:00:12 time: 0.0334 data_time: 0.0120 memory: 8837 2022/08/25 15:52:36 - mmengine - INFO - Epoch(val) [21][200/472] eta: 0:00:14 time: 0.0526 data_time: 0.0368 memory: 353 2022/08/25 15:52:39 - mmengine - INFO - Epoch(val) [21][300/472] eta: 0:00:07 time: 0.0452 data_time: 0.0216 memory: 388 2022/08/25 15:52:44 - mmengine - INFO - Epoch(val) [21][400/472] eta: 0:00:01 time: 0.0277 data_time: 0.0060 memory: 353 2022/08/25 15:52:48 - mmengine - INFO - Epoch(val) [21][472/472] kie/macro_f1: 0.8774 2022/08/25 15:53:08 - mmengine - INFO - Epoch(train) [22][100/317] lr: 1.0000e-03 eta: 0:46:41 time: 0.2697 data_time: 0.0097 memory: 7057 loss_node: 0.1509 loss_edge: 0.0000 acc_node: 95.6790 acc_edge: 100.0000 loss: 0.1509 2022/08/25 15:53:26 - mmengine - INFO - Epoch(train) [22][200/317] lr: 1.0000e-03 eta: 0:46:11 time: 0.2362 data_time: 0.0064 memory: 7458 loss_node: 0.1197 loss_edge: 0.0000 acc_node: 93.8775 acc_edge: 100.0000 loss: 0.1197 2022/08/25 15:53:43 - mmengine - INFO - Epoch(train) [22][300/317] lr: 1.0000e-03 eta: 0:45:38 time: 0.1599 data_time: 0.0050 memory: 7057 loss_node: 0.2345 loss_edge: 0.0000 acc_node: 90.3061 acc_edge: 100.0000 loss: 0.2346 2022/08/25 15:53:48 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:53:48 - mmengine - INFO - Saving checkpoint at 22 epochs 2022/08/25 15:53:54 - mmengine - INFO - Epoch(val) [22][100/472] eta: 0:00:18 time: 0.0487 data_time: 0.0266 memory: 9162 2022/08/25 15:53:57 - mmengine - INFO - Epoch(val) [22][200/472] eta: 0:00:10 time: 0.0397 data_time: 0.0241 memory: 353 2022/08/25 15:54:01 - mmengine - INFO - Epoch(val) [22][300/472] eta: 0:00:07 time: 0.0417 data_time: 0.0196 memory: 388 2022/08/25 15:54:04 - mmengine - INFO - Epoch(val) [22][400/472] eta: 0:00:02 time: 0.0309 data_time: 0.0060 memory: 353 2022/08/25 15:54:08 - mmengine - INFO - Epoch(val) [22][472/472] kie/macro_f1: 0.8630 2022/08/25 15:54:13 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:54:27 - mmengine - INFO - Epoch(train) [23][100/317] lr: 1.0000e-03 eta: 0:45:01 time: 0.2791 data_time: 0.0053 memory: 6403 loss_node: 0.1690 loss_edge: 0.0000 acc_node: 93.4959 acc_edge: 100.0000 loss: 0.1691 2022/08/25 15:54:51 - mmengine - INFO - Epoch(train) [23][200/317] lr: 1.0000e-03 eta: 0:44:42 time: 0.2730 data_time: 0.0057 memory: 11029 loss_node: 0.1884 loss_edge: 0.0000 acc_node: 95.0413 acc_edge: 100.0000 loss: 0.1884 2022/08/25 15:55:19 - mmengine - INFO - Epoch(train) [23][300/317] lr: 1.0000e-03 eta: 0:44:27 time: 0.1641 data_time: 0.0044 memory: 5387 loss_node: 0.2700 loss_edge: 0.0000 acc_node: 94.7712 acc_edge: 100.0000 loss: 0.2700 2022/08/25 15:55:27 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:55:27 - mmengine - INFO - Saving checkpoint at 23 epochs 2022/08/25 15:55:49 - mmengine - INFO - Epoch(val) [23][100/472] eta: 0:00:15 time: 0.0409 data_time: 0.0179 memory: 6999 2022/08/25 15:56:01 - mmengine - INFO - Epoch(val) [23][200/472] eta: 0:01:16 time: 0.2800 data_time: 0.2555 memory: 353 2022/08/25 15:56:19 - mmengine - INFO - Epoch(val) [23][300/472] eta: 0:00:11 time: 0.0654 data_time: 0.0419 memory: 388 2022/08/25 15:56:32 - mmengine - INFO - Epoch(val) [23][400/472] eta: 0:00:02 time: 0.0339 data_time: 0.0054 memory: 353 2022/08/25 15:56:40 - mmengine - INFO - Epoch(val) [23][472/472] kie/macro_f1: 0.8559 2022/08/25 15:57:07 - mmengine - INFO - Epoch(train) [24][100/317] lr: 1.0000e-03 eta: 0:44:07 time: 0.2561 data_time: 0.0416 memory: 7108 loss_node: 0.1206 loss_edge: 0.0000 acc_node: 96.6292 acc_edge: 100.0000 loss: 0.1206 2022/08/25 15:57:32 - mmengine - INFO - Epoch(train) [24][200/317] lr: 1.0000e-03 eta: 0:43:48 time: 0.2536 data_time: 0.0063 memory: 6402 loss_node: 0.1347 loss_edge: 0.0000 acc_node: 95.2128 acc_edge: 100.0000 loss: 0.1347 2022/08/25 15:57:53 - mmengine - INFO - Epoch(train) [24][300/317] lr: 1.0000e-03 eta: 0:43:21 time: 0.1624 data_time: 0.0054 memory: 7057 loss_node: 0.1517 loss_edge: 0.0000 acc_node: 97.4576 acc_edge: 100.0000 loss: 0.1517 2022/08/25 15:57:56 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:57:56 - mmengine - INFO - Saving checkpoint at 24 epochs 2022/08/25 15:58:05 - mmengine - INFO - Epoch(val) [24][100/472] eta: 0:00:46 time: 0.1257 data_time: 0.1037 memory: 6402 2022/08/25 15:58:12 - mmengine - INFO - Epoch(val) [24][200/472] eta: 0:00:15 time: 0.0585 data_time: 0.0337 memory: 353 2022/08/25 15:58:17 - mmengine - INFO - Epoch(val) [24][300/472] eta: 0:00:07 time: 0.0428 data_time: 0.0280 memory: 388 2022/08/25 15:58:20 - mmengine - INFO - Epoch(val) [24][400/472] eta: 0:00:01 time: 0.0231 data_time: 0.0016 memory: 353 2022/08/25 15:58:24 - mmengine - INFO - Epoch(val) [24][472/472] kie/macro_f1: 0.8608 2022/08/25 15:58:45 - mmengine - INFO - Epoch(train) [25][100/317] lr: 1.0000e-03 eta: 0:42:49 time: 0.1640 data_time: 0.0047 memory: 7057 loss_node: 0.1762 loss_edge: 0.0000 acc_node: 95.6835 acc_edge: 100.0000 loss: 0.1762 2022/08/25 15:59:02 - mmengine - INFO - Epoch(train) [25][200/317] lr: 1.0000e-03 eta: 0:42:18 time: 0.1556 data_time: 0.0049 memory: 5530 loss_node: 0.1932 loss_edge: 0.0000 acc_node: 90.8163 acc_edge: 100.0000 loss: 0.1932 2022/08/25 15:59:22 - mmengine - INFO - Epoch(train) [25][300/317] lr: 1.0000e-03 eta: 0:41:53 time: 0.1551 data_time: 0.0051 memory: 5530 loss_node: 0.2044 loss_edge: 0.0000 acc_node: 93.8650 acc_edge: 100.0000 loss: 0.2044 2022/08/25 15:59:27 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 15:59:27 - mmengine - INFO - Saving checkpoint at 25 epochs 2022/08/25 15:59:33 - mmengine - INFO - Epoch(val) [25][100/472] eta: 0:00:10 time: 0.0275 data_time: 0.0058 memory: 18373 2022/08/25 15:59:37 - mmengine - INFO - Epoch(val) [25][200/472] eta: 0:00:09 time: 0.0337 data_time: 0.0176 memory: 353 2022/08/25 15:59:40 - mmengine - INFO - Epoch(val) [25][300/472] eta: 0:00:05 time: 0.0345 data_time: 0.0123 memory: 388 2022/08/25 15:59:43 - mmengine - INFO - Epoch(val) [25][400/472] eta: 0:00:01 time: 0.0261 data_time: 0.0018 memory: 353 2022/08/25 15:59:46 - mmengine - INFO - Epoch(val) [25][472/472] kie/macro_f1: 0.8695 2022/08/25 16:00:00 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:00:04 - mmengine - INFO - Epoch(train) [26][100/317] lr: 1.0000e-03 eta: 0:41:17 time: 0.1667 data_time: 0.0106 memory: 7057 loss_node: 0.1410 loss_edge: 0.0000 acc_node: 98.5075 acc_edge: 100.0000 loss: 0.1410 2022/08/25 16:00:23 - mmengine - INFO - Epoch(train) [26][200/317] lr: 1.0000e-03 eta: 0:40:51 time: 0.1943 data_time: 0.0085 memory: 7245 loss_node: 0.1766 loss_edge: 0.0000 acc_node: 96.4497 acc_edge: 100.0000 loss: 0.1766 2022/08/25 16:00:41 - mmengine - INFO - Epoch(train) [26][300/317] lr: 1.0000e-03 eta: 0:40:22 time: 0.1541 data_time: 0.0044 memory: 7057 loss_node: 0.1649 loss_edge: 0.0000 acc_node: 96.3855 acc_edge: 100.0000 loss: 0.1649 2022/08/25 16:00:44 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:00:44 - mmengine - INFO - Saving checkpoint at 26 epochs 2022/08/25 16:00:50 - mmengine - INFO - Epoch(val) [26][100/472] eta: 0:00:10 time: 0.0279 data_time: 0.0081 memory: 6621 2022/08/25 16:00:53 - mmengine - INFO - Epoch(val) [26][200/472] eta: 0:00:10 time: 0.0382 data_time: 0.0164 memory: 353 2022/08/25 16:00:58 - mmengine - INFO - Epoch(val) [26][300/472] eta: 0:00:07 time: 0.0419 data_time: 0.0179 memory: 388 2022/08/25 16:01:01 - mmengine - INFO - Epoch(val) [26][400/472] eta: 0:00:02 time: 0.0280 data_time: 0.0031 memory: 353 2022/08/25 16:01:04 - mmengine - INFO - Epoch(val) [26][472/472] kie/macro_f1: 0.8730 2022/08/25 16:01:23 - mmengine - INFO - Epoch(train) [27][100/317] lr: 1.0000e-03 eta: 0:39:49 time: 0.1619 data_time: 0.0039 memory: 7057 loss_node: 0.1320 loss_edge: 0.0000 acc_node: 91.6084 acc_edge: 100.0000 loss: 0.1320 2022/08/25 16:01:43 - mmengine - INFO - Epoch(train) [27][200/317] lr: 1.0000e-03 eta: 0:39:23 time: 0.1510 data_time: 0.0048 memory: 7780 loss_node: 0.1373 loss_edge: 0.0000 acc_node: 94.4056 acc_edge: 100.0000 loss: 0.1373 2022/08/25 16:02:03 - mmengine - INFO - Epoch(train) [27][300/317] lr: 1.0000e-03 eta: 0:38:58 time: 0.1538 data_time: 0.0048 memory: 6621 loss_node: 0.2703 loss_edge: 0.0000 acc_node: 93.9189 acc_edge: 100.0000 loss: 0.2703 2022/08/25 16:02:07 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:02:07 - mmengine - INFO - Saving checkpoint at 27 epochs 2022/08/25 16:02:13 - mmengine - INFO - Epoch(val) [27][100/472] eta: 0:00:10 time: 0.0285 data_time: 0.0077 memory: 5530 2022/08/25 16:02:16 - mmengine - INFO - Epoch(val) [27][200/472] eta: 0:00:10 time: 0.0373 data_time: 0.0146 memory: 353 2022/08/25 16:02:20 - mmengine - INFO - Epoch(val) [27][300/472] eta: 0:00:13 time: 0.0758 data_time: 0.0226 memory: 388 2022/08/25 16:02:24 - mmengine - INFO - Epoch(val) [27][400/472] eta: 0:00:01 time: 0.0243 data_time: 0.0014 memory: 353 2022/08/25 16:02:28 - mmengine - INFO - Epoch(val) [27][472/472] kie/macro_f1: 0.8358 2022/08/25 16:02:48 - mmengine - INFO - Epoch(train) [28][100/317] lr: 1.0000e-03 eta: 0:38:28 time: 0.3408 data_time: 0.0134 memory: 7566 loss_node: 0.1778 loss_edge: 0.0000 acc_node: 94.3548 acc_edge: 100.0000 loss: 0.1779 2022/08/25 16:03:17 - mmengine - INFO - Epoch(train) [28][200/317] lr: 1.0000e-03 eta: 0:38:13 time: 0.1668 data_time: 0.0127 memory: 7277 loss_node: 0.1258 loss_edge: 0.0000 acc_node: 92.8144 acc_edge: 100.0000 loss: 0.1258 2022/08/25 16:03:39 - mmengine - INFO - Epoch(train) [28][300/317] lr: 1.0000e-03 eta: 0:37:51 time: 0.1555 data_time: 0.0054 memory: 7057 loss_node: 0.1422 loss_edge: 0.0000 acc_node: 92.8571 acc_edge: 100.0000 loss: 0.1422 2022/08/25 16:03:45 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:03:45 - mmengine - INFO - Saving checkpoint at 28 epochs 2022/08/25 16:03:56 - mmengine - INFO - Epoch(val) [28][100/472] eta: 0:00:18 time: 0.0499 data_time: 0.0252 memory: 5372 2022/08/25 16:04:03 - mmengine - INFO - Epoch(val) [28][200/472] eta: 0:00:10 time: 0.0398 data_time: 0.0158 memory: 353 2022/08/25 16:04:20 - mmengine - INFO - Epoch(val) [28][300/472] eta: 0:00:15 time: 0.0909 data_time: 0.0285 memory: 388 2022/08/25 16:04:29 - mmengine - INFO - Epoch(val) [28][400/472] eta: 0:00:01 time: 0.0271 data_time: 0.0037 memory: 353 2022/08/25 16:04:39 - mmengine - INFO - Epoch(val) [28][472/472] kie/macro_f1: 0.8764 2022/08/25 16:05:01 - mmengine - INFO - Epoch(train) [29][100/317] lr: 1.0000e-03 eta: 0:37:26 time: 0.2306 data_time: 0.0288 memory: 7057 loss_node: 0.1400 loss_edge: 0.0000 acc_node: 88.0435 acc_edge: 100.0000 loss: 0.1401 2022/08/25 16:05:06 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:05:27 - mmengine - INFO - Epoch(train) [29][200/317] lr: 1.0000e-03 eta: 0:37:07 time: 0.3430 data_time: 0.0041 memory: 10556 loss_node: 0.1271 loss_edge: 0.0000 acc_node: 98.3193 acc_edge: 100.0000 loss: 0.1271 2022/08/25 16:05:49 - mmengine - INFO - Epoch(train) [29][300/317] lr: 1.0000e-03 eta: 0:36:44 time: 0.1655 data_time: 0.0052 memory: 6404 loss_node: 0.1726 loss_edge: 0.0000 acc_node: 92.4658 acc_edge: 100.0000 loss: 0.1726 2022/08/25 16:05:52 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:05:52 - mmengine - INFO - Saving checkpoint at 29 epochs 2022/08/25 16:06:01 - mmengine - INFO - Epoch(val) [29][100/472] eta: 0:00:10 time: 0.0283 data_time: 0.0066 memory: 4875 2022/08/25 16:06:05 - mmengine - INFO - Epoch(val) [29][200/472] eta: 0:00:12 time: 0.0470 data_time: 0.0218 memory: 354 2022/08/25 16:06:11 - mmengine - INFO - Epoch(val) [29][300/472] eta: 0:00:06 time: 0.0391 data_time: 0.0183 memory: 389 2022/08/25 16:06:14 - mmengine - INFO - Epoch(val) [29][400/472] eta: 0:00:02 time: 0.0286 data_time: 0.0070 memory: 354 2022/08/25 16:06:16 - mmengine - INFO - Epoch(val) [29][472/472] kie/macro_f1: 0.8628 2022/08/25 16:06:33 - mmengine - INFO - Epoch(train) [30][100/317] lr: 1.0000e-03 eta: 0:36:10 time: 0.1554 data_time: 0.0046 memory: 6577 loss_node: 0.1222 loss_edge: 0.0000 acc_node: 96.5035 acc_edge: 100.0000 loss: 0.1222 2022/08/25 16:06:55 - mmengine - INFO - Epoch(train) [30][200/317] lr: 1.0000e-03 eta: 0:35:48 time: 0.1620 data_time: 0.0051 memory: 7057 loss_node: 0.1402 loss_edge: 0.0000 acc_node: 99.2593 acc_edge: 100.0000 loss: 0.1402 2022/08/25 16:07:16 - mmengine - INFO - Epoch(train) [30][300/317] lr: 1.0000e-03 eta: 0:35:24 time: 0.1694 data_time: 0.0055 memory: 6185 loss_node: 0.1472 loss_edge: 0.0000 acc_node: 95.8904 acc_edge: 100.0000 loss: 0.1473 2022/08/25 16:07:20 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:07:20 - mmengine - INFO - Saving checkpoint at 30 epochs 2022/08/25 16:07:25 - mmengine - INFO - Epoch(val) [30][100/472] eta: 0:00:10 time: 0.0287 data_time: 0.0077 memory: 6840 2022/08/25 16:07:29 - mmengine - INFO - Epoch(val) [30][200/472] eta: 0:00:09 time: 0.0353 data_time: 0.0112 memory: 353 2022/08/25 16:07:32 - mmengine - INFO - Epoch(val) [30][300/472] eta: 0:00:06 time: 0.0374 data_time: 0.0154 memory: 388 2022/08/25 16:07:35 - mmengine - INFO - Epoch(val) [30][400/472] eta: 0:00:01 time: 0.0248 data_time: 0.0029 memory: 353 2022/08/25 16:07:40 - mmengine - INFO - Epoch(val) [30][472/472] kie/macro_f1: 0.8717 2022/08/25 16:08:01 - mmengine - INFO - Epoch(train) [31][100/317] lr: 1.0000e-03 eta: 0:34:54 time: 0.4350 data_time: 0.0266 memory: 7058 loss_node: 0.1233 loss_edge: 0.0000 acc_node: 95.6897 acc_edge: 100.0000 loss: 0.1233 2022/08/25 16:08:21 - mmengine - INFO - Epoch(train) [31][200/317] lr: 1.0000e-03 eta: 0:34:29 time: 0.1540 data_time: 0.0048 memory: 6566 loss_node: 0.1370 loss_edge: 0.0000 acc_node: 95.0000 acc_edge: 100.0000 loss: 0.1370 2022/08/25 16:08:40 - mmengine - INFO - Epoch(train) [31][300/317] lr: 1.0000e-03 eta: 0:34:05 time: 0.4969 data_time: 0.0276 memory: 7058 loss_node: 0.1513 loss_edge: 0.0000 acc_node: 95.0980 acc_edge: 100.0000 loss: 0.1513 2022/08/25 16:08:43 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:08:43 - mmengine - INFO - Saving checkpoint at 31 epochs 2022/08/25 16:08:48 - mmengine - INFO - Epoch(val) [31][100/472] eta: 0:00:09 time: 0.0255 data_time: 0.0028 memory: 4875 2022/08/25 16:08:51 - mmengine - INFO - Epoch(val) [31][200/472] eta: 0:00:09 time: 0.0344 data_time: 0.0109 memory: 354 2022/08/25 16:08:54 - mmengine - INFO - Epoch(val) [31][300/472] eta: 0:00:05 time: 0.0349 data_time: 0.0128 memory: 388 2022/08/25 16:08:59 - mmengine - INFO - Epoch(val) [31][400/472] eta: 0:00:01 time: 0.0179 data_time: 0.0043 memory: 354 2022/08/25 16:09:01 - mmengine - INFO - Epoch(val) [31][472/472] kie/macro_f1: 0.8783 2022/08/25 16:09:21 - mmengine - INFO - Epoch(train) [32][100/317] lr: 1.0000e-03 eta: 0:33:34 time: 0.4320 data_time: 0.0146 memory: 6404 loss_node: 0.1338 loss_edge: 0.0000 acc_node: 94.4828 acc_edge: 100.0000 loss: 0.1339 2022/08/25 16:09:33 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:09:41 - mmengine - INFO - Epoch(train) [32][200/317] lr: 1.0000e-03 eta: 0:33:10 time: 0.5002 data_time: 0.0266 memory: 7058 loss_node: 0.1158 loss_edge: 0.0000 acc_node: 92.4370 acc_edge: 100.0000 loss: 0.1159 2022/08/25 16:10:03 - mmengine - INFO - Epoch(train) [32][300/317] lr: 1.0000e-03 eta: 0:32:48 time: 0.6617 data_time: 0.0339 memory: 6404 loss_node: 0.1175 loss_edge: 0.0000 acc_node: 96.4981 acc_edge: 100.0000 loss: 0.1175 2022/08/25 16:10:06 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:10:06 - mmengine - INFO - Saving checkpoint at 32 epochs 2022/08/25 16:10:11 - mmengine - INFO - Epoch(val) [32][100/472] eta: 0:00:10 time: 0.0283 data_time: 0.0059 memory: 5531 2022/08/25 16:10:15 - mmengine - INFO - Epoch(val) [32][200/472] eta: 0:00:13 time: 0.0507 data_time: 0.0282 memory: 354 2022/08/25 16:10:21 - mmengine - INFO - Epoch(val) [32][300/472] eta: 0:00:06 time: 0.0353 data_time: 0.0174 memory: 388 2022/08/25 16:10:24 - mmengine - INFO - Epoch(val) [32][400/472] eta: 0:00:01 time: 0.0249 data_time: 0.0034 memory: 354 2022/08/25 16:10:27 - mmengine - INFO - Epoch(val) [32][472/472] kie/macro_f1: 0.8825 2022/08/25 16:10:45 - mmengine - INFO - Epoch(train) [33][100/317] lr: 1.0000e-03 eta: 0:32:16 time: 0.1567 data_time: 0.0057 memory: 7058 loss_node: 0.1137 loss_edge: 0.0000 acc_node: 95.7895 acc_edge: 100.0000 loss: 0.1137 2022/08/25 16:11:09 - mmengine - INFO - Epoch(train) [33][200/317] lr: 1.0000e-03 eta: 0:31:56 time: 0.1613 data_time: 0.0089 memory: 7058 loss_node: 0.1434 loss_edge: 0.0000 acc_node: 96.4029 acc_edge: 100.0000 loss: 0.1434 2022/08/25 16:11:27 - mmengine - INFO - Epoch(train) [33][300/317] lr: 1.0000e-03 eta: 0:31:31 time: 0.1641 data_time: 0.0053 memory: 7058 loss_node: 0.1409 loss_edge: 0.0000 acc_node: 96.2687 acc_edge: 100.0000 loss: 0.1409 2022/08/25 16:11:33 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:11:33 - mmengine - INFO - Saving checkpoint at 33 epochs 2022/08/25 16:11:39 - mmengine - INFO - Epoch(val) [33][100/472] eta: 0:00:09 time: 0.0252 data_time: 0.0038 memory: 5942 2022/08/25 16:11:42 - mmengine - INFO - Epoch(val) [33][200/472] eta: 0:00:13 time: 0.0510 data_time: 0.0274 memory: 353 2022/08/25 16:11:46 - mmengine - INFO - Epoch(val) [33][300/472] eta: 0:00:06 time: 0.0387 data_time: 0.0156 memory: 388 2022/08/25 16:11:51 - mmengine - INFO - Epoch(val) [33][400/472] eta: 0:00:01 time: 0.0257 data_time: 0.0039 memory: 353 2022/08/25 16:11:55 - mmengine - INFO - Epoch(val) [33][472/472] kie/macro_f1: 0.8716 2022/08/25 16:12:17 - mmengine - INFO - Epoch(train) [34][100/317] lr: 1.0000e-03 eta: 0:31:05 time: 0.1922 data_time: 0.0220 memory: 6403 loss_node: 0.1521 loss_edge: 0.0000 acc_node: 96.4286 acc_edge: 100.0000 loss: 0.1521 2022/08/25 16:12:46 - mmengine - INFO - Epoch(train) [34][200/317] lr: 1.0000e-03 eta: 0:30:49 time: 0.1512 data_time: 0.0049 memory: 6587 loss_node: 0.1622 loss_edge: 0.0000 acc_node: 94.9275 acc_edge: 100.0000 loss: 0.1622 2022/08/25 16:13:26 - mmengine - INFO - Epoch(train) [34][300/317] lr: 1.0000e-03 eta: 0:30:40 time: 1.3002 data_time: 0.3826 memory: 6402 loss_node: 0.1193 loss_edge: 0.0000 acc_node: 99.1031 acc_edge: 100.0000 loss: 0.1193 2022/08/25 16:13:32 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:13:33 - mmengine - INFO - Saving checkpoint at 34 epochs 2022/08/25 16:13:43 - mmengine - INFO - Epoch(val) [34][100/472] eta: 0:00:25 time: 0.0686 data_time: 0.0453 memory: 7057 2022/08/25 16:13:53 - mmengine - INFO - Epoch(val) [34][200/472] eta: 0:00:14 time: 0.0533 data_time: 0.0281 memory: 353 2022/08/25 16:13:59 - mmengine - INFO - Epoch(val) [34][300/472] eta: 0:00:06 time: 0.0371 data_time: 0.0173 memory: 388 2022/08/25 16:14:10 - mmengine - INFO - Epoch(val) [34][400/472] eta: 0:00:08 time: 0.1188 data_time: 0.0952 memory: 353 2022/08/25 16:14:17 - mmengine - INFO - Epoch(val) [34][472/472] kie/macro_f1: 0.8646 2022/08/25 16:14:48 - mmengine - INFO - Epoch(train) [35][100/317] lr: 1.0000e-03 eta: 0:30:20 time: 0.1737 data_time: 0.0114 memory: 7058 loss_node: 0.0998 loss_edge: 0.0000 acc_node: 98.0469 acc_edge: 100.0000 loss: 0.0998 2022/08/25 16:15:11 - mmengine - INFO - Epoch(train) [35][200/317] lr: 1.0000e-03 eta: 0:29:58 time: 0.1502 data_time: 0.0048 memory: 5377 loss_node: 0.1318 loss_edge: 0.0000 acc_node: 94.9045 acc_edge: 100.0000 loss: 0.1318 2022/08/25 16:15:17 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:15:31 - mmengine - INFO - Epoch(train) [35][300/317] lr: 1.0000e-03 eta: 0:29:35 time: 0.1663 data_time: 0.0045 memory: 7058 loss_node: 0.1414 loss_edge: 0.0000 acc_node: 97.0414 acc_edge: 100.0000 loss: 0.1414 2022/08/25 16:15:34 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:15:34 - mmengine - INFO - Saving checkpoint at 35 epochs 2022/08/25 16:15:42 - mmengine - INFO - Epoch(val) [35][100/472] eta: 0:00:12 time: 0.0334 data_time: 0.0107 memory: 4073 2022/08/25 16:15:47 - mmengine - INFO - Epoch(val) [35][200/472] eta: 0:00:27 time: 0.1021 data_time: 0.0735 memory: 353 2022/08/25 16:15:52 - mmengine - INFO - Epoch(val) [35][300/472] eta: 0:00:08 time: 0.0469 data_time: 0.0228 memory: 388 2022/08/25 16:15:56 - mmengine - INFO - Epoch(val) [35][400/472] eta: 0:00:02 time: 0.0301 data_time: 0.0091 memory: 353 2022/08/25 16:16:01 - mmengine - INFO - Epoch(val) [35][472/472] kie/macro_f1: 0.8793 2022/08/25 16:16:21 - mmengine - INFO - Epoch(train) [36][100/317] lr: 1.0000e-03 eta: 0:29:06 time: 0.1810 data_time: 0.0174 memory: 6402 loss_node: 0.0872 loss_edge: 0.0000 acc_node: 93.4343 acc_edge: 100.0000 loss: 0.0872 2022/08/25 16:16:42 - mmengine - INFO - Epoch(train) [36][200/317] lr: 1.0000e-03 eta: 0:28:42 time: 0.1556 data_time: 0.0056 memory: 6647 loss_node: 0.1171 loss_edge: 0.0000 acc_node: 96.5812 acc_edge: 100.0000 loss: 0.1172 2022/08/25 16:17:01 - mmengine - INFO - Epoch(train) [36][300/317] lr: 1.0000e-03 eta: 0:28:18 time: 0.2164 data_time: 0.0088 memory: 6622 loss_node: 0.1689 loss_edge: 0.0000 acc_node: 96.5217 acc_edge: 100.0000 loss: 0.1689 2022/08/25 16:17:04 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:17:04 - mmengine - INFO - Saving checkpoint at 36 epochs 2022/08/25 16:17:10 - mmengine - INFO - Epoch(val) [36][100/472] eta: 0:00:14 time: 0.0393 data_time: 0.0182 memory: 7058 2022/08/25 16:17:13 - mmengine - INFO - Epoch(val) [36][200/472] eta: 0:00:12 time: 0.0469 data_time: 0.0249 memory: 353 2022/08/25 16:17:17 - mmengine - INFO - Epoch(val) [36][300/472] eta: 0:00:07 time: 0.0435 data_time: 0.0215 memory: 388 2022/08/25 16:17:22 - mmengine - INFO - Epoch(val) [36][400/472] eta: 0:00:01 time: 0.0211 data_time: 0.0054 memory: 353 2022/08/25 16:17:24 - mmengine - INFO - Epoch(val) [36][472/472] kie/macro_f1: 0.8858 2022/08/25 16:17:45 - mmengine - INFO - Epoch(train) [37][100/317] lr: 1.0000e-03 eta: 0:27:49 time: 0.1704 data_time: 0.0047 memory: 7371 loss_node: 0.1048 loss_edge: 0.0000 acc_node: 96.5278 acc_edge: 100.0000 loss: 0.1048 2022/08/25 16:18:07 - mmengine - INFO - Epoch(train) [37][200/317] lr: 1.0000e-03 eta: 0:27:27 time: 0.1726 data_time: 0.0048 memory: 7057 loss_node: 0.1326 loss_edge: 0.0000 acc_node: 96.7213 acc_edge: 100.0000 loss: 0.1327 2022/08/25 16:18:27 - mmengine - INFO - Epoch(train) [37][300/317] lr: 1.0000e-03 eta: 0:27:03 time: 0.1765 data_time: 0.0052 memory: 7539 loss_node: 0.1474 loss_edge: 0.0000 acc_node: 92.4812 acc_edge: 100.0000 loss: 0.1474 2022/08/25 16:18:30 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:18:30 - mmengine - INFO - Saving checkpoint at 37 epochs 2022/08/25 16:18:35 - mmengine - INFO - Epoch(val) [37][100/472] eta: 0:00:10 time: 0.0295 data_time: 0.0085 memory: 4213 2022/08/25 16:18:39 - mmengine - INFO - Epoch(val) [37][200/472] eta: 0:00:07 time: 0.0291 data_time: 0.0145 memory: 353 2022/08/25 16:18:42 - mmengine - INFO - Epoch(val) [37][300/472] eta: 0:00:06 time: 0.0350 data_time: 0.0136 memory: 388 2022/08/25 16:18:47 - mmengine - INFO - Epoch(val) [37][400/472] eta: 0:00:01 time: 0.0257 data_time: 0.0029 memory: 353 2022/08/25 16:18:50 - mmengine - INFO - Epoch(val) [37][472/472] kie/macro_f1: 0.8606 2022/08/25 16:19:08 - mmengine - INFO - Epoch(train) [38][100/317] lr: 1.0000e-03 eta: 0:26:34 time: 0.1665 data_time: 0.0081 memory: 6679 loss_node: 0.1329 loss_edge: 0.0000 acc_node: 95.2128 acc_edge: 100.0000 loss: 0.1329 2022/08/25 16:19:27 - mmengine - INFO - Epoch(train) [38][200/317] lr: 1.0000e-03 eta: 0:26:09 time: 0.1608 data_time: 0.0060 memory: 6402 loss_node: 0.1033 loss_edge: 0.0000 acc_node: 94.1176 acc_edge: 100.0000 loss: 0.1033 2022/08/25 16:19:42 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:19:46 - mmengine - INFO - Epoch(train) [38][300/317] lr: 1.0000e-03 eta: 0:25:46 time: 0.1609 data_time: 0.0059 memory: 7057 loss_node: 0.1486 loss_edge: 0.0000 acc_node: 97.2067 acc_edge: 100.0000 loss: 0.1486 2022/08/25 16:19:49 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:19:49 - mmengine - INFO - Saving checkpoint at 38 epochs 2022/08/25 16:19:55 - mmengine - INFO - Epoch(val) [38][100/472] eta: 0:00:10 time: 0.0277 data_time: 0.0067 memory: 7057 2022/08/25 16:19:59 - mmengine - INFO - Epoch(val) [38][200/472] eta: 0:00:12 time: 0.0446 data_time: 0.0198 memory: 353 2022/08/25 16:20:03 - mmengine - INFO - Epoch(val) [38][300/472] eta: 0:00:08 time: 0.0482 data_time: 0.0260 memory: 388 2022/08/25 16:20:06 - mmengine - INFO - Epoch(val) [38][400/472] eta: 0:00:02 time: 0.0291 data_time: 0.0051 memory: 353 2022/08/25 16:20:09 - mmengine - INFO - Epoch(val) [38][472/472] kie/macro_f1: 0.8745 2022/08/25 16:20:30 - mmengine - INFO - Epoch(train) [39][100/317] lr: 1.0000e-03 eta: 0:25:18 time: 0.3462 data_time: 0.0140 memory: 5612 loss_node: 0.1097 loss_edge: 0.0000 acc_node: 95.0704 acc_edge: 100.0000 loss: 0.1098 2022/08/25 16:20:55 - mmengine - INFO - Epoch(train) [39][200/317] lr: 1.0000e-03 eta: 0:24:58 time: 0.7596 data_time: 0.0187 memory: 7298 loss_node: 0.1180 loss_edge: 0.0000 acc_node: 92.9487 acc_edge: 100.0000 loss: 0.1180 2022/08/25 16:21:15 - mmengine - INFO - Epoch(train) [39][300/317] lr: 1.0000e-03 eta: 0:24:35 time: 0.1876 data_time: 0.0104 memory: 8310 loss_node: 0.1036 loss_edge: 0.0000 acc_node: 96.8354 acc_edge: 100.0000 loss: 0.1037 2022/08/25 16:21:21 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:21:21 - mmengine - INFO - Saving checkpoint at 39 epochs 2022/08/25 16:21:33 - mmengine - INFO - Epoch(val) [39][100/472] eta: 0:00:16 time: 0.0448 data_time: 0.0233 memory: 7622 2022/08/25 16:21:42 - mmengine - INFO - Epoch(val) [39][200/472] eta: 0:00:12 time: 0.0447 data_time: 0.0270 memory: 353 2022/08/25 16:21:48 - mmengine - INFO - Epoch(val) [39][300/472] eta: 0:00:15 time: 0.0898 data_time: 0.0648 memory: 388 2022/08/25 16:21:56 - mmengine - INFO - Epoch(val) [39][400/472] eta: 0:00:08 time: 0.1204 data_time: 0.0913 memory: 353 2022/08/25 16:22:08 - mmengine - INFO - Epoch(val) [39][472/472] kie/macro_f1: 0.8697 2022/08/25 16:22:32 - mmengine - INFO - Epoch(train) [40][100/317] lr: 1.0000e-03 eta: 0:24:09 time: 0.1515 data_time: 0.0055 memory: 7780 loss_node: 0.0896 loss_edge: 0.0000 acc_node: 95.8333 acc_edge: 100.0000 loss: 0.0896 2022/08/25 16:22:52 - mmengine - INFO - Epoch(train) [40][200/317] lr: 1.0000e-03 eta: 0:23:46 time: 0.1550 data_time: 0.0062 memory: 7057 loss_node: 0.1103 loss_edge: 0.0000 acc_node: 94.4882 acc_edge: 100.0000 loss: 0.1103 2022/08/25 16:23:17 - mmengine - INFO - Epoch(train) [40][300/317] lr: 1.0000e-03 eta: 0:23:25 time: 0.1717 data_time: 0.0230 memory: 6402 loss_node: 0.1391 loss_edge: 0.0000 acc_node: 95.3846 acc_edge: 100.0000 loss: 0.1392 2022/08/25 16:23:23 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:23:23 - mmengine - INFO - Saving checkpoint at 40 epochs 2022/08/25 16:23:31 - mmengine - INFO - Epoch(val) [40][100/472] eta: 0:00:20 time: 0.0547 data_time: 0.0316 memory: 4815 2022/08/25 16:23:36 - mmengine - INFO - Epoch(val) [40][200/472] eta: 0:00:10 time: 0.0404 data_time: 0.0150 memory: 353 2022/08/25 16:23:42 - mmengine - INFO - Epoch(val) [40][300/472] eta: 0:00:08 time: 0.0506 data_time: 0.0263 memory: 388 2022/08/25 16:23:46 - mmengine - INFO - Epoch(val) [40][400/472] eta: 0:00:02 time: 0.0297 data_time: 0.0056 memory: 353 2022/08/25 16:23:52 - mmengine - INFO - Epoch(val) [40][472/472] kie/macro_f1: 0.8788 2022/08/25 16:24:13 - mmengine - INFO - Epoch(train) [41][100/317] lr: 1.0000e-04 eta: 0:22:59 time: 0.1630 data_time: 0.0058 memory: 6402 loss_node: 0.0593 loss_edge: 0.0000 acc_node: 98.4000 acc_edge: 100.0000 loss: 0.0593 2022/08/25 16:24:31 - mmengine - INFO - Epoch(train) [41][200/317] lr: 1.0000e-04 eta: 0:22:35 time: 0.1479 data_time: 0.0083 memory: 7057 loss_node: 0.0719 loss_edge: 0.0000 acc_node: 97.4843 acc_edge: 100.0000 loss: 0.0719 2022/08/25 16:24:50 - mmengine - INFO - Epoch(train) [41][300/317] lr: 1.0000e-04 eta: 0:22:11 time: 0.3179 data_time: 0.0044 memory: 6402 loss_node: 0.0740 loss_edge: 0.0000 acc_node: 96.2963 acc_edge: 100.0000 loss: 0.0740 2022/08/25 16:24:54 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:24:54 - mmengine - INFO - Saving checkpoint at 41 epochs 2022/08/25 16:25:00 - mmengine - INFO - Epoch(val) [41][100/472] eta: 0:00:09 time: 0.0264 data_time: 0.0045 memory: 7268 2022/08/25 16:25:04 - mmengine - INFO - Epoch(val) [41][200/472] eta: 0:00:10 time: 0.0400 data_time: 0.0170 memory: 353 2022/08/25 16:25:07 - mmengine - INFO - Epoch(val) [41][300/472] eta: 0:00:06 time: 0.0367 data_time: 0.0155 memory: 388 2022/08/25 16:25:16 - mmengine - INFO - Epoch(val) [41][400/472] eta: 0:00:02 time: 0.0388 data_time: 0.0115 memory: 353 2022/08/25 16:25:19 - mmengine - INFO - Epoch(val) [41][472/472] kie/macro_f1: 0.8886 2022/08/25 16:25:21 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:25:40 - mmengine - INFO - Epoch(train) [42][100/317] lr: 1.0000e-04 eta: 0:21:44 time: 0.1594 data_time: 0.0061 memory: 7057 loss_node: 0.0555 loss_edge: 0.0000 acc_node: 99.2593 acc_edge: 100.0000 loss: 0.0556 2022/08/25 16:25:56 - mmengine - INFO - Epoch(train) [42][200/317] lr: 1.0000e-04 eta: 0:21:20 time: 0.1559 data_time: 0.0047 memory: 7057 loss_node: 0.0443 loss_edge: 0.0000 acc_node: 98.8950 acc_edge: 100.0000 loss: 0.0443 2022/08/25 16:26:17 - mmengine - INFO - Epoch(train) [42][300/317] lr: 1.0000e-04 eta: 0:20:57 time: 0.1556 data_time: 0.0061 memory: 6402 loss_node: 0.0419 loss_edge: 0.0000 acc_node: 98.5612 acc_edge: 100.0000 loss: 0.0419 2022/08/25 16:26:22 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:26:22 - mmengine - INFO - Saving checkpoint at 42 epochs 2022/08/25 16:26:27 - mmengine - INFO - Epoch(val) [42][100/472] eta: 0:00:09 time: 0.0267 data_time: 0.0056 memory: 7057 2022/08/25 16:26:30 - mmengine - INFO - Epoch(val) [42][200/472] eta: 0:00:08 time: 0.0321 data_time: 0.0184 memory: 353 2022/08/25 16:26:33 - mmengine - INFO - Epoch(val) [42][300/472] eta: 0:00:05 time: 0.0326 data_time: 0.0110 memory: 388 2022/08/25 16:26:36 - mmengine - INFO - Epoch(val) [42][400/472] eta: 0:00:01 time: 0.0269 data_time: 0.0036 memory: 353 2022/08/25 16:26:40 - mmengine - INFO - Epoch(val) [42][472/472] kie/macro_f1: 0.8905 2022/08/25 16:27:02 - mmengine - INFO - Epoch(train) [43][100/317] lr: 1.0000e-04 eta: 0:20:32 time: 0.1635 data_time: 0.0050 memory: 7057 loss_node: 0.0529 loss_edge: 0.0000 acc_node: 99.4186 acc_edge: 100.0000 loss: 0.0529 2022/08/25 16:27:24 - mmengine - INFO - Epoch(train) [43][200/317] lr: 1.0000e-04 eta: 0:20:10 time: 0.1647 data_time: 0.0048 memory: 7780 loss_node: 0.0357 loss_edge: 0.0000 acc_node: 99.3243 acc_edge: 100.0000 loss: 0.0357 2022/08/25 16:27:43 - mmengine - INFO - Epoch(train) [43][300/317] lr: 1.0000e-04 eta: 0:19:47 time: 0.1557 data_time: 0.0051 memory: 7092 loss_node: 0.0466 loss_edge: 0.0000 acc_node: 98.3871 acc_edge: 100.0000 loss: 0.0466 2022/08/25 16:27:46 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:27:46 - mmengine - INFO - Saving checkpoint at 43 epochs 2022/08/25 16:27:51 - mmengine - INFO - Epoch(val) [43][100/472] eta: 0:00:12 time: 0.0349 data_time: 0.0106 memory: 7058 2022/08/25 16:27:55 - mmengine - INFO - Epoch(val) [43][200/472] eta: 0:00:10 time: 0.0376 data_time: 0.0136 memory: 353 2022/08/25 16:27:59 - mmengine - INFO - Epoch(val) [43][300/472] eta: 0:00:27 time: 0.1600 data_time: 0.0251 memory: 389 2022/08/25 16:28:03 - mmengine - INFO - Epoch(val) [43][400/472] eta: 0:00:01 time: 0.0234 data_time: 0.0018 memory: 353 2022/08/25 16:28:06 - mmengine - INFO - Epoch(val) [43][472/472] kie/macro_f1: 0.8876 2022/08/25 16:28:27 - mmengine - INFO - Epoch(train) [44][100/317] lr: 1.0000e-04 eta: 0:19:20 time: 0.1552 data_time: 0.0047 memory: 7058 loss_node: 0.0393 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0394 2022/08/25 16:28:47 - mmengine - INFO - Epoch(train) [44][200/317] lr: 1.0000e-04 eta: 0:18:57 time: 0.1560 data_time: 0.0046 memory: 5374 loss_node: 0.0365 loss_edge: 0.0000 acc_node: 96.4602 acc_edge: 100.0000 loss: 0.0365 2022/08/25 16:29:21 - mmengine - INFO - Epoch(train) [44][300/317] lr: 1.0000e-04 eta: 0:18:40 time: 0.4693 data_time: 0.2749 memory: 7057 loss_node: 0.0496 loss_edge: 0.0000 acc_node: 98.3051 acc_edge: 100.0000 loss: 0.0496 2022/08/25 16:29:24 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:29:24 - mmengine - INFO - Saving checkpoint at 44 epochs 2022/08/25 16:29:41 - mmengine - INFO - Epoch(val) [44][100/472] eta: 0:02:00 time: 0.3227 data_time: 0.1686 memory: 7057 2022/08/25 16:29:54 - mmengine - INFO - Epoch(val) [44][200/472] eta: 0:00:13 time: 0.0505 data_time: 0.0260 memory: 354 2022/08/25 16:30:03 - mmengine - INFO - Epoch(val) [44][300/472] eta: 0:00:07 time: 0.0414 data_time: 0.0183 memory: 388 2022/08/25 16:30:21 - mmengine - INFO - Epoch(val) [44][400/472] eta: 0:00:09 time: 0.1267 data_time: 0.1009 memory: 354 2022/08/25 16:30:24 - mmengine - INFO - Epoch(val) [44][472/472] kie/macro_f1: 0.8858 2022/08/25 16:30:37 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:30:50 - mmengine - INFO - Epoch(train) [45][100/317] lr: 1.0000e-04 eta: 0:18:15 time: 0.5051 data_time: 0.0166 memory: 19641 loss_node: 0.0302 loss_edge: 0.0000 acc_node: 98.2143 acc_edge: 100.0000 loss: 0.0303 2022/08/25 16:31:11 - mmengine - INFO - Epoch(train) [45][200/317] lr: 1.0000e-04 eta: 0:17:53 time: 0.2682 data_time: 0.0105 memory: 7620 loss_node: 0.0411 loss_edge: 0.0000 acc_node: 98.2143 acc_edge: 100.0000 loss: 0.0412 2022/08/25 16:31:31 - mmengine - INFO - Epoch(train) [45][300/317] lr: 1.0000e-04 eta: 0:17:30 time: 0.3128 data_time: 0.0142 memory: 5853 loss_node: 0.0644 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0644 2022/08/25 16:31:34 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:31:34 - mmengine - INFO - Saving checkpoint at 45 epochs 2022/08/25 16:31:41 - mmengine - INFO - Epoch(val) [45][100/472] eta: 0:00:10 time: 0.0271 data_time: 0.0063 memory: 6403 2022/08/25 16:31:45 - mmengine - INFO - Epoch(val) [45][200/472] eta: 0:00:14 time: 0.0530 data_time: 0.0309 memory: 353 2022/08/25 16:31:48 - mmengine - INFO - Epoch(val) [45][300/472] eta: 0:00:08 time: 0.0493 data_time: 0.0254 memory: 388 2022/08/25 16:31:54 - mmengine - INFO - Epoch(val) [45][400/472] eta: 0:00:01 time: 0.0262 data_time: 0.0018 memory: 353 2022/08/25 16:31:56 - mmengine - INFO - Epoch(val) [45][472/472] kie/macro_f1: 0.8826 2022/08/25 16:32:16 - mmengine - INFO - Epoch(train) [46][100/317] lr: 1.0000e-04 eta: 0:17:03 time: 0.1526 data_time: 0.0055 memory: 7057 loss_node: 0.0432 loss_edge: 0.0000 acc_node: 98.8571 acc_edge: 100.0000 loss: 0.0432 2022/08/25 16:32:35 - mmengine - INFO - Epoch(train) [46][200/317] lr: 1.0000e-04 eta: 0:16:40 time: 0.1520 data_time: 0.0053 memory: 6403 loss_node: 0.0365 loss_edge: 0.0000 acc_node: 98.8764 acc_edge: 100.0000 loss: 0.0365 2022/08/25 16:32:53 - mmengine - INFO - Epoch(train) [46][300/317] lr: 1.0000e-04 eta: 0:16:17 time: 0.1845 data_time: 0.0066 memory: 6403 loss_node: 0.0513 loss_edge: 0.0000 acc_node: 98.3696 acc_edge: 100.0000 loss: 0.0513 2022/08/25 16:32:56 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:32:56 - mmengine - INFO - Saving checkpoint at 46 epochs 2022/08/25 16:33:04 - mmengine - INFO - Epoch(val) [46][100/472] eta: 0:00:14 time: 0.0385 data_time: 0.0168 memory: 5397 2022/08/25 16:33:08 - mmengine - INFO - Epoch(val) [46][200/472] eta: 0:00:11 time: 0.0428 data_time: 0.0190 memory: 353 2022/08/25 16:33:13 - mmengine - INFO - Epoch(val) [46][300/472] eta: 0:00:05 time: 0.0348 data_time: 0.0121 memory: 388 2022/08/25 16:33:17 - mmengine - INFO - Epoch(val) [46][400/472] eta: 0:00:01 time: 0.0255 data_time: 0.0031 memory: 353 2022/08/25 16:33:20 - mmengine - INFO - Epoch(val) [46][472/472] kie/macro_f1: 0.8840 2022/08/25 16:33:38 - mmengine - INFO - Epoch(train) [47][100/317] lr: 1.0000e-04 eta: 0:15:49 time: 0.1564 data_time: 0.0049 memory: 7540 loss_node: 0.0329 loss_edge: 0.0000 acc_node: 98.5401 acc_edge: 100.0000 loss: 0.0329 2022/08/25 16:33:58 - mmengine - INFO - Epoch(train) [47][200/317] lr: 1.0000e-04 eta: 0:15:27 time: 0.2848 data_time: 0.0063 memory: 7057 loss_node: 0.0319 loss_edge: 0.0000 acc_node: 98.7730 acc_edge: 100.0000 loss: 0.0319 2022/08/25 16:34:17 - mmengine - INFO - Epoch(train) [47][300/317] lr: 1.0000e-04 eta: 0:15:04 time: 0.1589 data_time: 0.0049 memory: 6403 loss_node: 0.0298 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0298 2022/08/25 16:34:20 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:34:20 - mmengine - INFO - Saving checkpoint at 47 epochs 2022/08/25 16:34:26 - mmengine - INFO - Epoch(val) [47][100/472] eta: 0:00:12 time: 0.0348 data_time: 0.0135 memory: 4875 2022/08/25 16:34:29 - mmengine - INFO - Epoch(val) [47][200/472] eta: 0:00:15 time: 0.0583 data_time: 0.0328 memory: 353 2022/08/25 16:34:33 - mmengine - INFO - Epoch(val) [47][300/472] eta: 0:00:07 time: 0.0447 data_time: 0.0203 memory: 388 2022/08/25 16:34:37 - mmengine - INFO - Epoch(val) [47][400/472] eta: 0:00:02 time: 0.0373 data_time: 0.0141 memory: 353 2022/08/25 16:34:44 - mmengine - INFO - Epoch(val) [47][472/472] kie/macro_f1: 0.8817 2022/08/25 16:35:06 - mmengine - INFO - Epoch(train) [48][100/317] lr: 1.0000e-04 eta: 0:14:38 time: 0.1564 data_time: 0.0050 memory: 7540 loss_node: 0.0235 loss_edge: 0.0000 acc_node: 98.5507 acc_edge: 100.0000 loss: 0.0235 2022/08/25 16:35:06 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:35:26 - mmengine - INFO - Epoch(train) [48][200/317] lr: 1.0000e-04 eta: 0:14:16 time: 0.1559 data_time: 0.0039 memory: 7059 loss_node: 0.0243 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0243 2022/08/25 16:35:47 - mmengine - INFO - Epoch(train) [48][300/317] lr: 1.0000e-04 eta: 0:13:54 time: 0.1683 data_time: 0.0057 memory: 7059 loss_node: 0.0511 loss_edge: 0.0000 acc_node: 97.5610 acc_edge: 100.0000 loss: 0.0511 2022/08/25 16:35:51 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:35:51 - mmengine - INFO - Saving checkpoint at 48 epochs 2022/08/25 16:35:56 - mmengine - INFO - Epoch(val) [48][100/472] eta: 0:00:09 time: 0.0248 data_time: 0.0032 memory: 10024 2022/08/25 16:36:00 - mmengine - INFO - Epoch(val) [48][200/472] eta: 0:00:20 time: 0.0748 data_time: 0.0289 memory: 353 2022/08/25 16:36:03 - mmengine - INFO - Epoch(val) [48][300/472] eta: 0:00:07 time: 0.0461 data_time: 0.0221 memory: 388 2022/08/25 16:36:06 - mmengine - INFO - Epoch(val) [48][400/472] eta: 0:00:02 time: 0.0292 data_time: 0.0059 memory: 353 2022/08/25 16:36:09 - mmengine - INFO - Epoch(val) [48][472/472] kie/macro_f1: 0.8835 2022/08/25 16:36:30 - mmengine - INFO - Epoch(train) [49][100/317] lr: 1.0000e-04 eta: 0:13:27 time: 0.3544 data_time: 0.0049 memory: 7852 loss_node: 0.0360 loss_edge: 0.0000 acc_node: 98.6014 acc_edge: 100.0000 loss: 0.0360 2022/08/25 16:36:49 - mmengine - INFO - Epoch(train) [49][200/317] lr: 1.0000e-04 eta: 0:13:05 time: 0.1647 data_time: 0.0073 memory: 8473 loss_node: 0.0255 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0255 2022/08/25 16:37:26 - mmengine - INFO - Epoch(train) [49][300/317] lr: 1.0000e-04 eta: 0:12:46 time: 0.4149 data_time: 0.2462 memory: 6404 loss_node: 0.0351 loss_edge: 0.0000 acc_node: 99.2248 acc_edge: 100.0000 loss: 0.0351 2022/08/25 16:37:32 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:37:32 - mmengine - INFO - Saving checkpoint at 49 epochs 2022/08/25 16:37:46 - mmengine - INFO - Epoch(val) [49][100/472] eta: 0:00:20 time: 0.0553 data_time: 0.0312 memory: 4877 2022/08/25 16:37:55 - mmengine - INFO - Epoch(val) [49][200/472] eta: 0:00:18 time: 0.0695 data_time: 0.0465 memory: 354 2022/08/25 16:37:59 - mmengine - INFO - Epoch(val) [49][300/472] eta: 0:00:06 time: 0.0368 data_time: 0.0136 memory: 389 2022/08/25 16:38:05 - mmengine - INFO - Epoch(val) [49][400/472] eta: 0:00:09 time: 0.1295 data_time: 0.1043 memory: 354 2022/08/25 16:38:09 - mmengine - INFO - Epoch(val) [49][472/472] kie/macro_f1: 0.8815 2022/08/25 16:38:35 - mmengine - INFO - Epoch(train) [50][100/317] lr: 1.0000e-04 eta: 0:12:22 time: 0.1617 data_time: 0.0050 memory: 7060 loss_node: 0.0285 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0285 2022/08/25 16:38:56 - mmengine - INFO - Epoch(train) [50][200/317] lr: 1.0000e-04 eta: 0:12:00 time: 0.1667 data_time: 0.0052 memory: 7060 loss_node: 0.0182 loss_edge: 0.0000 acc_node: 99.2000 acc_edge: 100.0000 loss: 0.0182 2022/08/25 16:39:19 - mmengine - INFO - Epoch(train) [50][300/317] lr: 1.0000e-04 eta: 0:11:38 time: 0.1674 data_time: 0.0048 memory: 8189 loss_node: 0.0279 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0279 2022/08/25 16:39:23 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:39:23 - mmengine - INFO - Saving checkpoint at 50 epochs 2022/08/25 16:39:31 - mmengine - INFO - Epoch(val) [50][100/472] eta: 0:00:11 time: 0.0308 data_time: 0.0094 memory: 5966 2022/08/25 16:39:35 - mmengine - INFO - Epoch(val) [50][200/472] eta: 0:00:10 time: 0.0389 data_time: 0.0154 memory: 354 2022/08/25 16:39:38 - mmengine - INFO - Epoch(val) [50][300/472] eta: 0:00:07 time: 0.0437 data_time: 0.0216 memory: 388 2022/08/25 16:39:44 - mmengine - INFO - Epoch(val) [50][400/472] eta: 0:00:02 time: 0.0320 data_time: 0.0077 memory: 354 2022/08/25 16:39:49 - mmengine - INFO - Epoch(val) [50][472/472] kie/macro_f1: 0.8837 2022/08/25 16:40:11 - mmengine - INFO - Epoch(train) [51][100/317] lr: 1.0000e-05 eta: 0:11:12 time: 0.1999 data_time: 0.0049 memory: 7057 loss_node: 0.0266 loss_edge: 0.0000 acc_node: 98.2659 acc_edge: 100.0000 loss: 0.0266 2022/08/25 16:40:19 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:40:29 - mmengine - INFO - Epoch(train) [51][200/317] lr: 1.0000e-05 eta: 0:10:50 time: 0.1531 data_time: 0.0048 memory: 7057 loss_node: 0.0254 loss_edge: 0.0000 acc_node: 99.3464 acc_edge: 100.0000 loss: 0.0254 2022/08/25 16:40:49 - mmengine - INFO - Epoch(train) [51][300/317] lr: 1.0000e-05 eta: 0:10:27 time: 0.1665 data_time: 0.0048 memory: 6403 loss_node: 0.0177 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0178 2022/08/25 16:40:53 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:40:53 - mmengine - INFO - Saving checkpoint at 51 epochs 2022/08/25 16:40:59 - mmengine - INFO - Epoch(val) [51][100/472] eta: 0:00:15 time: 0.0413 data_time: 0.0189 memory: 6582 2022/08/25 16:41:04 - mmengine - INFO - Epoch(val) [51][200/472] eta: 0:00:10 time: 0.0387 data_time: 0.0133 memory: 353 2022/08/25 16:41:08 - mmengine - INFO - Epoch(val) [51][300/472] eta: 0:00:06 time: 0.0363 data_time: 0.0141 memory: 388 2022/08/25 16:41:13 - mmengine - INFO - Epoch(val) [51][400/472] eta: 0:00:01 time: 0.0278 data_time: 0.0023 memory: 353 2022/08/25 16:41:17 - mmengine - INFO - Epoch(val) [51][472/472] kie/macro_f1: 0.8851 2022/08/25 16:41:39 - mmengine - INFO - Epoch(train) [52][100/317] lr: 1.0000e-05 eta: 0:10:01 time: 0.1848 data_time: 0.0051 memory: 7780 loss_node: 0.0307 loss_edge: 0.0000 acc_node: 99.3865 acc_edge: 100.0000 loss: 0.0307 2022/08/25 16:41:58 - mmengine - INFO - Epoch(train) [52][200/317] lr: 1.0000e-05 eta: 0:09:39 time: 0.1696 data_time: 0.0048 memory: 6437 loss_node: 0.0251 loss_edge: 0.0000 acc_node: 98.5612 acc_edge: 100.0000 loss: 0.0251 2022/08/25 16:42:20 - mmengine - INFO - Epoch(train) [52][300/317] lr: 1.0000e-05 eta: 0:09:17 time: 0.1844 data_time: 0.0047 memory: 6402 loss_node: 0.0177 loss_edge: 0.0000 acc_node: 98.4962 acc_edge: 100.0000 loss: 0.0177 2022/08/25 16:42:23 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:42:23 - mmengine - INFO - Saving checkpoint at 52 epochs 2022/08/25 16:42:29 - mmengine - INFO - Epoch(val) [52][100/472] eta: 0:00:09 time: 0.0244 data_time: 0.0037 memory: 6402 2022/08/25 16:42:33 - mmengine - INFO - Epoch(val) [52][200/472] eta: 0:00:09 time: 0.0362 data_time: 0.0202 memory: 354 2022/08/25 16:42:36 - mmengine - INFO - Epoch(val) [52][300/472] eta: 0:00:06 time: 0.0361 data_time: 0.0140 memory: 388 2022/08/25 16:42:41 - mmengine - INFO - Epoch(val) [52][400/472] eta: 0:00:02 time: 0.0318 data_time: 0.0104 memory: 354 2022/08/25 16:42:43 - mmengine - INFO - Epoch(val) [52][472/472] kie/macro_f1: 0.8852 2022/08/25 16:43:02 - mmengine - INFO - Epoch(train) [53][100/317] lr: 1.0000e-05 eta: 0:08:51 time: 0.1529 data_time: 0.0048 memory: 7058 loss_node: 0.0183 loss_edge: 0.0000 acc_node: 99.3464 acc_edge: 100.0000 loss: 0.0183 2022/08/25 16:43:24 - mmengine - INFO - Epoch(train) [53][200/317] lr: 1.0000e-05 eta: 0:08:29 time: 0.1677 data_time: 0.0055 memory: 6402 loss_node: 0.0185 loss_edge: 0.0000 acc_node: 99.0826 acc_edge: 100.0000 loss: 0.0185 2022/08/25 16:43:44 - mmengine - INFO - Epoch(train) [53][300/317] lr: 1.0000e-05 eta: 0:08:07 time: 0.1603 data_time: 0.0050 memory: 5853 loss_node: 0.0245 loss_edge: 0.0000 acc_node: 99.5575 acc_edge: 100.0000 loss: 0.0245 2022/08/25 16:43:48 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:43:48 - mmengine - INFO - Saving checkpoint at 53 epochs 2022/08/25 16:43:53 - mmengine - INFO - Epoch(val) [53][100/472] eta: 0:00:09 time: 0.0267 data_time: 0.0055 memory: 4877 2022/08/25 16:43:57 - mmengine - INFO - Epoch(val) [53][200/472] eta: 0:00:09 time: 0.0339 data_time: 0.0118 memory: 353 2022/08/25 16:44:01 - mmengine - INFO - Epoch(val) [53][300/472] eta: 0:00:23 time: 0.1395 data_time: 0.0177 memory: 388 2022/08/25 16:44:05 - mmengine - INFO - Epoch(val) [53][400/472] eta: 0:00:02 time: 0.0326 data_time: 0.0089 memory: 353 2022/08/25 16:44:08 - mmengine - INFO - Epoch(val) [53][472/472] kie/macro_f1: 0.8854 2022/08/25 16:44:28 - mmengine - INFO - Epoch(train) [54][100/317] lr: 1.0000e-05 eta: 0:07:41 time: 0.1573 data_time: 0.0053 memory: 7058 loss_node: 0.0232 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0232 2022/08/25 16:44:46 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:44:47 - mmengine - INFO - Epoch(train) [54][200/317] lr: 1.0000e-05 eta: 0:07:19 time: 0.1707 data_time: 0.0073 memory: 7058 loss_node: 0.0230 loss_edge: 0.0000 acc_node: 99.2647 acc_edge: 100.0000 loss: 0.0230 2022/08/25 16:45:16 - mmengine - INFO - Epoch(train) [54][300/317] lr: 1.0000e-05 eta: 0:06:58 time: 0.1624 data_time: 0.0055 memory: 7058 loss_node: 0.0180 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0180 2022/08/25 16:45:19 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:45:19 - mmengine - INFO - Saving checkpoint at 54 epochs 2022/08/25 16:45:37 - mmengine - INFO - Epoch(val) [54][100/472] eta: 0:00:33 time: 0.0909 data_time: 0.0679 memory: 5206 2022/08/25 16:45:50 - mmengine - INFO - Epoch(val) [54][200/472] eta: 0:00:34 time: 0.1254 data_time: 0.0993 memory: 353 2022/08/25 16:45:59 - mmengine - INFO - Epoch(val) [54][300/472] eta: 0:00:14 time: 0.0842 data_time: 0.0610 memory: 389 2022/08/25 16:46:07 - mmengine - INFO - Epoch(val) [54][400/472] eta: 0:00:01 time: 0.0273 data_time: 0.0041 memory: 353 2022/08/25 16:46:14 - mmengine - INFO - Epoch(val) [54][472/472] kie/macro_f1: 0.8849 2022/08/25 16:46:42 - mmengine - INFO - Epoch(train) [55][100/317] lr: 1.0000e-05 eta: 0:06:33 time: 0.3227 data_time: 0.0313 memory: 7058 loss_node: 0.0288 loss_edge: 0.0000 acc_node: 99.3590 acc_edge: 100.0000 loss: 0.0288 2022/08/25 16:47:03 - mmengine - INFO - Epoch(train) [55][200/317] lr: 1.0000e-05 eta: 0:06:11 time: 0.4510 data_time: 0.0058 memory: 6579 loss_node: 0.0184 loss_edge: 0.0000 acc_node: 97.5758 acc_edge: 100.0000 loss: 0.0184 2022/08/25 16:47:25 - mmengine - INFO - Epoch(train) [55][300/317] lr: 1.0000e-05 eta: 0:05:49 time: 0.2007 data_time: 0.0571 memory: 7058 loss_node: 0.0243 loss_edge: 0.0000 acc_node: 99.4012 acc_edge: 100.0000 loss: 0.0243 2022/08/25 16:47:28 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:47:28 - mmengine - INFO - Saving checkpoint at 55 epochs 2022/08/25 16:47:48 - mmengine - INFO - Epoch(val) [55][100/472] eta: 0:00:12 time: 0.0348 data_time: 0.0127 memory: 7058 2022/08/25 16:47:56 - mmengine - INFO - Epoch(val) [55][200/472] eta: 0:00:11 time: 0.0434 data_time: 0.0184 memory: 353 2022/08/25 16:48:02 - mmengine - INFO - Epoch(val) [55][300/472] eta: 0:00:07 time: 0.0429 data_time: 0.0285 memory: 388 2022/08/25 16:48:07 - mmengine - INFO - Epoch(val) [55][400/472] eta: 0:00:03 time: 0.0447 data_time: 0.0230 memory: 353 2022/08/25 16:48:12 - mmengine - INFO - Epoch(val) [55][472/472] kie/macro_f1: 0.8851 2022/08/25 16:48:33 - mmengine - INFO - Epoch(train) [56][100/317] lr: 1.0000e-05 eta: 0:05:24 time: 0.3285 data_time: 0.0205 memory: 6628 loss_node: 0.0157 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0157 2022/08/25 16:48:54 - mmengine - INFO - Epoch(train) [56][200/317] lr: 1.0000e-05 eta: 0:05:02 time: 0.4919 data_time: 0.0260 memory: 6628 loss_node: 0.0221 loss_edge: 0.0000 acc_node: 97.0149 acc_edge: 100.0000 loss: 0.0221 2022/08/25 16:49:11 - mmengine - INFO - Epoch(train) [56][300/317] lr: 1.0000e-05 eta: 0:04:39 time: 0.2107 data_time: 0.0053 memory: 7058 loss_node: 0.0177 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0177 2022/08/25 16:49:15 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:49:15 - mmengine - INFO - Saving checkpoint at 56 epochs 2022/08/25 16:49:20 - mmengine - INFO - Epoch(val) [56][100/472] eta: 0:00:11 time: 0.0306 data_time: 0.0100 memory: 6143 2022/08/25 16:49:23 - mmengine - INFO - Epoch(val) [56][200/472] eta: 0:00:09 time: 0.0360 data_time: 0.0160 memory: 353 2022/08/25 16:49:27 - mmengine - INFO - Epoch(val) [56][300/472] eta: 0:00:06 time: 0.0398 data_time: 0.0179 memory: 388 2022/08/25 16:49:32 - mmengine - INFO - Epoch(val) [56][400/472] eta: 0:00:02 time: 0.0320 data_time: 0.0096 memory: 353 2022/08/25 16:49:35 - mmengine - INFO - Epoch(val) [56][472/472] kie/macro_f1: 0.8841 2022/08/25 16:49:54 - mmengine - INFO - Epoch(train) [57][100/317] lr: 1.0000e-05 eta: 0:04:14 time: 0.1618 data_time: 0.0057 memory: 6621 loss_node: 0.0171 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0171 2022/08/25 16:50:11 - mmengine - INFO - Epoch(train) [57][200/317] lr: 1.0000e-05 eta: 0:03:52 time: 0.2316 data_time: 0.0046 memory: 7539 loss_node: 0.0249 loss_edge: 0.0000 acc_node: 99.3976 acc_edge: 100.0000 loss: 0.0249 2022/08/25 16:50:22 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:50:34 - mmengine - INFO - Epoch(train) [57][300/317] lr: 1.0000e-05 eta: 0:03:30 time: 0.4807 data_time: 0.0212 memory: 6402 loss_node: 0.0163 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0163 2022/08/25 16:50:37 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:50:37 - mmengine - INFO - Saving checkpoint at 57 epochs 2022/08/25 16:50:43 - mmengine - INFO - Epoch(val) [57][100/472] eta: 0:00:12 time: 0.0333 data_time: 0.0124 memory: 5751 2022/08/25 16:50:46 - mmengine - INFO - Epoch(val) [57][200/472] eta: 0:00:11 time: 0.0408 data_time: 0.0161 memory: 353 2022/08/25 16:50:49 - mmengine - INFO - Epoch(val) [57][300/472] eta: 0:00:07 time: 0.0450 data_time: 0.0221 memory: 388 2022/08/25 16:50:55 - mmengine - INFO - Epoch(val) [57][400/472] eta: 0:00:01 time: 0.0249 data_time: 0.0040 memory: 353 2022/08/25 16:50:58 - mmengine - INFO - Epoch(val) [57][472/472] kie/macro_f1: 0.8830 2022/08/25 16:51:18 - mmengine - INFO - Epoch(train) [58][100/317] lr: 1.0000e-05 eta: 0:03:04 time: 0.1558 data_time: 0.0049 memory: 10299 loss_node: 0.0246 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0246 2022/08/25 16:51:39 - mmengine - INFO - Epoch(train) [58][200/317] lr: 1.0000e-05 eta: 0:02:43 time: 0.1690 data_time: 0.0049 memory: 7300 loss_node: 0.0268 loss_edge: 0.0000 acc_node: 99.4565 acc_edge: 100.0000 loss: 0.0268 2022/08/25 16:52:00 - mmengine - INFO - Epoch(train) [58][300/317] lr: 1.0000e-05 eta: 0:02:21 time: 0.1541 data_time: 0.0047 memory: 6405 loss_node: 0.0154 loss_edge: 0.0000 acc_node: 99.4253 acc_edge: 100.0000 loss: 0.0154 2022/08/25 16:52:04 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:52:04 - mmengine - INFO - Saving checkpoint at 58 epochs 2022/08/25 16:52:09 - mmengine - INFO - Epoch(val) [58][100/472] eta: 0:00:08 time: 0.0241 data_time: 0.0038 memory: 7059 2022/08/25 16:52:13 - mmengine - INFO - Epoch(val) [58][200/472] eta: 0:00:11 time: 0.0416 data_time: 0.0124 memory: 353 2022/08/25 16:52:16 - mmengine - INFO - Epoch(val) [58][300/472] eta: 0:00:05 time: 0.0337 data_time: 0.0116 memory: 388 2022/08/25 16:52:19 - mmengine - INFO - Epoch(val) [58][400/472] eta: 0:00:01 time: 0.0216 data_time: 0.0039 memory: 353 2022/08/25 16:52:25 - mmengine - INFO - Epoch(val) [58][472/472] kie/macro_f1: 0.8833 2022/08/25 16:52:44 - mmengine - INFO - Epoch(train) [59][100/317] lr: 1.0000e-05 eta: 0:01:55 time: 0.3503 data_time: 0.0170 memory: 7058 loss_node: 0.0122 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0122 2022/08/25 16:53:07 - mmengine - INFO - Epoch(train) [59][200/317] lr: 1.0000e-05 eta: 0:01:34 time: 0.2056 data_time: 0.0531 memory: 19529 loss_node: 0.0256 loss_edge: 0.0000 acc_node: 97.3118 acc_edge: 100.0000 loss: 0.0256 2022/08/25 16:53:34 - mmengine - INFO - Epoch(train) [59][300/317] lr: 1.0000e-05 eta: 0:01:12 time: 0.1516 data_time: 0.0049 memory: 7057 loss_node: 0.0239 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0239 2022/08/25 16:53:39 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:53:39 - mmengine - INFO - Saving checkpoint at 59 epochs 2022/08/25 16:53:50 - mmengine - INFO - Epoch(val) [59][100/472] eta: 0:00:27 time: 0.0747 data_time: 0.0542 memory: 7280 2022/08/25 16:54:03 - mmengine - INFO - Epoch(val) [59][200/472] eta: 0:00:13 time: 0.0508 data_time: 0.0271 memory: 353 2022/08/25 16:54:08 - mmengine - INFO - Epoch(val) [59][300/472] eta: 0:00:11 time: 0.0652 data_time: 0.0404 memory: 388 2022/08/25 16:54:15 - mmengine - INFO - Epoch(val) [59][400/472] eta: 0:00:05 time: 0.0802 data_time: 0.0633 memory: 353 2022/08/25 16:54:19 - mmengine - INFO - Epoch(val) [59][472/472] kie/macro_f1: 0.8837 2022/08/25 16:54:40 - mmengine - INFO - Epoch(train) [60][100/317] lr: 1.0000e-05 eta: 0:00:47 time: 0.1626 data_time: 0.0053 memory: 6096 loss_node: 0.0205 loss_edge: 0.0000 acc_node: 99.1228 acc_edge: 100.0000 loss: 0.0205 2022/08/25 16:55:07 - mmengine - INFO - Epoch(train) [60][200/317] lr: 1.0000e-05 eta: 0:00:25 time: 0.1490 data_time: 0.0051 memory: 7057 loss_node: 0.0208 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0208 2022/08/25 16:55:30 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:55:31 - mmengine - INFO - Epoch(train) [60][300/317] lr: 1.0000e-05 eta: 0:00:03 time: 0.1697 data_time: 0.0105 memory: 6403 loss_node: 0.0215 loss_edge: 0.0000 acc_node: 100.0000 acc_edge: 100.0000 loss: 0.0215 2022/08/25 16:55:35 - mmengine - INFO - Exp name: sdmgr_unet16_60e_wildreceipt_20220825_151648 2022/08/25 16:55:35 - mmengine - INFO - Saving checkpoint at 60 epochs 2022/08/25 16:55:57 - mmengine - INFO - Epoch(val) [60][100/472] eta: 0:00:10 time: 0.0277 data_time: 0.0138 memory: 5906 2022/08/25 16:56:04 - mmengine - INFO - Epoch(val) [60][200/472] eta: 0:00:21 time: 0.0804 data_time: 0.0545 memory: 353 2022/08/25 16:56:08 - mmengine - INFO - Epoch(val) [60][300/472] eta: 0:00:07 time: 0.0458 data_time: 0.0224 memory: 388 2022/08/25 16:56:16 - mmengine - INFO - Epoch(val) [60][400/472] eta: 0:00:01 time: 0.0246 data_time: 0.0017 memory: 353 2022/08/25 16:56:20 - mmengine - INFO - Epoch(val) [60][472/472] kie/macro_f1: 0.8832