2023/04/13 22:42:44 - mmengine - INFO - ------------------------------------------------------------ System environment: sys.platform: linux Python: 3.7.16 (default, Jan 17 2023, 22:20:44) [GCC 11.2.0] CUDA available: True numpy_random_seed: 83034293 GPU 0,1,2,3,4,5,6,7: NVIDIA A100-SXM4-80GB CUDA_HOME: None GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44) PyTorch: 1.13.1+cu116 PyTorch compiling details: PyTorch built with: - GCC 9.3 - C++ Version: 201402 - Intel(R) Math Kernel Library Version 2020.0.0 Product Build 20191122 for Intel(R) 64 architecture applications - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) - OpenMP 201511 (a.k.a. OpenMP 4.5) - LAPACK is enabled (usually provided by MKL) - NNPACK is enabled - CPU capability usage: AVX2 - CUDA Runtime 11.6 - 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 8.3.2 (built against CUDA 11.5) - Magma 2.6.1 - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.6, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wunused-local-typedefs -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.13.1, 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, USE_ROCM=OFF, TorchVision: 0.14.1+cu116 OpenCV: 4.7.0 MMEngine: 0.7.0 Runtime environment: cudnn_benchmark: False mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} dist_cfg: {'backend': 'nccl'} seed: None deterministic: False Distributed launcher: pytorch Distributed training: True GPU number: 8 ------------------------------------------------------------ 2023/04/13 22:42:45 - mmengine - INFO - Config: model = dict( _scope_='mmrazor', type='MMArchitectureQuant', data_preprocessor=dict( type='mmcls.ClsDataPreprocessor', num_classes=1000, mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True), architecture=dict( type='ImageClassifier', backbone=dict( type='ResNet', depth=18, num_stages=4, out_indices=(3, ), style='pytorch'), neck=dict(type='GlobalAveragePooling'), head=dict( type='LinearClsHead', num_classes=1000, in_channels=512, loss=dict(type='CrossEntropyLoss', loss_weight=1.0), topk=(1, 5)), _scope_='mmcls'), float_checkpoint= 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet18_8xb32_in1k_20210831-fbbb1da6.pth', quantizer=dict( type='mmrazor.OpenVINOQuantizer', global_qconfig=dict( w_observer=dict(type='mmrazor.LSQPerChannelObserver'), a_observer=dict(type='mmrazor.LSQObserver'), w_fake_quant=dict(type='mmrazor.LearnableFakeQuantize'), a_fake_quant=dict(type='mmrazor.LearnableFakeQuantize'), w_qscheme=dict( qdtype='qint8', bit=8, is_symmetry=True, is_symmetric_range=True), a_qscheme=dict(qdtype='quint8', bit=8, is_symmetry=True)), tracer=dict( type='mmrazor.CustomTracer', skipped_methods=[ 'mmcls.models.heads.ClsHead._get_loss', 'mmcls.models.heads.ClsHead._get_predictions' ]))) dataset_type = 'ImageNet' data_preprocessor = dict( num_classes=1000, mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True) train_pipeline = [ dict(type='LoadImageFromFile', _scope_='mmcls'), dict(type='RandomResizedCrop', scale=224, _scope_='mmcls'), dict(type='RandomFlip', prob=0.5, direction='horizontal', _scope_='mmcls'), dict(type='PackClsInputs', _scope_='mmcls') ] test_pipeline = [ dict(type='LoadImageFromFile', _scope_='mmcls'), dict(type='ResizeEdge', scale=256, edge='short', _scope_='mmcls'), dict(type='CenterCrop', crop_size=224, _scope_='mmcls'), dict(type='PackClsInputs', _scope_='mmcls') ] train_dataloader = dict( batch_size=32, num_workers=5, dataset=dict( type='ImageNet', data_root='data/imagenet', ann_file='meta/train.txt', data_prefix='train', pipeline=[ dict(type='LoadImageFromFile'), dict(type='RandomResizedCrop', scale=224), dict(type='RandomFlip', prob=0.5, direction='horizontal'), dict(type='PackClsInputs') ], _scope_='mmcls'), sampler=dict(type='DefaultSampler', shuffle=True, _scope_='mmcls')) val_dataloader = dict( batch_size=32, num_workers=5, dataset=dict( type='ImageNet', data_root='data/imagenet', ann_file='meta/val.txt', data_prefix='val', pipeline=[ dict(type='LoadImageFromFile'), dict(type='ResizeEdge', scale=256, edge='short'), dict(type='CenterCrop', crop_size=224), dict(type='PackClsInputs') ], _scope_='mmcls'), sampler=dict(type='DefaultSampler', shuffle=False, _scope_='mmcls')) val_evaluator = dict(type='Accuracy', topk=(1, 5), _scope_='mmcls') test_dataloader = dict( batch_size=32, num_workers=5, dataset=dict( type='ImageNet', data_root='data/imagenet', ann_file='meta/val.txt', data_prefix='val', pipeline=[ dict(type='LoadImageFromFile'), dict(type='ResizeEdge', scale=256, edge='short'), dict(type='CenterCrop', crop_size=224), dict(type='PackClsInputs') ], _scope_='mmcls'), sampler=dict(type='DefaultSampler', shuffle=False, _scope_='mmcls')) test_evaluator = dict(type='Accuracy', topk=(1, 5), _scope_='mmcls') optim_wrapper = dict( optimizer=dict( type='SGD', lr=0.0001, momentum=0.9, weight_decay=0.0001, _scope_='mmcls')) param_scheduler = dict(type='ConstantLR', factor=1.0, by_epoch=True) train_cfg = dict( type='mmrazor.LSQEpochBasedLoop', max_epochs=10, val_interval=1) val_cfg = dict(type='mmrazor.QATValLoop') test_cfg = dict(type='mmrazor.QATValLoop') auto_scale_lr = dict(base_batch_size=256) default_scope = 'mmcls' default_hooks = dict( timer=dict(type='IterTimerHook', _scope_='mmcls'), logger=dict(type='LoggerHook', interval=100, _scope_='mmcls'), param_scheduler=dict(type='ParamSchedulerHook', _scope_='mmcls'), checkpoint=dict(type='CheckpointHook', interval=1, _scope_='mmcls'), sampler_seed=dict(type='DistSamplerSeedHook', _scope_='mmcls'), visualization=dict( type='VisualizationHook', enable=False, _scope_='mmcls'), sync=dict(type='SyncBuffersHook')) env_cfg = dict( cudnn_benchmark=False, mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), dist_cfg=dict(backend='nccl')) vis_backends = [dict(type='LocalVisBackend', _scope_='mmcls')] visualizer = dict( type='ClsVisualizer', vis_backends=[dict(type='LocalVisBackend')], _scope_='mmcls') log_level = 'INFO' load_from = None resume = False randomness = dict(seed=None, deterministic=False) resnet = dict( type='ImageClassifier', backbone=dict( type='ResNet', depth=18, num_stages=4, out_indices=(3, ), style='pytorch'), neck=dict(type='GlobalAveragePooling'), head=dict( type='LinearClsHead', num_classes=1000, in_channels=512, loss=dict(type='CrossEntropyLoss', loss_weight=1.0), topk=(1, 5)), _scope_='mmcls') float_checkpoint = 'https://download.openmmlab.com/mmclassification/v0/resnet/resnet18_8xb32_in1k_20210831-fbbb1da6.pth' global_qconfig = dict( w_observer=dict(type='mmrazor.LSQPerChannelObserver'), a_observer=dict(type='mmrazor.LSQObserver'), w_fake_quant=dict(type='mmrazor.LearnableFakeQuantize'), a_fake_quant=dict(type='mmrazor.LearnableFakeQuantize'), w_qscheme=dict( qdtype='qint8', bit=8, is_symmetry=True, is_symmetric_range=True), a_qscheme=dict(qdtype='quint8', bit=8, is_symmetry=True)) model_wrapper_cfg = dict( type='mmrazor.MMArchitectureQuantDDP', broadcast_buffers=False, find_unused_parameters=True) launcher = 'pytorch' work_dir = 'lsq_r18_new' 2023/04/13 22:42:45 - mmengine - WARNING - The "task util" registry in mmrazor did not set import location. Fallback to call `mmrazor.utils.register_all_modules` instead. 2023/04/13 22:42:47 - mmengine - WARNING - The "model_wrapper" registry in mmrazor did not set import location. Fallback to call `mmrazor.utils.register_all_modules` instead. 2023/04/13 22:43:04 - mmengine - INFO - Hooks will be executed in the following order: before_run: (VERY_HIGH ) RuntimeInfoHook (BELOW_NORMAL) LoggerHook -------------------- before_train: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (VERY_LOW ) CheckpointHook -------------------- before_train_epoch: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (NORMAL ) DistSamplerSeedHook -------------------- before_train_iter: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook -------------------- after_train_iter: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (BELOW_NORMAL) LoggerHook (LOW ) ParamSchedulerHook (VERY_LOW ) CheckpointHook -------------------- after_train_epoch: (NORMAL ) IterTimerHook (NORMAL ) SyncBuffersHook (LOW ) ParamSchedulerHook (VERY_LOW ) CheckpointHook -------------------- before_val_epoch: (NORMAL ) IterTimerHook -------------------- before_val_iter: (NORMAL ) IterTimerHook -------------------- after_val_iter: (NORMAL ) IterTimerHook (NORMAL ) VisualizationHook (BELOW_NORMAL) LoggerHook -------------------- after_val_epoch: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (BELOW_NORMAL) LoggerHook (LOW ) ParamSchedulerHook (VERY_LOW ) CheckpointHook -------------------- before_test_epoch: (NORMAL ) IterTimerHook -------------------- before_test_iter: (NORMAL ) IterTimerHook -------------------- after_test_iter: (NORMAL ) IterTimerHook (NORMAL ) VisualizationHook (BELOW_NORMAL) LoggerHook -------------------- after_test_epoch: (VERY_HIGH ) RuntimeInfoHook (NORMAL ) IterTimerHook (BELOW_NORMAL) LoggerHook -------------------- after_run: (BELOW_NORMAL) LoggerHook -------------------- 2023/04/13 22:43:04 - mmengine - WARNING - The "loop" registry in mmrazor did not set import location. Fallback to call `mmrazor.utils.register_all_modules` instead. 2023/04/13 22:43:12 - mmengine - WARNING - init_weights of ImageClassifier has been called more than once. Name of parameter - Initialization information architecture.backbone.conv1.weight - torch.Size([64, 3, 7, 7]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.bn1.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.bn1.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer1.0.conv1.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer1.0.bn1.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer1.0.bn1.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer1.0.bn2.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer1.0.bn2.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer1.1.conv1.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer1.1.bn1.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer1.1.bn1.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer1.1.bn2.weight - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer1.1.bn2.bias - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer2.0.conv1.weight - torch.Size([128, 64, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer2.0.bn1.weight - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer2.0.bn1.bias - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer2.0.bn2.weight - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer2.0.bn2.bias - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer2.0.downsample.0.weight - torch.Size([128, 64, 1, 1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer2.0.downsample.1.weight - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer2.0.downsample.1.bias - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer2.1.conv1.weight - torch.Size([128, 128, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer2.1.bn1.weight - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer2.1.bn1.bias - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer2.1.bn2.weight - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer2.1.bn2.bias - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer3.0.conv1.weight - torch.Size([256, 128, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer3.0.bn1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer3.0.bn1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer3.0.bn2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer3.0.bn2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer3.0.downsample.0.weight - torch.Size([256, 128, 1, 1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer3.0.downsample.1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer3.0.downsample.1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer3.1.conv1.weight - torch.Size([256, 256, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer3.1.bn1.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer3.1.bn1.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer3.1.bn2.weight - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer3.1.bn2.bias - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer4.0.conv1.weight - torch.Size([512, 256, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer4.0.bn1.weight - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer4.0.bn1.bias - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer4.0.bn2.weight - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer4.0.bn2.bias - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer4.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer4.0.downsample.1.weight - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer4.0.downsample.1.bias - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer4.1.conv1.weight - torch.Size([512, 512, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer4.1.bn1.weight - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer4.1.bn1.bias - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer4.1.bn2.weight - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.backbone.layer4.1.bn2.bias - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.head.fc.weight - torch.Size([1000, 512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant architecture.head.fc.bias - torch.Size([1000]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_0.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.conv1.weight_fake_quant.scale - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer1.0.conv1.weight_fake_quant.scale - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer1.0.conv2.weight_fake_quant.scale - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer1.1.conv1.weight_fake_quant.scale - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer1.1.conv2.weight_fake_quant.scale - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer2.0.conv1.weight_fake_quant.scale - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer2.0.conv2.weight_fake_quant.scale - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer2.0.downsample.0.weight_fake_quant.scale - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer2.1.conv1.weight_fake_quant.scale - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer2.1.conv2.weight_fake_quant.scale - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer3.0.conv1.weight_fake_quant.scale - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer3.0.conv2.weight_fake_quant.scale - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer3.0.downsample.0.weight_fake_quant.scale - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer3.1.conv1.weight_fake_quant.scale - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer3.1.conv2.weight_fake_quant.scale - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer4.0.conv1.weight_fake_quant.scale - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer4.0.conv2.weight_fake_quant.scale - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer4.0.downsample.0.weight_fake_quant.scale - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer4.1.conv1.weight_fake_quant.scale - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.backbone.layer4.1.conv2.weight_fake_quant.scale - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_1.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_3.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_5.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_6.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_7.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_9.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_10.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_11.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_13.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_14.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_15.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_16.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_18.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_19.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_20.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_22.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_23.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_24.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_25.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_27.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_28.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_29.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_31.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_32.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_33.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_34.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.activation_post_process_36.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.tensor.head.fc.weight_fake_quant.scale - torch.Size([1000]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_0.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.conv1.weight_fake_quant.scale - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer1.0.conv1.weight_fake_quant.scale - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer1.0.conv2.weight_fake_quant.scale - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer1.1.conv1.weight_fake_quant.scale - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer1.1.conv2.weight_fake_quant.scale - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer2.0.conv1.weight_fake_quant.scale - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer2.0.conv2.weight_fake_quant.scale - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer2.0.downsample.0.weight_fake_quant.scale - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer2.1.conv1.weight_fake_quant.scale - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer2.1.conv2.weight_fake_quant.scale - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer3.0.conv1.weight_fake_quant.scale - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer3.0.conv2.weight_fake_quant.scale - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer3.0.downsample.0.weight_fake_quant.scale - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer3.1.conv1.weight_fake_quant.scale - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer3.1.conv2.weight_fake_quant.scale - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer4.0.conv1.weight_fake_quant.scale - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer4.0.conv2.weight_fake_quant.scale - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer4.0.downsample.0.weight_fake_quant.scale - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer4.1.conv1.weight_fake_quant.scale - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.backbone.layer4.1.conv2.weight_fake_quant.scale - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_1.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_3.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_5.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_6.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_7.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_9.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_10.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_11.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_13.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_14.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_15.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_16.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_18.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_19.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_20.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_22.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_23.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_24.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_25.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_27.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_28.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_29.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_31.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_32.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_33.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_34.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.activation_post_process_36.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.predict.head.fc.weight_fake_quant.scale - torch.Size([1000]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_0.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.conv1.weight_fake_quant.scale - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer1.0.conv1.weight_fake_quant.scale - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer1.0.conv2.weight_fake_quant.scale - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer1.1.conv1.weight_fake_quant.scale - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer1.1.conv2.weight_fake_quant.scale - torch.Size([64]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer2.0.conv1.weight_fake_quant.scale - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer2.0.conv2.weight_fake_quant.scale - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer2.0.downsample.0.weight_fake_quant.scale - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer2.1.conv1.weight_fake_quant.scale - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer2.1.conv2.weight_fake_quant.scale - torch.Size([128]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer3.0.conv1.weight_fake_quant.scale - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer3.0.conv2.weight_fake_quant.scale - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer3.0.downsample.0.weight_fake_quant.scale - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer3.1.conv1.weight_fake_quant.scale - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer3.1.conv2.weight_fake_quant.scale - torch.Size([256]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer4.0.conv1.weight_fake_quant.scale - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer4.0.conv2.weight_fake_quant.scale - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer4.0.downsample.0.weight_fake_quant.scale - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer4.1.conv1.weight_fake_quant.scale - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.backbone.layer4.1.conv2.weight_fake_quant.scale - torch.Size([512]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_1.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_3.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_5.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_6.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_7.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_9.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_10.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_11.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_13.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_14.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_15.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_16.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_18.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_19.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_20.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_22.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_23.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_24.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_25.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_27.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_28.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_29.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_31.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_32.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_33.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_34.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.activation_post_process_36.scale - torch.Size([1]): The value is the same before and after calling `init_weights` of MMArchitectureQuant qmodels.loss.head.fc.weight_fake_quant.scale - torch.Size([1000]): The value is the same before and after calling `init_weights` of MMArchitectureQuant 2023/04/13 22:43:14 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io 2023/04/13 22:43:14 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. 2023/04/13 22:43:14 - mmengine - INFO - Checkpoints will be saved to /nvme/caoweihan.p/projects/mmrazor/lsq_r18_new. 2023/04/13 22:44:09 - mmengine - INFO - Epoch(train) [1][ 100/5005] lr: 1.0000e-04 eta: 7:37:21 time: 0.5416 data_time: 0.0077 memory: 1685 loss: 1.2171 2023/04/13 22:45:00 - mmengine - INFO - Epoch(train) [1][ 200/5005] lr: 1.0000e-04 eta: 7:16:53 time: 0.5001 data_time: 0.0078 memory: 1685 loss: 1.3102 2023/04/13 22:45:51 - mmengine - INFO - Epoch(train) [1][ 300/5005] lr: 1.0000e-04 eta: 7:13:46 time: 0.4839 data_time: 0.0074 memory: 1685 loss: 1.5241 2023/04/13 22:46:46 - mmengine - INFO - Epoch(train) [1][ 400/5005] lr: 1.0000e-04 eta: 7:17:46 time: 0.5553 data_time: 0.0075 memory: 1685 loss: 1.3661 2023/04/13 22:47:41 - mmengine - INFO - Epoch(train) [1][ 500/5005] lr: 1.0000e-04 eta: 7:19:57 time: 0.5301 data_time: 0.0077 memory: 1685 loss: 1.4851 2023/04/13 22:48:32 - mmengine - INFO - Epoch(train) [1][ 600/5005] lr: 1.0000e-04 eta: 7:16:36 time: 0.4734 data_time: 0.0077 memory: 1685 loss: 1.3720 2023/04/13 22:49:28 - mmengine - INFO - Epoch(train) [1][ 700/5005] lr: 1.0000e-04 eta: 7:18:34 time: 0.5563 data_time: 0.0107 memory: 1685 loss: 1.4967 2023/04/13 22:50:23 - mmengine - INFO - Epoch(train) [1][ 800/5005] lr: 1.0000e-04 eta: 7:19:27 time: 0.5873 data_time: 0.0104 memory: 1685 loss: 1.3825 2023/04/13 22:51:15 - mmengine - INFO - Epoch(train) [1][ 900/5005] lr: 1.0000e-04 eta: 7:17:31 time: 0.5135 data_time: 0.0088 memory: 1685 loss: 1.3218 2023/04/13 22:52:10 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/13 22:52:10 - mmengine - INFO - Epoch(train) [1][1000/5005] lr: 1.0000e-04 eta: 7:18:01 time: 0.5385 data_time: 0.0109 memory: 1685 loss: 1.2135 2023/04/13 22:53:04 - mmengine - INFO - Epoch(train) [1][1100/5005] lr: 1.0000e-04 eta: 7:17:02 time: 0.5043 data_time: 0.0106 memory: 1685 loss: 1.4194 2023/04/13 22:53:59 - mmengine - INFO - Epoch(train) [1][1200/5005] lr: 1.0000e-04 eta: 7:17:02 time: 0.5291 data_time: 0.0118 memory: 1685 loss: 1.4523 2023/04/13 22:54:53 - mmengine - INFO - Epoch(train) [1][1300/5005] lr: 1.0000e-04 eta: 7:16:33 time: 0.5912 data_time: 0.0107 memory: 1685 loss: 1.2280 2023/04/13 22:55:44 - mmengine - INFO - Epoch(train) [1][1400/5005] lr: 1.0000e-04 eta: 7:14:15 time: 0.5634 data_time: 0.0113 memory: 1685 loss: 1.4407 2023/04/13 22:56:37 - mmengine - INFO - Epoch(train) [1][1500/5005] lr: 1.0000e-04 eta: 7:13:02 time: 0.5224 data_time: 0.0112 memory: 1685 loss: 1.4619 2023/04/13 22:57:31 - mmengine - INFO - Epoch(train) [1][1600/5005] lr: 1.0000e-04 eta: 7:12:10 time: 0.5324 data_time: 0.0105 memory: 1685 loss: 1.4448 2023/04/13 22:58:26 - mmengine - INFO - Epoch(train) [1][1700/5005] lr: 1.0000e-04 eta: 7:12:08 time: 0.5672 data_time: 0.0125 memory: 1685 loss: 1.3685 2023/04/13 22:59:19 - mmengine - INFO - Epoch(train) [1][1800/5005] lr: 1.0000e-04 eta: 7:11:05 time: 0.5562 data_time: 0.0112 memory: 1685 loss: 1.2678 2023/04/13 23:00:12 - mmengine - INFO - Epoch(train) [1][1900/5005] lr: 1.0000e-04 eta: 7:09:55 time: 0.5515 data_time: 0.0113 memory: 1685 loss: 1.1856 2023/04/13 23:01:07 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/13 23:01:07 - mmengine - INFO - Epoch(train) [1][2000/5005] lr: 1.0000e-04 eta: 7:09:31 time: 0.6245 data_time: 0.0101 memory: 1685 loss: 1.4513 2023/04/13 23:02:02 - mmengine - INFO - Epoch(train) [1][2100/5005] lr: 1.0000e-04 eta: 7:09:11 time: 0.5635 data_time: 0.0122 memory: 1685 loss: 1.4038 2023/04/13 23:02:57 - mmengine - INFO - Epoch(train) [1][2200/5005] lr: 1.0000e-04 eta: 7:08:33 time: 0.6015 data_time: 0.0116 memory: 1685 loss: 1.2741 2023/04/13 23:03:50 - mmengine - INFO - Epoch(train) [1][2300/5005] lr: 1.0000e-04 eta: 7:07:31 time: 0.5273 data_time: 0.0103 memory: 1685 loss: 1.1181 2023/04/13 23:04:43 - mmengine - INFO - Epoch(train) [1][2400/5005] lr: 1.0000e-04 eta: 7:06:31 time: 0.4786 data_time: 0.0104 memory: 1685 loss: 1.2893 2023/04/13 23:05:38 - mmengine - INFO - Epoch(train) [1][2500/5005] lr: 1.0000e-04 eta: 7:05:47 time: 0.5410 data_time: 0.0114 memory: 1685 loss: 1.1987 2023/04/13 23:06:32 - mmengine - INFO - Epoch(train) [1][2600/5005] lr: 1.0000e-04 eta: 7:05:13 time: 0.4927 data_time: 0.0106 memory: 1685 loss: 1.2159 2023/04/13 23:07:25 - mmengine - INFO - Epoch(train) [1][2700/5005] lr: 1.0000e-04 eta: 7:03:58 time: 0.5130 data_time: 0.0106 memory: 1685 loss: 1.2357 2023/04/13 23:08:20 - mmengine - INFO - Epoch(train) [1][2800/5005] lr: 1.0000e-04 eta: 7:03:21 time: 0.5227 data_time: 0.0094 memory: 1685 loss: 1.3303 2023/04/13 23:09:05 - mmengine - INFO - Epoch(train) [1][2900/5005] lr: 1.0000e-04 eta: 7:00:13 time: 0.3855 data_time: 0.0109 memory: 1685 loss: 1.4337 2023/04/13 23:09:59 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/13 23:09:59 - mmengine - INFO - Epoch(train) [1][3000/5005] lr: 1.0000e-04 eta: 6:59:22 time: 1.2532 data_time: 0.0095 memory: 1685 loss: 1.5872 2023/04/13 23:10:54 - mmengine - INFO - Epoch(train) [1][3100/5005] lr: 1.0000e-04 eta: 6:59:00 time: 0.6641 data_time: 0.0088 memory: 1685 loss: 1.1594 2023/04/13 23:12:12 - mmengine - INFO - Epoch(train) [1][3200/5005] lr: 1.0000e-04 eta: 7:03:54 time: 0.5377 data_time: 0.0108 memory: 1685 loss: 1.1659 2023/04/13 23:13:04 - mmengine - INFO - Epoch(train) [1][3300/5005] lr: 1.0000e-04 eta: 7:02:29 time: 0.4793 data_time: 0.0120 memory: 1685 loss: 1.4186 2023/04/13 23:13:57 - mmengine - INFO - Epoch(train) [1][3400/5005] lr: 1.0000e-04 eta: 7:01:16 time: 0.5088 data_time: 0.0113 memory: 1685 loss: 1.4583 2023/04/13 23:14:49 - mmengine - INFO - Epoch(train) [1][3500/5005] lr: 1.0000e-04 eta: 6:59:54 time: 0.5727 data_time: 0.0117 memory: 1685 loss: 1.2520 2023/04/13 23:15:42 - mmengine - INFO - Epoch(train) [1][3600/5005] lr: 1.0000e-04 eta: 6:58:48 time: 0.5882 data_time: 0.0130 memory: 1685 loss: 1.3190 2023/04/13 23:16:38 - mmengine - INFO - Epoch(train) [1][3700/5005] lr: 1.0000e-04 eta: 6:58:14 time: 0.5237 data_time: 0.0119 memory: 1685 loss: 1.2687 2023/04/13 23:17:30 - mmengine - INFO - Epoch(train) [1][3800/5005] lr: 1.0000e-04 eta: 6:57:02 time: 0.5744 data_time: 0.0108 memory: 1685 loss: 1.3417 2023/04/13 23:18:25 - mmengine - INFO - Epoch(train) [1][3900/5005] lr: 1.0000e-04 eta: 6:56:10 time: 0.5656 data_time: 0.0119 memory: 1685 loss: 1.1930 2023/04/13 23:19:20 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/13 23:19:20 - mmengine - INFO - Epoch(train) [1][4000/5005] lr: 1.0000e-04 eta: 6:55:34 time: 0.6102 data_time: 0.0113 memory: 1685 loss: 1.3669 2023/04/13 23:20:15 - mmengine - INFO - Epoch(train) [1][4100/5005] lr: 1.0000e-04 eta: 6:54:42 time: 0.5654 data_time: 0.0108 memory: 1685 loss: 1.2413 2023/04/13 23:21:10 - mmengine - INFO - Epoch(train) [1][4200/5005] lr: 1.0000e-04 eta: 6:53:58 time: 0.6059 data_time: 0.0112 memory: 1685 loss: 1.2910 2023/04/13 23:22:01 - mmengine - INFO - Epoch(train) [1][4300/5005] lr: 1.0000e-04 eta: 6:52:36 time: 0.6370 data_time: 0.0114 memory: 1685 loss: 1.2639 2023/04/13 23:22:58 - mmengine - INFO - Epoch(train) [1][4400/5005] lr: 1.0000e-04 eta: 6:52:04 time: 0.5438 data_time: 0.0110 memory: 1685 loss: 1.4034 2023/04/13 23:23:54 - mmengine - INFO - Epoch(train) [1][4500/5005] lr: 1.0000e-04 eta: 6:51:36 time: 0.5661 data_time: 0.0114 memory: 1685 loss: 1.4908 2023/04/13 23:24:50 - mmengine - INFO - Epoch(train) [1][4600/5005] lr: 1.0000e-04 eta: 6:50:53 time: 0.6026 data_time: 0.0107 memory: 1685 loss: 1.4240 2023/04/13 23:25:46 - mmengine - INFO - Epoch(train) [1][4700/5005] lr: 1.0000e-04 eta: 6:50:16 time: 0.5945 data_time: 0.0110 memory: 1685 loss: 1.3294 2023/04/13 23:26:42 - mmengine - INFO - Epoch(train) [1][4800/5005] lr: 1.0000e-04 eta: 6:49:37 time: 0.5289 data_time: 0.0113 memory: 1685 loss: 1.4545 2023/04/13 23:27:37 - mmengine - INFO - Epoch(train) [1][4900/5005] lr: 1.0000e-04 eta: 6:48:51 time: 0.5703 data_time: 0.0107 memory: 1685 loss: 1.2514 2023/04/13 23:28:30 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/13 23:28:30 - mmengine - INFO - Epoch(train) [1][5000/5005] lr: 1.0000e-04 eta: 6:47:46 time: 0.2513 data_time: 0.0095 memory: 1685 loss: 1.2419 2023/04/13 23:28:31 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/13 23:28:32 - mmengine - INFO - Saving checkpoint at 1 epochs 2023/04/13 23:28:44 - mmengine - INFO - Epoch(val) [1][100/196] eta: 0:00:09 time: 0.0947 data_time: 0.0116 memory: 1685 2023/04/13 23:30:18 - mmengine - INFO - Epoch(val) [1][196/196] qat.accuracy/top1: 68.8900 qat.accuracy/top5: 88.7120data_time: 0.0238 time: 0.2124 2023/04/13 23:30:22 - mmengine - INFO - Epoch(val) [1][100/196] eta: 0:00:03 time: 0.0350 data_time: 0.0190 memory: 969 2023/04/13 23:31:19 - mmengine - INFO - Epoch(val) [1][196/196] original.accuracy/top1: 68.9700 original.accuracy/top5: 88.7080data_time: 0.0165 time: 0.0359 2023/04/13 23:32:11 - mmengine - INFO - Epoch(train) [2][ 100/5005] lr: 1.0000e-04 eta: 6:46:09 time: 0.4853 data_time: 0.0081 memory: 1685 loss: 1.4775 2023/04/13 23:33:00 - mmengine - INFO - Epoch(train) [2][ 200/5005] lr: 1.0000e-04 eta: 6:44:33 time: 0.5065 data_time: 0.0083 memory: 1685 loss: 1.3803 2023/04/13 23:33:49 - mmengine - INFO - Epoch(train) [2][ 300/5005] lr: 1.0000e-04 eta: 6:42:55 time: 0.5071 data_time: 0.0083 memory: 1685 loss: 1.3978 2023/04/13 23:34:38 - mmengine - INFO - Epoch(train) [2][ 400/5005] lr: 1.0000e-04 eta: 6:41:22 time: 0.4798 data_time: 0.0082 memory: 1685 loss: 1.3244 2023/04/13 23:35:29 - mmengine - INFO - Epoch(train) [2][ 500/5005] lr: 1.0000e-04 eta: 6:40:00 time: 0.5136 data_time: 0.0080 memory: 1685 loss: 1.3869 2023/04/13 23:36:19 - mmengine - INFO - Epoch(train) [2][ 600/5005] lr: 1.0000e-04 eta: 6:38:36 time: 0.5059 data_time: 0.0088 memory: 1685 loss: 1.2445 2023/04/13 23:37:09 - mmengine - INFO - Epoch(train) [2][ 700/5005] lr: 1.0000e-04 eta: 6:37:10 time: 0.4902 data_time: 0.0091 memory: 1685 loss: 1.3893 2023/04/13 23:38:00 - mmengine - INFO - Epoch(train) [2][ 800/5005] lr: 1.0000e-04 eta: 6:35:56 time: 0.4952 data_time: 0.0092 memory: 1685 loss: 1.2894 2023/04/13 23:38:52 - mmengine - INFO - Epoch(train) [2][ 900/5005] lr: 1.0000e-04 eta: 6:34:53 time: 0.5301 data_time: 0.0105 memory: 1685 loss: 1.1935 2023/04/13 23:39:42 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/13 23:39:45 - mmengine - INFO - Epoch(train) [2][1000/5005] lr: 1.0000e-04 eta: 6:33:53 time: 0.5280 data_time: 0.0112 memory: 1685 loss: 1.2602 2023/04/13 23:40:39 - mmengine - INFO - Epoch(train) [2][1100/5005] lr: 1.0000e-04 eta: 6:33:03 time: 0.4854 data_time: 0.0114 memory: 1685 loss: 1.2690 2023/04/13 23:41:33 - mmengine - INFO - Epoch(train) [2][1200/5005] lr: 1.0000e-04 eta: 6:32:13 time: 0.5289 data_time: 0.0136 memory: 1685 loss: 1.1564 2023/04/13 23:42:26 - mmengine - INFO - Epoch(train) [2][1300/5005] lr: 1.0000e-04 eta: 6:31:16 time: 0.5351 data_time: 0.0110 memory: 1685 loss: 1.2241 2023/04/13 23:43:21 - mmengine - INFO - Epoch(train) [2][1400/5005] lr: 1.0000e-04 eta: 6:30:27 time: 0.5411 data_time: 0.0104 memory: 1685 loss: 1.1579 2023/04/13 23:44:14 - mmengine - INFO - Epoch(train) [2][1500/5005] lr: 1.0000e-04 eta: 6:29:28 time: 0.5128 data_time: 0.0100 memory: 1685 loss: 1.2879 2023/04/13 23:45:07 - mmengine - INFO - Epoch(train) [2][1600/5005] lr: 1.0000e-04 eta: 6:28:34 time: 0.5379 data_time: 0.0092 memory: 1685 loss: 1.2200 2023/04/13 23:46:00 - mmengine - INFO - Epoch(train) [2][1700/5005] lr: 1.0000e-04 eta: 6:27:35 time: 0.5636 data_time: 0.0118 memory: 1685 loss: 1.1917 2023/04/13 23:46:52 - mmengine - INFO - Epoch(train) [2][1800/5005] lr: 1.0000e-04 eta: 6:26:33 time: 0.4972 data_time: 0.0115 memory: 1685 loss: 1.2901 2023/04/13 23:47:47 - mmengine - INFO - Epoch(train) [2][1900/5005] lr: 1.0000e-04 eta: 6:25:43 time: 0.5206 data_time: 0.0096 memory: 1685 loss: 1.3610 2023/04/13 23:48:38 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/13 23:48:40 - mmengine - INFO - Epoch(train) [2][2000/5005] lr: 1.0000e-04 eta: 6:24:47 time: 0.5132 data_time: 0.0095 memory: 1685 loss: 1.4232 2023/04/13 23:49:34 - mmengine - INFO - Epoch(train) [2][2100/5005] lr: 1.0000e-04 eta: 6:23:58 time: 0.5347 data_time: 0.0116 memory: 1685 loss: 1.3291 2023/04/13 23:50:29 - mmengine - INFO - Epoch(train) [2][2200/5005] lr: 1.0000e-04 eta: 6:23:08 time: 0.5311 data_time: 0.0084 memory: 1685 loss: 1.2441 2023/04/13 23:51:24 - mmengine - INFO - Epoch(train) [2][2300/5005] lr: 1.0000e-04 eta: 6:22:23 time: 0.5695 data_time: 0.0123 memory: 1685 loss: 1.2997 2023/04/13 23:52:18 - mmengine - INFO - Epoch(train) [2][2400/5005] lr: 1.0000e-04 eta: 6:21:30 time: 0.5592 data_time: 0.0095 memory: 1685 loss: 1.3543 2023/04/13 23:53:11 - mmengine - INFO - Epoch(train) [2][2500/5005] lr: 1.0000e-04 eta: 6:20:37 time: 0.5199 data_time: 0.0102 memory: 1685 loss: 1.4072 2023/04/13 23:54:03 - mmengine - INFO - Epoch(train) [2][2600/5005] lr: 1.0000e-04 eta: 6:19:34 time: 0.5146 data_time: 0.0084 memory: 1685 loss: 1.2871 2023/04/13 23:54:56 - mmengine - INFO - Epoch(train) [2][2700/5005] lr: 1.0000e-04 eta: 6:18:33 time: 0.4968 data_time: 0.0092 memory: 1685 loss: 1.4370 2023/04/13 23:55:45 - mmengine - INFO - Epoch(train) [2][2800/5005] lr: 1.0000e-04 eta: 6:17:15 time: 0.4973 data_time: 0.0082 memory: 1685 loss: 1.2449 2023/04/13 23:56:34 - mmengine - INFO - Epoch(train) [2][2900/5005] lr: 1.0000e-04 eta: 6:15:57 time: 0.4763 data_time: 0.0085 memory: 1685 loss: 1.2787 2023/04/13 23:57:21 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/13 23:57:23 - mmengine - INFO - Epoch(train) [2][3000/5005] lr: 1.0000e-04 eta: 6:14:43 time: 0.4505 data_time: 0.0089 memory: 1685 loss: 1.2416 2023/04/13 23:58:14 - mmengine - INFO - Epoch(train) [2][3100/5005] lr: 1.0000e-04 eta: 6:13:32 time: 0.4878 data_time: 0.0081 memory: 1685 loss: 1.2438 2023/04/13 23:59:05 - mmengine - INFO - Epoch(train) [2][3200/5005] lr: 1.0000e-04 eta: 6:12:27 time: 0.4862 data_time: 0.0081 memory: 1685 loss: 1.4459 2023/04/13 23:59:53 - mmengine - INFO - Epoch(train) [2][3300/5005] lr: 1.0000e-04 eta: 6:11:10 time: 0.4936 data_time: 0.0081 memory: 1685 loss: 1.2307 2023/04/14 00:00:42 - mmengine - INFO - Epoch(train) [2][3400/5005] lr: 1.0000e-04 eta: 6:09:55 time: 0.5136 data_time: 0.0079 memory: 1685 loss: 1.4655 2023/04/14 00:01:27 - mmengine - INFO - Epoch(train) [2][3500/5005] lr: 1.0000e-04 eta: 6:08:17 time: 0.4244 data_time: 0.0080 memory: 1685 loss: 1.3709 2023/04/14 00:02:41 - mmengine - INFO - Epoch(train) [2][3600/5005] lr: 1.0000e-04 eta: 6:09:07 time: 1.2472 data_time: 0.0079 memory: 1685 loss: 1.4449 2023/04/14 00:03:55 - mmengine - INFO - Epoch(train) [2][3700/5005] lr: 1.0000e-04 eta: 6:09:50 time: 0.5307 data_time: 0.0079 memory: 1685 loss: 1.2903 2023/04/14 00:04:44 - mmengine - INFO - Epoch(train) [2][3800/5005] lr: 1.0000e-04 eta: 6:08:37 time: 0.4929 data_time: 0.0083 memory: 1685 loss: 1.4934 2023/04/14 00:05:34 - mmengine - INFO - Epoch(train) [2][3900/5005] lr: 1.0000e-04 eta: 6:07:26 time: 0.4920 data_time: 0.0079 memory: 1685 loss: 1.4190 2023/04/14 00:06:22 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 00:06:24 - mmengine - INFO - Epoch(train) [2][4000/5005] lr: 1.0000e-04 eta: 6:06:14 time: 0.4702 data_time: 0.0081 memory: 1685 loss: 1.5032 2023/04/14 00:07:14 - mmengine - INFO - Epoch(train) [2][4100/5005] lr: 1.0000e-04 eta: 6:05:06 time: 0.4964 data_time: 0.0092 memory: 1685 loss: 1.4647 2023/04/14 00:08:04 - mmengine - INFO - Epoch(train) [2][4200/5005] lr: 1.0000e-04 eta: 6:03:58 time: 0.4932 data_time: 0.0089 memory: 1685 loss: 1.4049 2023/04/14 00:08:55 - mmengine - INFO - Epoch(train) [2][4300/5005] lr: 1.0000e-04 eta: 6:02:53 time: 0.4897 data_time: 0.0092 memory: 1685 loss: 1.3874 2023/04/14 00:09:46 - mmengine - INFO - Epoch(train) [2][4400/5005] lr: 1.0000e-04 eta: 6:01:48 time: 0.5563 data_time: 0.0092 memory: 1685 loss: 1.3308 2023/04/14 00:10:40 - mmengine - INFO - Epoch(train) [2][4500/5005] lr: 1.0000e-04 eta: 6:00:55 time: 0.5144 data_time: 0.0121 memory: 1685 loss: 1.2509 2023/04/14 00:11:35 - mmengine - INFO - Epoch(train) [2][4600/5005] lr: 1.0000e-04 eta: 6:00:09 time: 0.5522 data_time: 0.0106 memory: 1685 loss: 1.1196 2023/04/14 00:12:30 - mmengine - INFO - Epoch(train) [2][4700/5005] lr: 1.0000e-04 eta: 5:59:21 time: 0.5561 data_time: 0.0105 memory: 1685 loss: 1.2536 2023/04/14 00:13:24 - mmengine - INFO - Epoch(train) [2][4800/5005] lr: 1.0000e-04 eta: 5:58:33 time: 0.5575 data_time: 0.0103 memory: 1685 loss: 1.2955 2023/04/14 00:14:18 - mmengine - INFO - Epoch(train) [2][4900/5005] lr: 1.0000e-04 eta: 5:57:41 time: 0.4814 data_time: 0.0093 memory: 1685 loss: 1.5378 2023/04/14 00:15:09 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 00:15:12 - mmengine - INFO - Epoch(train) [2][5000/5005] lr: 1.0000e-04 eta: 5:56:48 time: 0.5516 data_time: 0.0215 memory: 1685 loss: 1.2999 2023/04/14 00:15:14 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 00:15:16 - mmengine - INFO - Saving checkpoint at 2 epochs 2023/04/14 00:15:48 - mmengine - INFO - Epoch(val) [2][100/196] eta: 0:00:29 time: 0.2909 data_time: 0.0108 memory: 1685 2023/04/14 00:17:13 - mmengine - INFO - Epoch(val) [2][196/196] qat.accuracy/top1: 69.0820 qat.accuracy/top5: 88.9520data_time: 0.0103 time: 0.5722 2023/04/14 00:17:17 - mmengine - INFO - Epoch(val) [2][100/196] eta: 0:00:04 time: 0.0476 data_time: 0.0292 memory: 969 2023/04/14 00:18:14 - mmengine - INFO - Epoch(val) [2][196/196] original.accuracy/top1: 69.0580 original.accuracy/top5: 88.8420data_time: 0.0233 time: 0.0424 2023/04/14 00:19:06 - mmengine - INFO - Epoch(train) [3][ 100/5005] lr: 1.0000e-04 eta: 5:55:43 time: 0.5036 data_time: 0.0088 memory: 1685 loss: 1.1535 2023/04/14 00:19:56 - mmengine - INFO - Epoch(train) [3][ 200/5005] lr: 1.0000e-04 eta: 5:54:38 time: 0.5005 data_time: 0.0088 memory: 1685 loss: 1.0970 2023/04/14 00:20:48 - mmengine - INFO - Epoch(train) [3][ 300/5005] lr: 1.0000e-04 eta: 5:53:39 time: 0.5039 data_time: 0.0082 memory: 1685 loss: 1.2676 2023/04/14 00:21:38 - mmengine - INFO - Epoch(train) [3][ 400/5005] lr: 1.0000e-04 eta: 5:52:33 time: 0.5140 data_time: 0.0095 memory: 1685 loss: 1.1114 2023/04/14 00:22:32 - mmengine - INFO - Epoch(train) [3][ 500/5005] lr: 1.0000e-04 eta: 5:51:40 time: 0.5587 data_time: 0.0081 memory: 1685 loss: 1.2077 2023/04/14 00:23:23 - mmengine - INFO - Epoch(train) [3][ 600/5005] lr: 1.0000e-04 eta: 5:50:39 time: 0.5498 data_time: 0.0087 memory: 1685 loss: 1.1765 2023/04/14 00:24:15 - mmengine - INFO - Epoch(train) [3][ 700/5005] lr: 1.0000e-04 eta: 5:49:39 time: 0.5249 data_time: 0.0087 memory: 1685 loss: 1.4642 2023/04/14 00:25:06 - mmengine - INFO - Epoch(train) [3][ 800/5005] lr: 1.0000e-04 eta: 5:48:37 time: 0.5277 data_time: 0.0088 memory: 1685 loss: 1.5966 2023/04/14 00:25:55 - mmengine - INFO - Epoch(train) [3][ 900/5005] lr: 1.0000e-04 eta: 5:47:31 time: 0.4537 data_time: 0.0095 memory: 1685 loss: 1.2710 2023/04/14 00:26:44 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 00:26:49 - mmengine - INFO - Epoch(train) [3][1000/5005] lr: 1.0000e-04 eta: 5:46:40 time: 0.5074 data_time: 0.0085 memory: 1685 loss: 1.4888 2023/04/14 00:27:39 - mmengine - INFO - Epoch(train) [3][1100/5005] lr: 1.0000e-04 eta: 5:45:35 time: 0.5374 data_time: 0.0085 memory: 1685 loss: 1.2199 2023/04/14 00:28:31 - mmengine - INFO - Epoch(train) [3][1200/5005] lr: 1.0000e-04 eta: 5:44:35 time: 0.4888 data_time: 0.0080 memory: 1685 loss: 1.2025 2023/04/14 00:29:20 - mmengine - INFO - Epoch(train) [3][1300/5005] lr: 1.0000e-04 eta: 5:43:30 time: 0.4871 data_time: 0.0079 memory: 1685 loss: 1.2553 2023/04/14 00:30:11 - mmengine - INFO - Epoch(train) [3][1400/5005] lr: 1.0000e-04 eta: 5:42:29 time: 0.5004 data_time: 0.0083 memory: 1685 loss: 1.3244 2023/04/14 00:31:01 - mmengine - INFO - Epoch(train) [3][1500/5005] lr: 1.0000e-04 eta: 5:41:25 time: 0.4734 data_time: 0.0083 memory: 1685 loss: 1.5175 2023/04/14 00:31:51 - mmengine - INFO - Epoch(train) [3][1600/5005] lr: 1.0000e-04 eta: 5:40:21 time: 0.4796 data_time: 0.0080 memory: 1685 loss: 1.2027 2023/04/14 00:32:42 - mmengine - INFO - Epoch(train) [3][1700/5005] lr: 1.0000e-04 eta: 5:39:20 time: 0.4964 data_time: 0.0079 memory: 1685 loss: 1.2320 2023/04/14 00:33:32 - mmengine - INFO - Epoch(train) [3][1800/5005] lr: 1.0000e-04 eta: 5:38:17 time: 0.5086 data_time: 0.0085 memory: 1685 loss: 1.4208 2023/04/14 00:34:22 - mmengine - INFO - Epoch(train) [3][1900/5005] lr: 1.0000e-04 eta: 5:37:15 time: 0.5042 data_time: 0.0082 memory: 1685 loss: 1.3530 2023/04/14 00:35:08 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 00:35:13 - mmengine - INFO - Epoch(train) [3][2000/5005] lr: 1.0000e-04 eta: 5:36:15 time: 0.5230 data_time: 0.0082 memory: 1685 loss: 1.3569 2023/04/14 00:36:04 - mmengine - INFO - Epoch(train) [3][2100/5005] lr: 1.0000e-04 eta: 5:35:14 time: 0.4930 data_time: 0.0080 memory: 1685 loss: 1.4308 2023/04/14 00:36:55 - mmengine - INFO - Epoch(train) [3][2200/5005] lr: 1.0000e-04 eta: 5:34:15 time: 0.4734 data_time: 0.0082 memory: 1685 loss: 1.2295 2023/04/14 00:37:45 - mmengine - INFO - Epoch(train) [3][2300/5005] lr: 1.0000e-04 eta: 5:33:14 time: 0.5276 data_time: 0.0083 memory: 1685 loss: 1.3605 2023/04/14 00:38:36 - mmengine - INFO - Epoch(train) [3][2400/5005] lr: 1.0000e-04 eta: 5:32:15 time: 0.5019 data_time: 0.0084 memory: 1685 loss: 1.2141 2023/04/14 00:39:28 - mmengine - INFO - Epoch(train) [3][2500/5005] lr: 1.0000e-04 eta: 5:31:18 time: 0.4889 data_time: 0.0097 memory: 1685 loss: 1.5570 2023/04/14 00:40:20 - mmengine - INFO - Epoch(train) [3][2600/5005] lr: 1.0000e-04 eta: 5:30:22 time: 0.5179 data_time: 0.0095 memory: 1685 loss: 1.3223 2023/04/14 00:41:10 - mmengine - INFO - Epoch(train) [3][2700/5005] lr: 1.0000e-04 eta: 5:29:22 time: 0.5112 data_time: 0.0085 memory: 1685 loss: 1.2127 2023/04/14 00:42:01 - mmengine - INFO - Epoch(train) [3][2800/5005] lr: 1.0000e-04 eta: 5:28:24 time: 0.5196 data_time: 0.0084 memory: 1685 loss: 1.4454 2023/04/14 00:42:53 - mmengine - INFO - Epoch(train) [3][2900/5005] lr: 1.0000e-04 eta: 5:27:26 time: 0.5584 data_time: 0.0077 memory: 1685 loss: 1.3207 2023/04/14 00:43:38 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 00:43:43 - mmengine - INFO - Epoch(train) [3][3000/5005] lr: 1.0000e-04 eta: 5:26:26 time: 0.5027 data_time: 0.0083 memory: 1685 loss: 1.2444 2023/04/14 00:44:34 - mmengine - INFO - Epoch(train) [3][3100/5005] lr: 1.0000e-04 eta: 5:25:28 time: 0.5418 data_time: 0.0082 memory: 1685 loss: 1.2760 2023/04/14 00:45:25 - mmengine - INFO - Epoch(train) [3][3200/5005] lr: 1.0000e-04 eta: 5:24:30 time: 0.5341 data_time: 0.0078 memory: 1685 loss: 1.2564 2023/04/14 00:46:16 - mmengine - INFO - Epoch(train) [3][3300/5005] lr: 1.0000e-04 eta: 5:23:33 time: 0.5380 data_time: 0.0077 memory: 1685 loss: 1.2473 2023/04/14 00:47:09 - mmengine - INFO - Epoch(train) [3][3400/5005] lr: 1.0000e-04 eta: 5:22:40 time: 0.5138 data_time: 0.0111 memory: 1685 loss: 1.2008 2023/04/14 00:48:05 - mmengine - INFO - Epoch(train) [3][3500/5005] lr: 1.0000e-04 eta: 5:21:54 time: 0.5423 data_time: 0.0122 memory: 1685 loss: 1.2328 2023/04/14 00:49:00 - mmengine - INFO - Epoch(train) [3][3600/5005] lr: 1.0000e-04 eta: 5:21:08 time: 0.5636 data_time: 0.0108 memory: 1685 loss: 1.0889 2023/04/14 00:49:56 - mmengine - INFO - Epoch(train) [3][3700/5005] lr: 1.0000e-04 eta: 5:20:22 time: 0.5226 data_time: 0.0097 memory: 1685 loss: 1.3114 2023/04/14 00:50:49 - mmengine - INFO - Epoch(train) [3][3800/5005] lr: 1.0000e-04 eta: 5:19:32 time: 0.5363 data_time: 0.0103 memory: 1685 loss: 1.1974 2023/04/14 00:51:46 - mmengine - INFO - Epoch(train) [3][3900/5005] lr: 1.0000e-04 eta: 5:18:49 time: 0.5589 data_time: 0.0110 memory: 1685 loss: 1.3176 2023/04/14 00:52:36 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 00:52:42 - mmengine - INFO - Epoch(train) [3][4000/5005] lr: 1.0000e-04 eta: 5:18:04 time: 0.5993 data_time: 0.0110 memory: 1685 loss: 1.3264 2023/04/14 00:53:36 - mmengine - INFO - Epoch(train) [3][4100/5005] lr: 1.0000e-04 eta: 5:17:13 time: 0.5384 data_time: 0.0110 memory: 1685 loss: 1.4351 2023/04/14 00:54:27 - mmengine - INFO - Epoch(train) [3][4200/5005] lr: 1.0000e-04 eta: 5:16:16 time: 0.4665 data_time: 0.0101 memory: 1685 loss: 1.2768 2023/04/14 00:55:16 - mmengine - INFO - Epoch(train) [3][4300/5005] lr: 1.0000e-04 eta: 5:15:12 time: 0.2850 data_time: 0.0116 memory: 1685 loss: 1.3867 2023/04/14 00:56:43 - mmengine - INFO - Epoch(train) [3][4400/5005] lr: 1.0000e-04 eta: 5:15:43 time: 1.3206 data_time: 0.0085 memory: 1685 loss: 1.4838 2023/04/14 00:57:48 - mmengine - INFO - Epoch(train) [3][4500/5005] lr: 1.0000e-04 eta: 5:15:21 time: 0.5033 data_time: 0.0092 memory: 1685 loss: 1.3163 2023/04/14 00:58:38 - mmengine - INFO - Epoch(train) [3][4600/5005] lr: 1.0000e-04 eta: 5:14:19 time: 0.4902 data_time: 0.0085 memory: 1685 loss: 1.2108 2023/04/14 00:59:28 - mmengine - INFO - Epoch(train) [3][4700/5005] lr: 1.0000e-04 eta: 5:13:18 time: 0.5084 data_time: 0.0084 memory: 1685 loss: 1.2135 2023/04/14 01:00:18 - mmengine - INFO - Epoch(train) [3][4800/5005] lr: 1.0000e-04 eta: 5:12:17 time: 0.5168 data_time: 0.0080 memory: 1685 loss: 1.3309 2023/04/14 01:01:08 - mmengine - INFO - Epoch(train) [3][4900/5005] lr: 1.0000e-04 eta: 5:11:17 time: 0.5394 data_time: 0.0083 memory: 1685 loss: 1.2680 2023/04/14 01:01:52 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 01:01:57 - mmengine - INFO - Epoch(train) [3][5000/5005] lr: 1.0000e-04 eta: 5:10:14 time: 0.5403 data_time: 0.0096 memory: 1685 loss: 1.2584 2023/04/14 01:02:00 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 01:02:01 - mmengine - INFO - Saving checkpoint at 3 epochs 2023/04/14 01:02:31 - mmengine - INFO - Epoch(val) [3][100/196] eta: 0:00:27 time: 0.2683 data_time: 0.0082 memory: 1685 2023/04/14 01:03:56 - mmengine - INFO - Epoch(val) [3][196/196] qat.accuracy/top1: 69.3580 qat.accuracy/top5: 88.9700data_time: 0.0076 time: 0.5736 2023/04/14 01:04:01 - mmengine - INFO - Epoch(val) [3][100/196] eta: 0:00:04 time: 0.0407 data_time: 0.0223 memory: 969 2023/04/14 01:04:57 - mmengine - INFO - Epoch(val) [3][196/196] original.accuracy/top1: 69.0940 original.accuracy/top5: 88.8420data_time: 0.0254 time: 0.0453 2023/04/14 01:05:49 - mmengine - INFO - Epoch(train) [4][ 100/5005] lr: 1.0000e-04 eta: 5:09:14 time: 0.4871 data_time: 0.0093 memory: 1685 loss: 1.3311 2023/04/14 01:06:40 - mmengine - INFO - Epoch(train) [4][ 200/5005] lr: 1.0000e-04 eta: 5:08:14 time: 0.5104 data_time: 0.0081 memory: 1685 loss: 1.2948 2023/04/14 01:07:30 - mmengine - INFO - Epoch(train) [4][ 300/5005] lr: 1.0000e-04 eta: 5:07:15 time: 0.4842 data_time: 0.0076 memory: 1685 loss: 1.3639 2023/04/14 01:08:21 - mmengine - INFO - Epoch(train) [4][ 400/5005] lr: 1.0000e-04 eta: 5:06:16 time: 0.5247 data_time: 0.0078 memory: 1685 loss: 1.3939 2023/04/14 01:09:11 - mmengine - INFO - Epoch(train) [4][ 500/5005] lr: 1.0000e-04 eta: 5:05:17 time: 0.5016 data_time: 0.0076 memory: 1685 loss: 1.2940 2023/04/14 01:10:01 - mmengine - INFO - Epoch(train) [4][ 600/5005] lr: 1.0000e-04 eta: 5:04:17 time: 0.4913 data_time: 0.0086 memory: 1685 loss: 1.2984 2023/04/14 01:10:53 - mmengine - INFO - Epoch(train) [4][ 700/5005] lr: 1.0000e-04 eta: 5:03:21 time: 0.5175 data_time: 0.0082 memory: 1685 loss: 1.4286 2023/04/14 01:11:42 - mmengine - INFO - Epoch(train) [4][ 800/5005] lr: 1.0000e-04 eta: 5:02:21 time: 0.5154 data_time: 0.0080 memory: 1685 loss: 1.3337 2023/04/14 01:12:33 - mmengine - INFO - Epoch(train) [4][ 900/5005] lr: 1.0000e-04 eta: 5:01:24 time: 0.4916 data_time: 0.0095 memory: 1685 loss: 1.1680 2023/04/14 01:13:16 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 01:13:23 - mmengine - INFO - Epoch(train) [4][1000/5005] lr: 1.0000e-04 eta: 5:00:25 time: 0.4744 data_time: 0.0077 memory: 1685 loss: 1.3383 2023/04/14 01:14:14 - mmengine - INFO - Epoch(train) [4][1100/5005] lr: 1.0000e-04 eta: 4:59:27 time: 0.5158 data_time: 0.0078 memory: 1685 loss: 1.4216 2023/04/14 01:15:05 - mmengine - INFO - Epoch(train) [4][1200/5005] lr: 1.0000e-04 eta: 4:58:30 time: 0.4884 data_time: 0.0079 memory: 1685 loss: 1.2761 2023/04/14 01:15:56 - mmengine - INFO - Epoch(train) [4][1300/5005] lr: 1.0000e-04 eta: 4:57:33 time: 0.4956 data_time: 0.0077 memory: 1685 loss: 1.3115 2023/04/14 01:16:47 - mmengine - INFO - Epoch(train) [4][1400/5005] lr: 1.0000e-04 eta: 4:56:35 time: 0.5355 data_time: 0.0076 memory: 1685 loss: 1.2009 2023/04/14 01:17:37 - mmengine - INFO - Epoch(train) [4][1500/5005] lr: 1.0000e-04 eta: 4:55:36 time: 0.4917 data_time: 0.0079 memory: 1685 loss: 1.2842 2023/04/14 01:18:26 - mmengine - INFO - Epoch(train) [4][1600/5005] lr: 1.0000e-04 eta: 4:54:37 time: 0.5031 data_time: 0.0077 memory: 1685 loss: 1.3136 2023/04/14 01:19:17 - mmengine - INFO - Epoch(train) [4][1700/5005] lr: 1.0000e-04 eta: 4:53:40 time: 0.5021 data_time: 0.0075 memory: 1685 loss: 1.2183 2023/04/14 01:20:11 - mmengine - INFO - Epoch(train) [4][1800/5005] lr: 1.0000e-04 eta: 4:52:49 time: 0.5070 data_time: 0.0078 memory: 1685 loss: 1.3754 2023/04/14 01:21:01 - mmengine - INFO - Epoch(train) [4][1900/5005] lr: 1.0000e-04 eta: 4:51:51 time: 0.4812 data_time: 0.0077 memory: 1685 loss: 1.2106 2023/04/14 01:21:45 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 01:21:53 - mmengine - INFO - Epoch(train) [4][2000/5005] lr: 1.0000e-04 eta: 4:50:57 time: 0.5719 data_time: 0.0076 memory: 1685 loss: 1.3197 2023/04/14 01:22:43 - mmengine - INFO - Epoch(train) [4][2100/5005] lr: 1.0000e-04 eta: 4:49:58 time: 0.5257 data_time: 0.0075 memory: 1685 loss: 1.4039 2023/04/14 01:23:34 - mmengine - INFO - Epoch(train) [4][2200/5005] lr: 1.0000e-04 eta: 4:49:01 time: 0.4819 data_time: 0.0075 memory: 1685 loss: 1.1897 2023/04/14 01:24:24 - mmengine - INFO - Epoch(train) [4][2300/5005] lr: 1.0000e-04 eta: 4:48:04 time: 0.4750 data_time: 0.0075 memory: 1685 loss: 1.1536 2023/04/14 01:25:15 - mmengine - INFO - Epoch(train) [4][2400/5005] lr: 1.0000e-04 eta: 4:47:06 time: 0.4948 data_time: 0.0077 memory: 1685 loss: 1.4205 2023/04/14 01:26:05 - mmengine - INFO - Epoch(train) [4][2500/5005] lr: 1.0000e-04 eta: 4:46:08 time: 0.4774 data_time: 0.0076 memory: 1685 loss: 1.0800 2023/04/14 01:26:55 - mmengine - INFO - Epoch(train) [4][2600/5005] lr: 1.0000e-04 eta: 4:45:12 time: 0.4873 data_time: 0.0076 memory: 1685 loss: 1.3761 2023/04/14 01:27:46 - mmengine - INFO - Epoch(train) [4][2700/5005] lr: 1.0000e-04 eta: 4:44:15 time: 0.5158 data_time: 0.0076 memory: 1685 loss: 1.1852 2023/04/14 01:28:36 - mmengine - INFO - Epoch(train) [4][2800/5005] lr: 1.0000e-04 eta: 4:43:18 time: 0.5140 data_time: 0.0084 memory: 1685 loss: 1.2589 2023/04/14 01:29:28 - mmengine - INFO - Epoch(train) [4][2900/5005] lr: 1.0000e-04 eta: 4:42:23 time: 0.5298 data_time: 0.0076 memory: 1685 loss: 1.3269 2023/04/14 01:30:10 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 01:30:18 - mmengine - INFO - Epoch(train) [4][3000/5005] lr: 1.0000e-04 eta: 4:41:26 time: 0.5113 data_time: 0.0075 memory: 1685 loss: 1.2181 2023/04/14 01:31:09 - mmengine - INFO - Epoch(train) [4][3100/5005] lr: 1.0000e-04 eta: 4:40:31 time: 0.5336 data_time: 0.0071 memory: 1685 loss: 1.2238 2023/04/14 01:32:00 - mmengine - INFO - Epoch(train) [4][3200/5005] lr: 1.0000e-04 eta: 4:39:34 time: 0.5511 data_time: 0.0079 memory: 1685 loss: 1.2643 2023/04/14 01:32:51 - mmengine - INFO - Epoch(train) [4][3300/5005] lr: 1.0000e-04 eta: 4:38:39 time: 0.5894 data_time: 0.0076 memory: 1685 loss: 1.4197 2023/04/14 01:33:42 - mmengine - INFO - Epoch(train) [4][3400/5005] lr: 1.0000e-04 eta: 4:37:43 time: 0.5210 data_time: 0.0076 memory: 1685 loss: 1.3034 2023/04/14 01:34:32 - mmengine - INFO - Epoch(train) [4][3500/5005] lr: 1.0000e-04 eta: 4:36:47 time: 0.4757 data_time: 0.0076 memory: 1685 loss: 1.4141 2023/04/14 01:35:24 - mmengine - INFO - Epoch(train) [4][3600/5005] lr: 1.0000e-04 eta: 4:35:53 time: 0.5011 data_time: 0.0075 memory: 1685 loss: 1.1801 2023/04/14 01:36:14 - mmengine - INFO - Epoch(train) [4][3700/5005] lr: 1.0000e-04 eta: 4:34:56 time: 0.4780 data_time: 0.0070 memory: 1685 loss: 1.3278 2023/04/14 01:37:05 - mmengine - INFO - Epoch(train) [4][3800/5005] lr: 1.0000e-04 eta: 4:33:59 time: 0.4813 data_time: 0.0081 memory: 1685 loss: 1.3021 2023/04/14 01:37:55 - mmengine - INFO - Epoch(train) [4][3900/5005] lr: 1.0000e-04 eta: 4:33:03 time: 0.5128 data_time: 0.0078 memory: 1685 loss: 1.1602 2023/04/14 01:38:38 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 01:38:45 - mmengine - INFO - Epoch(train) [4][4000/5005] lr: 1.0000e-04 eta: 4:32:06 time: 0.5033 data_time: 0.0077 memory: 1685 loss: 1.2275 2023/04/14 01:39:37 - mmengine - INFO - Epoch(train) [4][4100/5005] lr: 1.0000e-04 eta: 4:31:12 time: 0.5108 data_time: 0.0078 memory: 1685 loss: 1.1892 2023/04/14 01:40:27 - mmengine - INFO - Epoch(train) [4][4200/5005] lr: 1.0000e-04 eta: 4:30:16 time: 0.5374 data_time: 0.0077 memory: 1685 loss: 1.1676 2023/04/14 01:41:17 - mmengine - INFO - Epoch(train) [4][4300/5005] lr: 1.0000e-04 eta: 4:29:19 time: 0.4781 data_time: 0.0076 memory: 1685 loss: 1.2420 2023/04/14 01:42:09 - mmengine - INFO - Epoch(train) [4][4400/5005] lr: 1.0000e-04 eta: 4:28:25 time: 0.5045 data_time: 0.0189 memory: 1685 loss: 1.5300 2023/04/14 01:43:00 - mmengine - INFO - Epoch(train) [4][4500/5005] lr: 1.0000e-04 eta: 4:27:30 time: 0.5006 data_time: 0.0077 memory: 1685 loss: 1.4469 2023/04/14 01:43:50 - mmengine - INFO - Epoch(train) [4][4600/5005] lr: 1.0000e-04 eta: 4:26:33 time: 0.5014 data_time: 0.0084 memory: 1685 loss: 1.3074 2023/04/14 01:44:39 - mmengine - INFO - Epoch(train) [4][4700/5005] lr: 1.0000e-04 eta: 4:25:36 time: 0.5036 data_time: 0.0078 memory: 1685 loss: 1.4360 2023/04/14 01:45:29 - mmengine - INFO - Epoch(train) [4][4800/5005] lr: 1.0000e-04 eta: 4:24:39 time: 0.5328 data_time: 0.0086 memory: 1685 loss: 1.3378 2023/04/14 01:46:18 - mmengine - INFO - Epoch(train) [4][4900/5005] lr: 1.0000e-04 eta: 4:23:41 time: 0.5164 data_time: 0.0080 memory: 1685 loss: 1.0631 2023/04/14 01:46:55 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 01:47:02 - mmengine - INFO - Epoch(train) [4][5000/5005] lr: 1.0000e-04 eta: 4:22:35 time: 0.4366 data_time: 0.0082 memory: 1685 loss: 1.3463 2023/04/14 01:47:04 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 01:47:04 - mmengine - INFO - Saving checkpoint at 4 epochs 2023/04/14 01:47:29 - mmengine - INFO - Epoch(val) [4][100/196] eta: 0:00:22 time: 0.2497 data_time: 0.0077 memory: 1685 2023/04/14 01:49:24 - mmengine - INFO - Epoch(val) [4][196/196] qat.accuracy/top1: 69.2420 qat.accuracy/top5: 88.9120data_time: 0.0072 time: 0.1995 2023/04/14 01:49:28 - mmengine - INFO - Epoch(val) [4][100/196] eta: 0:00:03 time: 0.0311 data_time: 0.0149 memory: 969 2023/04/14 01:50:25 - mmengine - INFO - Epoch(val) [4][196/196] original.accuracy/top1: 69.1540 original.accuracy/top5: 88.8480data_time: 0.0319 time: 0.0443 2023/04/14 01:50:50 - mmengine - INFO - Epoch(train) [5][ 100/5005] lr: 1.0000e-04 eta: 4:20:59 time: 0.2380 data_time: 0.0085 memory: 1685 loss: 1.2934 2023/04/14 01:51:15 - mmengine - INFO - Epoch(train) [5][ 200/5005] lr: 1.0000e-04 eta: 4:19:27 time: 0.2353 data_time: 0.0091 memory: 1685 loss: 1.2166 2023/04/14 01:51:39 - mmengine - INFO - Epoch(train) [5][ 300/5005] lr: 1.0000e-04 eta: 4:17:53 time: 0.2381 data_time: 0.0089 memory: 1685 loss: 1.2873 2023/04/14 01:52:04 - mmengine - INFO - Epoch(train) [5][ 400/5005] lr: 1.0000e-04 eta: 4:16:21 time: 0.2423 data_time: 0.0095 memory: 1685 loss: 1.3542 2023/04/14 01:52:28 - mmengine - INFO - Epoch(train) [5][ 500/5005] lr: 1.0000e-04 eta: 4:14:50 time: 0.2310 data_time: 0.0089 memory: 1685 loss: 1.1442 2023/04/14 01:52:53 - mmengine - INFO - Epoch(train) [5][ 600/5005] lr: 1.0000e-04 eta: 4:13:20 time: 0.2975 data_time: 0.0083 memory: 1685 loss: 1.2142 2023/04/14 01:53:17 - mmengine - INFO - Epoch(train) [5][ 700/5005] lr: 1.0000e-04 eta: 4:11:48 time: 0.2327 data_time: 0.0085 memory: 1685 loss: 1.1566 2023/04/14 01:53:41 - mmengine - INFO - Epoch(train) [5][ 800/5005] lr: 1.0000e-04 eta: 4:10:18 time: 0.2349 data_time: 0.0083 memory: 1685 loss: 1.3897 2023/04/14 01:54:05 - mmengine - INFO - Epoch(train) [5][ 900/5005] lr: 1.0000e-04 eta: 4:08:49 time: 0.3432 data_time: 0.0236 memory: 1685 loss: 1.4631 2023/04/14 01:54:24 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 01:54:29 - mmengine - INFO - Epoch(train) [5][1000/5005] lr: 1.0000e-04 eta: 4:07:21 time: 0.3029 data_time: 0.0096 memory: 1685 loss: 1.2536 2023/04/14 01:54:54 - mmengine - INFO - Epoch(train) [5][1100/5005] lr: 1.0000e-04 eta: 4:05:53 time: 0.2438 data_time: 0.0088 memory: 1685 loss: 1.4519 2023/04/14 01:55:20 - mmengine - INFO - Epoch(train) [5][1200/5005] lr: 1.0000e-04 eta: 4:04:28 time: 0.2472 data_time: 0.0082 memory: 1685 loss: 0.9584 2023/04/14 01:55:43 - mmengine - INFO - Epoch(train) [5][1300/5005] lr: 1.0000e-04 eta: 4:03:00 time: 0.2400 data_time: 0.0081 memory: 1685 loss: 1.2102 2023/04/14 01:56:07 - mmengine - INFO - Epoch(train) [5][1400/5005] lr: 1.0000e-04 eta: 4:01:34 time: 0.2392 data_time: 0.0086 memory: 1685 loss: 1.2416 2023/04/14 01:56:31 - mmengine - INFO - Epoch(train) [5][1500/5005] lr: 1.0000e-04 eta: 4:00:07 time: 0.2338 data_time: 0.0085 memory: 1685 loss: 1.4380 2023/04/14 01:56:55 - mmengine - INFO - Epoch(train) [5][1600/5005] lr: 1.0000e-04 eta: 3:58:42 time: 0.2350 data_time: 0.0086 memory: 1685 loss: 1.3665 2023/04/14 01:57:19 - mmengine - INFO - Epoch(train) [5][1700/5005] lr: 1.0000e-04 eta: 3:57:17 time: 0.2866 data_time: 0.0082 memory: 1685 loss: 1.2418 2023/04/14 01:57:43 - mmengine - INFO - Epoch(train) [5][1800/5005] lr: 1.0000e-04 eta: 3:55:53 time: 0.2406 data_time: 0.0082 memory: 1685 loss: 1.2967 2023/04/14 01:58:07 - mmengine - INFO - Epoch(train) [5][1900/5005] lr: 1.0000e-04 eta: 3:54:29 time: 0.2400 data_time: 0.0091 memory: 1685 loss: 1.1685 2023/04/14 01:58:27 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 01:58:32 - mmengine - INFO - Epoch(train) [5][2000/5005] lr: 1.0000e-04 eta: 3:53:07 time: 0.2398 data_time: 0.0083 memory: 1685 loss: 1.1697 2023/04/14 01:58:56 - mmengine - INFO - Epoch(train) [5][2100/5005] lr: 1.0000e-04 eta: 3:51:45 time: 0.2412 data_time: 0.0084 memory: 1685 loss: 1.1168 2023/04/14 01:59:20 - mmengine - INFO - Epoch(train) [5][2200/5005] lr: 1.0000e-04 eta: 3:50:23 time: 0.2818 data_time: 0.0096 memory: 1685 loss: 1.2509 2023/04/14 01:59:46 - mmengine - INFO - Epoch(train) [5][2300/5005] lr: 1.0000e-04 eta: 3:49:04 time: 0.2414 data_time: 0.0085 memory: 1685 loss: 1.2282 2023/04/14 02:00:10 - mmengine - INFO - Epoch(train) [5][2400/5005] lr: 1.0000e-04 eta: 3:47:43 time: 0.2383 data_time: 0.0099 memory: 1685 loss: 1.2798 2023/04/14 02:00:35 - mmengine - INFO - Epoch(train) [5][2500/5005] lr: 1.0000e-04 eta: 3:46:23 time: 0.2543 data_time: 0.0349 memory: 1685 loss: 1.1851 2023/04/14 02:01:01 - mmengine - INFO - Epoch(train) [5][2600/5005] lr: 1.0000e-04 eta: 3:45:06 time: 0.3134 data_time: 0.0086 memory: 1685 loss: 1.3071 2023/04/14 02:01:26 - mmengine - INFO - Epoch(train) [5][2700/5005] lr: 1.0000e-04 eta: 3:43:47 time: 0.2447 data_time: 0.0103 memory: 1685 loss: 1.2784 2023/04/14 02:01:53 - mmengine - INFO - Epoch(train) [5][2800/5005] lr: 1.0000e-04 eta: 3:42:31 time: 0.2424 data_time: 0.0114 memory: 1685 loss: 1.4795 2023/04/14 02:02:17 - mmengine - INFO - Epoch(train) [5][2900/5005] lr: 1.0000e-04 eta: 3:41:13 time: 0.2448 data_time: 0.0088 memory: 1685 loss: 1.3821 2023/04/14 02:02:36 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:02:41 - mmengine - INFO - Epoch(train) [5][3000/5005] lr: 1.0000e-04 eta: 3:39:55 time: 0.2416 data_time: 0.0106 memory: 1685 loss: 1.2472 2023/04/14 02:03:08 - mmengine - INFO - Epoch(train) [5][3100/5005] lr: 1.0000e-04 eta: 3:38:40 time: 0.2533 data_time: 0.0095 memory: 1685 loss: 1.3502 2023/04/14 02:03:34 - mmengine - INFO - Epoch(train) [5][3200/5005] lr: 1.0000e-04 eta: 3:37:25 time: 0.2406 data_time: 0.0093 memory: 1685 loss: 1.3389 2023/04/14 02:03:59 - mmengine - INFO - Epoch(train) [5][3300/5005] lr: 1.0000e-04 eta: 3:36:10 time: 0.2392 data_time: 0.0083 memory: 1685 loss: 1.1773 2023/04/14 02:04:24 - mmengine - INFO - Epoch(train) [5][3400/5005] lr: 1.0000e-04 eta: 3:34:54 time: 0.2353 data_time: 0.0079 memory: 1685 loss: 1.3417 2023/04/14 02:04:48 - mmengine - INFO - Epoch(train) [5][3500/5005] lr: 1.0000e-04 eta: 3:33:39 time: 0.2340 data_time: 0.0081 memory: 1685 loss: 1.3200 2023/04/14 02:05:13 - mmengine - INFO - Epoch(train) [5][3600/5005] lr: 1.0000e-04 eta: 3:32:24 time: 0.2408 data_time: 0.0086 memory: 1685 loss: 1.3093 2023/04/14 02:05:38 - mmengine - INFO - Epoch(train) [5][3700/5005] lr: 1.0000e-04 eta: 3:31:10 time: 0.2452 data_time: 0.0083 memory: 1685 loss: 1.2790 2023/04/14 02:06:04 - mmengine - INFO - Epoch(train) [5][3800/5005] lr: 1.0000e-04 eta: 3:29:58 time: 0.2778 data_time: 0.0081 memory: 1685 loss: 1.4382 2023/04/14 02:06:29 - mmengine - INFO - Epoch(train) [5][3900/5005] lr: 1.0000e-04 eta: 3:28:45 time: 0.2327 data_time: 0.0062 memory: 1685 loss: 1.1224 2023/04/14 02:06:49 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:06:54 - mmengine - INFO - Epoch(train) [5][4000/5005] lr: 1.0000e-04 eta: 3:27:32 time: 0.2436 data_time: 0.0089 memory: 1685 loss: 1.3689 2023/04/14 02:07:20 - mmengine - INFO - Epoch(train) [5][4100/5005] lr: 1.0000e-04 eta: 3:26:20 time: 0.2368 data_time: 0.0089 memory: 1685 loss: 1.2491 2023/04/14 02:07:46 - mmengine - INFO - Epoch(train) [5][4200/5005] lr: 1.0000e-04 eta: 3:25:09 time: 0.2442 data_time: 0.0083 memory: 1685 loss: 1.2678 2023/04/14 02:08:09 - mmengine - INFO - Epoch(train) [5][4300/5005] lr: 1.0000e-04 eta: 3:23:55 time: 0.1788 data_time: 0.0100 memory: 1685 loss: 1.4072 2023/04/14 02:08:36 - mmengine - INFO - Epoch(train) [5][4400/5005] lr: 1.0000e-04 eta: 3:22:46 time: 0.2493 data_time: 0.0086 memory: 1685 loss: 1.2737 2023/04/14 02:09:00 - mmengine - INFO - Epoch(train) [5][4500/5005] lr: 1.0000e-04 eta: 3:21:34 time: 0.2370 data_time: 0.0081 memory: 1685 loss: 1.1650 2023/04/14 02:09:25 - mmengine - INFO - Epoch(train) [5][4600/5005] lr: 1.0000e-04 eta: 3:20:24 time: 0.2481 data_time: 0.0084 memory: 1685 loss: 1.3345 2023/04/14 02:09:49 - mmengine - INFO - Epoch(train) [5][4700/5005] lr: 1.0000e-04 eta: 3:19:13 time: 0.2365 data_time: 0.0083 memory: 1685 loss: 1.1655 2023/04/14 02:10:13 - mmengine - INFO - Epoch(train) [5][4800/5005] lr: 1.0000e-04 eta: 3:18:02 time: 0.2500 data_time: 0.0083 memory: 1685 loss: 1.2022 2023/04/14 02:10:38 - mmengine - INFO - Epoch(train) [5][4900/5005] lr: 1.0000e-04 eta: 3:16:53 time: 0.2447 data_time: 0.0092 memory: 1685 loss: 1.2983 2023/04/14 02:10:57 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:11:02 - mmengine - INFO - Epoch(train) [5][5000/5005] lr: 1.0000e-04 eta: 3:15:43 time: 0.2408 data_time: 0.0114 memory: 1685 loss: 1.2626 2023/04/14 02:11:03 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:11:04 - mmengine - INFO - Saving checkpoint at 5 epochs 2023/04/14 02:11:21 - mmengine - INFO - Epoch(val) [5][100/196] eta: 0:00:15 time: 0.1639 data_time: 0.0078 memory: 1685 2023/04/14 02:12:38 - mmengine - INFO - Epoch(val) [5][196/196] qat.accuracy/top1: 69.3020 qat.accuracy/top5: 89.0140data_time: 0.0073 time: 0.2796 2023/04/14 02:12:41 - mmengine - INFO - Epoch(val) [5][100/196] eta: 0:00:03 time: 0.0273 data_time: 0.0108 memory: 969 2023/04/14 02:13:42 - mmengine - INFO - Epoch(val) [5][196/196] original.accuracy/top1: 69.2120 original.accuracy/top5: 88.9160data_time: 0.0292 time: 0.0411 2023/04/14 02:14:08 - mmengine - INFO - Epoch(train) [6][ 100/5005] lr: 1.0000e-04 eta: 3:14:31 time: 0.2347 data_time: 0.0084 memory: 1685 loss: 1.2917 2023/04/14 02:14:33 - mmengine - INFO - Epoch(train) [6][ 200/5005] lr: 1.0000e-04 eta: 3:13:23 time: 0.2686 data_time: 0.0086 memory: 1685 loss: 1.3332 2023/04/14 02:14:58 - mmengine - INFO - Epoch(train) [6][ 300/5005] lr: 1.0000e-04 eta: 3:12:15 time: 0.2355 data_time: 0.0082 memory: 1685 loss: 1.3764 2023/04/14 02:15:22 - mmengine - INFO - Epoch(train) [6][ 400/5005] lr: 1.0000e-04 eta: 3:11:06 time: 0.2411 data_time: 0.0081 memory: 1685 loss: 1.3726 2023/04/14 02:15:47 - mmengine - INFO - Epoch(train) [6][ 500/5005] lr: 1.0000e-04 eta: 3:09:59 time: 0.2348 data_time: 0.0086 memory: 1685 loss: 1.1845 2023/04/14 02:16:12 - mmengine - INFO - Epoch(train) [6][ 600/5005] lr: 1.0000e-04 eta: 3:08:52 time: 0.2402 data_time: 0.0092 memory: 1685 loss: 1.1395 2023/04/14 02:16:38 - mmengine - INFO - Epoch(train) [6][ 700/5005] lr: 1.0000e-04 eta: 3:07:46 time: 0.2468 data_time: 0.0081 memory: 1685 loss: 1.0904 2023/04/14 02:17:03 - mmengine - INFO - Epoch(train) [6][ 800/5005] lr: 1.0000e-04 eta: 3:06:40 time: 0.2436 data_time: 0.0084 memory: 1685 loss: 1.2369 2023/04/14 02:17:28 - mmengine - INFO - Epoch(train) [6][ 900/5005] lr: 1.0000e-04 eta: 3:05:34 time: 0.2452 data_time: 0.0083 memory: 1685 loss: 1.3389 2023/04/14 02:17:48 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:17:54 - mmengine - INFO - Epoch(train) [6][1000/5005] lr: 1.0000e-04 eta: 3:04:29 time: 0.2389 data_time: 0.0094 memory: 1685 loss: 1.3412 2023/04/14 02:18:18 - mmengine - INFO - Epoch(train) [6][1100/5005] lr: 1.0000e-04 eta: 3:03:23 time: 0.2357 data_time: 0.0083 memory: 1685 loss: 1.2697 2023/04/14 02:18:42 - mmengine - INFO - Epoch(train) [6][1200/5005] lr: 1.0000e-04 eta: 3:02:17 time: 0.2298 data_time: 0.0082 memory: 1685 loss: 1.2202 2023/04/14 02:19:07 - mmengine - INFO - Epoch(train) [6][1300/5005] lr: 1.0000e-04 eta: 3:01:12 time: 0.2420 data_time: 0.0088 memory: 1685 loss: 1.3698 2023/04/14 02:19:31 - mmengine - INFO - Epoch(train) [6][1400/5005] lr: 1.0000e-04 eta: 3:00:07 time: 0.2342 data_time: 0.0084 memory: 1685 loss: 1.3822 2023/04/14 02:19:57 - mmengine - INFO - Epoch(train) [6][1500/5005] lr: 1.0000e-04 eta: 2:59:03 time: 0.2486 data_time: 0.0085 memory: 1685 loss: 1.2310 2023/04/14 02:20:21 - mmengine - INFO - Epoch(train) [6][1600/5005] lr: 1.0000e-04 eta: 2:57:58 time: 0.2326 data_time: 0.0104 memory: 1685 loss: 1.3026 2023/04/14 02:20:45 - mmengine - INFO - Epoch(train) [6][1700/5005] lr: 1.0000e-04 eta: 2:56:54 time: 0.2461 data_time: 0.0086 memory: 1685 loss: 1.4124 2023/04/14 02:21:10 - mmengine - INFO - Epoch(train) [6][1800/5005] lr: 1.0000e-04 eta: 2:55:51 time: 0.2425 data_time: 0.0195 memory: 1685 loss: 1.2056 2023/04/14 02:21:35 - mmengine - INFO - Epoch(train) [6][1900/5005] lr: 1.0000e-04 eta: 2:54:48 time: 0.2495 data_time: 0.0087 memory: 1685 loss: 1.3997 2023/04/14 02:21:53 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:22:00 - mmengine - INFO - Epoch(train) [6][2000/5005] lr: 1.0000e-04 eta: 2:53:45 time: 0.2921 data_time: 0.0088 memory: 1685 loss: 1.2450 2023/04/14 02:22:23 - mmengine - INFO - Epoch(train) [6][2100/5005] lr: 1.0000e-04 eta: 2:52:41 time: 0.2379 data_time: 0.0082 memory: 1685 loss: 1.2274 2023/04/14 02:22:48 - mmengine - INFO - Epoch(train) [6][2200/5005] lr: 1.0000e-04 eta: 2:51:39 time: 0.2845 data_time: 0.0086 memory: 1685 loss: 1.2765 2023/04/14 02:23:12 - mmengine - INFO - Epoch(train) [6][2300/5005] lr: 1.0000e-04 eta: 2:50:37 time: 0.2697 data_time: 0.0080 memory: 1685 loss: 1.2808 2023/04/14 02:23:37 - mmengine - INFO - Epoch(train) [6][2400/5005] lr: 1.0000e-04 eta: 2:49:35 time: 0.2348 data_time: 0.0079 memory: 1685 loss: 1.3292 2023/04/14 02:24:00 - mmengine - INFO - Epoch(train) [6][2500/5005] lr: 1.0000e-04 eta: 2:48:32 time: 0.2373 data_time: 0.0082 memory: 1685 loss: 1.0479 2023/04/14 02:24:24 - mmengine - INFO - Epoch(train) [6][2600/5005] lr: 1.0000e-04 eta: 2:47:30 time: 0.2443 data_time: 0.0080 memory: 1685 loss: 1.3707 2023/04/14 02:24:48 - mmengine - INFO - Epoch(train) [6][2700/5005] lr: 1.0000e-04 eta: 2:46:29 time: 0.2451 data_time: 0.0079 memory: 1685 loss: 1.2311 2023/04/14 02:25:13 - mmengine - INFO - Epoch(train) [6][2800/5005] lr: 1.0000e-04 eta: 2:45:28 time: 0.2280 data_time: 0.0083 memory: 1685 loss: 1.1360 2023/04/14 02:25:37 - mmengine - INFO - Epoch(train) [6][2900/5005] lr: 1.0000e-04 eta: 2:44:27 time: 0.2480 data_time: 0.0079 memory: 1685 loss: 1.1723 2023/04/14 02:25:56 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:26:02 - mmengine - INFO - Epoch(train) [6][3000/5005] lr: 1.0000e-04 eta: 2:43:27 time: 0.2374 data_time: 0.0080 memory: 1685 loss: 1.3955 2023/04/14 02:26:26 - mmengine - INFO - Epoch(train) [6][3100/5005] lr: 1.0000e-04 eta: 2:42:26 time: 0.2320 data_time: 0.0082 memory: 1685 loss: 1.3568 2023/04/14 02:26:50 - mmengine - INFO - Epoch(train) [6][3200/5005] lr: 1.0000e-04 eta: 2:41:26 time: 0.2293 data_time: 0.0094 memory: 1685 loss: 1.3149 2023/04/14 02:27:15 - mmengine - INFO - Epoch(train) [6][3300/5005] lr: 1.0000e-04 eta: 2:40:27 time: 0.2354 data_time: 0.0081 memory: 1685 loss: 1.2464 2023/04/14 02:27:38 - mmengine - INFO - Epoch(train) [6][3400/5005] lr: 1.0000e-04 eta: 2:39:26 time: 0.2331 data_time: 0.0080 memory: 1685 loss: 1.1570 2023/04/14 02:28:03 - mmengine - INFO - Epoch(train) [6][3500/5005] lr: 1.0000e-04 eta: 2:38:28 time: 0.2816 data_time: 0.0197 memory: 1685 loss: 1.1907 2023/04/14 02:28:26 - mmengine - INFO - Epoch(train) [6][3600/5005] lr: 1.0000e-04 eta: 2:37:28 time: 0.2256 data_time: 0.0087 memory: 1685 loss: 1.3342 2023/04/14 02:28:49 - mmengine - INFO - Epoch(train) [6][3700/5005] lr: 1.0000e-04 eta: 2:36:28 time: 0.2257 data_time: 0.0081 memory: 1685 loss: 1.2705 2023/04/14 02:29:13 - mmengine - INFO - Epoch(train) [6][3800/5005] lr: 1.0000e-04 eta: 2:35:29 time: 0.2294 data_time: 0.0083 memory: 1685 loss: 1.1612 2023/04/14 02:29:36 - mmengine - INFO - Epoch(train) [6][3900/5005] lr: 1.0000e-04 eta: 2:34:30 time: 0.2272 data_time: 0.0082 memory: 1685 loss: 1.4370 2023/04/14 02:29:55 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:30:00 - mmengine - INFO - Epoch(train) [6][4000/5005] lr: 1.0000e-04 eta: 2:33:31 time: 0.2283 data_time: 0.0083 memory: 1685 loss: 1.3243 2023/04/14 02:30:24 - mmengine - INFO - Epoch(train) [6][4100/5005] lr: 1.0000e-04 eta: 2:32:33 time: 0.2343 data_time: 0.0080 memory: 1685 loss: 1.1741 2023/04/14 02:30:48 - mmengine - INFO - Epoch(train) [6][4200/5005] lr: 1.0000e-04 eta: 2:31:35 time: 0.2385 data_time: 0.0083 memory: 1685 loss: 1.2999 2023/04/14 02:31:12 - mmengine - INFO - Epoch(train) [6][4300/5005] lr: 1.0000e-04 eta: 2:30:38 time: 0.2521 data_time: 0.0082 memory: 1685 loss: 1.3095 2023/04/14 02:31:36 - mmengine - INFO - Epoch(train) [6][4400/5005] lr: 1.0000e-04 eta: 2:29:40 time: 0.2782 data_time: 0.0083 memory: 1685 loss: 1.2014 2023/04/14 02:32:00 - mmengine - INFO - Epoch(train) [6][4500/5005] lr: 1.0000e-04 eta: 2:28:43 time: 0.2376 data_time: 0.0084 memory: 1685 loss: 1.3871 2023/04/14 02:32:24 - mmengine - INFO - Epoch(train) [6][4600/5005] lr: 1.0000e-04 eta: 2:27:47 time: 0.2345 data_time: 0.0080 memory: 1685 loss: 1.1666 2023/04/14 02:32:48 - mmengine - INFO - Epoch(train) [6][4700/5005] lr: 1.0000e-04 eta: 2:26:50 time: 0.2442 data_time: 0.0081 memory: 1685 loss: 1.3106 2023/04/14 02:33:12 - mmengine - INFO - Epoch(train) [6][4800/5005] lr: 1.0000e-04 eta: 2:25:53 time: 0.2288 data_time: 0.0087 memory: 1685 loss: 1.2850 2023/04/14 02:33:35 - mmengine - INFO - Epoch(train) [6][4900/5005] lr: 1.0000e-04 eta: 2:24:56 time: 0.2321 data_time: 0.0085 memory: 1685 loss: 1.2198 2023/04/14 02:33:53 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:33:59 - mmengine - INFO - Epoch(train) [6][5000/5005] lr: 1.0000e-04 eta: 2:24:00 time: 0.2447 data_time: 0.0090 memory: 1685 loss: 1.1018 2023/04/14 02:34:00 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:34:02 - mmengine - INFO - Saving checkpoint at 6 epochs 2023/04/14 02:34:18 - mmengine - INFO - Epoch(val) [6][100/196] eta: 0:00:14 time: 0.1278 data_time: 0.0084 memory: 1685 2023/04/14 02:35:29 - mmengine - INFO - Epoch(val) [6][196/196] qat.accuracy/top1: 69.2000 qat.accuracy/top5: 89.0460data_time: 0.0070 time: 0.3112 2023/04/14 02:35:33 - mmengine - INFO - Epoch(val) [6][100/196] eta: 0:00:03 time: 0.0308 data_time: 0.0158 memory: 969 2023/04/14 02:36:30 - mmengine - INFO - Epoch(val) [6][196/196] original.accuracy/top1: 69.2180 original.accuracy/top5: 88.9180data_time: 0.0182 time: 0.0315 2023/04/14 02:36:57 - mmengine - INFO - Epoch(train) [7][ 100/5005] lr: 1.0000e-04 eta: 2:23:04 time: 0.2408 data_time: 0.0082 memory: 1685 loss: 1.2366 2023/04/14 02:37:21 - mmengine - INFO - Epoch(train) [7][ 200/5005] lr: 1.0000e-04 eta: 2:22:08 time: 0.2335 data_time: 0.0083 memory: 1685 loss: 1.1512 2023/04/14 02:37:45 - mmengine - INFO - Epoch(train) [7][ 300/5005] lr: 1.0000e-04 eta: 2:21:13 time: 0.2312 data_time: 0.0085 memory: 1685 loss: 0.9835 2023/04/14 02:38:09 - mmengine - INFO - Epoch(train) [7][ 400/5005] lr: 1.0000e-04 eta: 2:20:18 time: 0.2486 data_time: 0.0083 memory: 1685 loss: 1.1132 2023/04/14 02:38:33 - mmengine - INFO - Epoch(train) [7][ 500/5005] lr: 1.0000e-04 eta: 2:19:23 time: 0.2343 data_time: 0.0091 memory: 1685 loss: 1.2383 2023/04/14 02:38:58 - mmengine - INFO - Epoch(train) [7][ 600/5005] lr: 1.0000e-04 eta: 2:18:28 time: 0.2874 data_time: 0.0084 memory: 1685 loss: 1.0620 2023/04/14 02:39:22 - mmengine - INFO - Epoch(train) [7][ 700/5005] lr: 1.0000e-04 eta: 2:17:34 time: 0.2408 data_time: 0.0081 memory: 1685 loss: 1.2255 2023/04/14 02:39:47 - mmengine - INFO - Epoch(train) [7][ 800/5005] lr: 1.0000e-04 eta: 2:16:40 time: 0.2455 data_time: 0.0088 memory: 1685 loss: 1.3291 2023/04/14 02:40:12 - mmengine - INFO - Epoch(train) [7][ 900/5005] lr: 1.0000e-04 eta: 2:15:46 time: 0.2466 data_time: 0.0083 memory: 1685 loss: 1.2662 2023/04/14 02:40:30 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:40:38 - mmengine - INFO - Epoch(train) [7][1000/5005] lr: 1.0000e-04 eta: 2:14:54 time: 0.2508 data_time: 0.0083 memory: 1685 loss: 1.2070 2023/04/14 02:41:04 - mmengine - INFO - Epoch(train) [7][1100/5005] lr: 1.0000e-04 eta: 2:14:01 time: 0.2513 data_time: 0.0084 memory: 1685 loss: 1.2094 2023/04/14 02:41:29 - mmengine - INFO - Epoch(train) [7][1200/5005] lr: 1.0000e-04 eta: 2:13:08 time: 0.2566 data_time: 0.0083 memory: 1685 loss: 1.1604 2023/04/14 02:41:54 - mmengine - INFO - Epoch(train) [7][1300/5005] lr: 1.0000e-04 eta: 2:12:15 time: 0.2485 data_time: 0.0087 memory: 1685 loss: 1.3109 2023/04/14 02:42:20 - mmengine - INFO - Epoch(train) [7][1400/5005] lr: 1.0000e-04 eta: 2:11:23 time: 0.2684 data_time: 0.0084 memory: 1685 loss: 1.2069 2023/04/14 02:42:45 - mmengine - INFO - Epoch(train) [7][1500/5005] lr: 1.0000e-04 eta: 2:10:30 time: 0.2303 data_time: 0.0082 memory: 1685 loss: 1.2486 2023/04/14 02:43:09 - mmengine - INFO - Epoch(train) [7][1600/5005] lr: 1.0000e-04 eta: 2:09:38 time: 0.2448 data_time: 0.0081 memory: 1685 loss: 1.3219 2023/04/14 02:43:34 - mmengine - INFO - Epoch(train) [7][1700/5005] lr: 1.0000e-04 eta: 2:08:45 time: 0.2457 data_time: 0.0081 memory: 1685 loss: 1.2728 2023/04/14 02:44:00 - mmengine - INFO - Epoch(train) [7][1800/5005] lr: 1.0000e-04 eta: 2:07:54 time: 0.2495 data_time: 0.0100 memory: 1685 loss: 1.3057 2023/04/14 02:44:25 - mmengine - INFO - Epoch(train) [7][1900/5005] lr: 1.0000e-04 eta: 2:07:02 time: 0.2310 data_time: 0.0084 memory: 1685 loss: 1.2684 2023/04/14 02:44:42 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:44:49 - mmengine - INFO - Epoch(train) [7][2000/5005] lr: 1.0000e-04 eta: 2:06:10 time: 0.2328 data_time: 0.0089 memory: 1685 loss: 1.4815 2023/04/14 02:45:14 - mmengine - INFO - Epoch(train) [7][2100/5005] lr: 1.0000e-04 eta: 2:05:18 time: 0.2407 data_time: 0.0085 memory: 1685 loss: 1.1589 2023/04/14 02:45:39 - mmengine - INFO - Epoch(train) [7][2200/5005] lr: 1.0000e-04 eta: 2:04:27 time: 0.2470 data_time: 0.0080 memory: 1685 loss: 1.1789 2023/04/14 02:46:03 - mmengine - INFO - Epoch(train) [7][2300/5005] lr: 1.0000e-04 eta: 2:03:35 time: 0.2334 data_time: 0.0095 memory: 1685 loss: 1.4155 2023/04/14 02:46:27 - mmengine - INFO - Epoch(train) [7][2400/5005] lr: 1.0000e-04 eta: 2:02:44 time: 0.2300 data_time: 0.0084 memory: 1685 loss: 1.1912 2023/04/14 02:46:51 - mmengine - INFO - Epoch(train) [7][2500/5005] lr: 1.0000e-04 eta: 2:01:52 time: 0.2413 data_time: 0.0082 memory: 1685 loss: 1.1313 2023/04/14 02:47:15 - mmengine - INFO - Epoch(train) [7][2600/5005] lr: 1.0000e-04 eta: 2:01:01 time: 0.2423 data_time: 0.0087 memory: 1685 loss: 1.3419 2023/04/14 02:47:40 - mmengine - INFO - Epoch(train) [7][2700/5005] lr: 1.0000e-04 eta: 2:00:11 time: 0.2374 data_time: 0.0081 memory: 1685 loss: 1.4087 2023/04/14 02:48:05 - mmengine - INFO - Epoch(train) [7][2800/5005] lr: 1.0000e-04 eta: 1:59:20 time: 0.2567 data_time: 0.0085 memory: 1685 loss: 1.1449 2023/04/14 02:48:29 - mmengine - INFO - Epoch(train) [7][2900/5005] lr: 1.0000e-04 eta: 1:58:30 time: 0.2442 data_time: 0.0086 memory: 1685 loss: 1.1795 2023/04/14 02:48:46 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:48:53 - mmengine - INFO - Epoch(train) [7][3000/5005] lr: 1.0000e-04 eta: 1:57:39 time: 0.2528 data_time: 0.0079 memory: 1685 loss: 1.2109 2023/04/14 02:49:19 - mmengine - INFO - Epoch(train) [7][3100/5005] lr: 1.0000e-04 eta: 1:56:50 time: 0.2429 data_time: 0.0079 memory: 1685 loss: 1.1807 2023/04/14 02:49:44 - mmengine - INFO - Epoch(train) [7][3200/5005] lr: 1.0000e-04 eta: 1:56:00 time: 0.2413 data_time: 0.0079 memory: 1685 loss: 1.4232 2023/04/14 02:50:09 - mmengine - INFO - Epoch(train) [7][3300/5005] lr: 1.0000e-04 eta: 1:55:10 time: 0.2416 data_time: 0.0088 memory: 1685 loss: 1.1962 2023/04/14 02:50:35 - mmengine - INFO - Epoch(train) [7][3400/5005] lr: 1.0000e-04 eta: 1:54:21 time: 0.2633 data_time: 0.0090 memory: 1685 loss: 1.2416 2023/04/14 02:51:01 - mmengine - INFO - Epoch(train) [7][3500/5005] lr: 1.0000e-04 eta: 1:53:32 time: 0.2631 data_time: 0.0101 memory: 1685 loss: 1.0733 2023/04/14 02:51:26 - mmengine - INFO - Epoch(train) [7][3600/5005] lr: 1.0000e-04 eta: 1:52:43 time: 0.2526 data_time: 0.0087 memory: 1685 loss: 1.2221 2023/04/14 02:51:46 - mmengine - INFO - Epoch(train) [7][3700/5005] lr: 1.0000e-04 eta: 1:51:52 time: 0.1953 data_time: 0.0131 memory: 1685 loss: 1.1557 2023/04/14 02:52:13 - mmengine - INFO - Epoch(train) [7][3800/5005] lr: 1.0000e-04 eta: 1:51:04 time: 0.2262 data_time: 0.0094 memory: 1685 loss: 1.2416 2023/04/14 02:52:38 - mmengine - INFO - Epoch(train) [7][3900/5005] lr: 1.0000e-04 eta: 1:50:15 time: 0.2363 data_time: 0.0090 memory: 1685 loss: 1.4208 2023/04/14 02:52:56 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:53:04 - mmengine - INFO - Epoch(train) [7][4000/5005] lr: 1.0000e-04 eta: 1:49:27 time: 0.2916 data_time: 0.0106 memory: 1685 loss: 1.4611 2023/04/14 02:53:28 - mmengine - INFO - Epoch(train) [7][4100/5005] lr: 1.0000e-04 eta: 1:48:38 time: 0.2557 data_time: 0.0087 memory: 1685 loss: 1.3470 2023/04/14 02:53:51 - mmengine - INFO - Epoch(train) [7][4200/5005] lr: 1.0000e-04 eta: 1:47:49 time: 0.2430 data_time: 0.0085 memory: 1685 loss: 1.2430 2023/04/14 02:54:16 - mmengine - INFO - Epoch(train) [7][4300/5005] lr: 1.0000e-04 eta: 1:47:00 time: 0.2893 data_time: 0.0088 memory: 1685 loss: 1.2764 2023/04/14 02:54:40 - mmengine - INFO - Epoch(train) [7][4400/5005] lr: 1.0000e-04 eta: 1:46:12 time: 0.2320 data_time: 0.0092 memory: 1685 loss: 1.3110 2023/04/14 02:55:05 - mmengine - INFO - Epoch(train) [7][4500/5005] lr: 1.0000e-04 eta: 1:45:24 time: 0.3036 data_time: 0.0095 memory: 1685 loss: 1.3277 2023/04/14 02:55:29 - mmengine - INFO - Epoch(train) [7][4600/5005] lr: 1.0000e-04 eta: 1:44:36 time: 0.2306 data_time: 0.0092 memory: 1685 loss: 1.3756 2023/04/14 02:55:53 - mmengine - INFO - Epoch(train) [7][4700/5005] lr: 1.0000e-04 eta: 1:43:48 time: 0.2597 data_time: 0.0082 memory: 1685 loss: 1.2851 2023/04/14 02:56:17 - mmengine - INFO - Epoch(train) [7][4800/5005] lr: 1.0000e-04 eta: 1:43:00 time: 0.2318 data_time: 0.0085 memory: 1685 loss: 1.5602 2023/04/14 02:56:40 - mmengine - INFO - Epoch(train) [7][4900/5005] lr: 1.0000e-04 eta: 1:42:12 time: 0.2417 data_time: 0.0082 memory: 1685 loss: 1.2574 2023/04/14 02:56:58 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:57:05 - mmengine - INFO - Epoch(train) [7][5000/5005] lr: 1.0000e-04 eta: 1:41:25 time: 0.2300 data_time: 0.0101 memory: 1685 loss: 1.3839 2023/04/14 02:57:06 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 02:57:07 - mmengine - INFO - Saving checkpoint at 7 epochs 2023/04/14 02:57:23 - mmengine - INFO - Epoch(val) [7][100/196] eta: 0:00:13 time: 0.1316 data_time: 0.0080 memory: 1685 2023/04/14 02:58:32 - mmengine - INFO - Epoch(val) [7][196/196] qat.accuracy/top1: 69.3780 qat.accuracy/top5: 88.9600data_time: 0.0208 time: 0.3466 2023/04/14 02:58:36 - mmengine - INFO - Epoch(val) [7][100/196] eta: 0:00:04 time: 0.0321 data_time: 0.0162 memory: 969 2023/04/14 02:59:33 - mmengine - INFO - Epoch(val) [7][196/196] original.accuracy/top1: 69.1700 original.accuracy/top5: 88.8360data_time: 0.0291 time: 0.0430 2023/04/14 02:59:59 - mmengine - INFO - Epoch(train) [8][ 100/5005] lr: 1.0000e-04 eta: 1:40:35 time: 0.2513 data_time: 0.0087 memory: 1685 loss: 1.3568 2023/04/14 03:00:24 - mmengine - INFO - Epoch(train) [8][ 200/5005] lr: 1.0000e-04 eta: 1:39:49 time: 0.2664 data_time: 0.0089 memory: 1685 loss: 1.5285 2023/04/14 03:00:48 - mmengine - INFO - Epoch(train) [8][ 300/5005] lr: 1.0000e-04 eta: 1:39:01 time: 0.2333 data_time: 0.0087 memory: 1685 loss: 1.3419 2023/04/14 03:01:14 - mmengine - INFO - Epoch(train) [8][ 400/5005] lr: 1.0000e-04 eta: 1:38:15 time: 0.2511 data_time: 0.0090 memory: 1685 loss: 1.2688 2023/04/14 03:01:38 - mmengine - INFO - Epoch(train) [8][ 500/5005] lr: 1.0000e-04 eta: 1:37:28 time: 0.2337 data_time: 0.0089 memory: 1685 loss: 1.2000 2023/04/14 03:02:04 - mmengine - INFO - Epoch(train) [8][ 600/5005] lr: 1.0000e-04 eta: 1:36:42 time: 0.2396 data_time: 0.0097 memory: 1685 loss: 1.2756 2023/04/14 03:02:28 - mmengine - INFO - Epoch(train) [8][ 700/5005] lr: 1.0000e-04 eta: 1:35:55 time: 0.2358 data_time: 0.0094 memory: 1685 loss: 1.3297 2023/04/14 03:02:52 - mmengine - INFO - Epoch(train) [8][ 800/5005] lr: 1.0000e-04 eta: 1:35:08 time: 0.2635 data_time: 0.0094 memory: 1685 loss: 1.2704 2023/04/14 03:03:17 - mmengine - INFO - Epoch(train) [8][ 900/5005] lr: 1.0000e-04 eta: 1:34:22 time: 0.2445 data_time: 0.0096 memory: 1685 loss: 1.3944 2023/04/14 03:03:33 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 03:03:42 - mmengine - INFO - Epoch(train) [8][1000/5005] lr: 1.0000e-04 eta: 1:33:36 time: 0.2271 data_time: 0.0094 memory: 1685 loss: 1.2023 2023/04/14 03:04:08 - mmengine - INFO - Epoch(train) [8][1100/5005] lr: 1.0000e-04 eta: 1:32:51 time: 0.2941 data_time: 0.0097 memory: 1685 loss: 1.4243 2023/04/14 03:04:32 - mmengine - INFO - Epoch(train) [8][1200/5005] lr: 1.0000e-04 eta: 1:32:05 time: 0.2512 data_time: 0.0093 memory: 1685 loss: 1.3232 2023/04/14 03:04:56 - mmengine - INFO - Epoch(train) [8][1300/5005] lr: 1.0000e-04 eta: 1:31:19 time: 0.2329 data_time: 0.0094 memory: 1685 loss: 1.4617 2023/04/14 03:05:21 - mmengine - INFO - Epoch(train) [8][1400/5005] lr: 1.0000e-04 eta: 1:30:33 time: 0.2406 data_time: 0.0107 memory: 1685 loss: 1.4335 2023/04/14 03:05:46 - mmengine - INFO - Epoch(train) [8][1500/5005] lr: 1.0000e-04 eta: 1:29:47 time: 0.2348 data_time: 0.0089 memory: 1685 loss: 1.1627 2023/04/14 03:06:11 - mmengine - INFO - Epoch(train) [8][1600/5005] lr: 1.0000e-04 eta: 1:29:02 time: 0.2465 data_time: 0.0089 memory: 1685 loss: 1.1385 2023/04/14 03:06:35 - mmengine - INFO - Epoch(train) [8][1700/5005] lr: 1.0000e-04 eta: 1:28:17 time: 0.2315 data_time: 0.0084 memory: 1685 loss: 1.2622 2023/04/14 03:07:00 - mmengine - INFO - Epoch(train) [8][1800/5005] lr: 1.0000e-04 eta: 1:27:32 time: 0.2853 data_time: 0.0084 memory: 1685 loss: 1.3941 2023/04/14 03:07:24 - mmengine - INFO - Epoch(train) [8][1900/5005] lr: 1.0000e-04 eta: 1:26:46 time: 0.2402 data_time: 0.0103 memory: 1685 loss: 1.3012 2023/04/14 03:07:40 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 03:07:48 - mmengine - INFO - Epoch(train) [8][2000/5005] lr: 1.0000e-04 eta: 1:26:01 time: 0.2293 data_time: 0.0085 memory: 1685 loss: 1.3159 2023/04/14 03:08:12 - mmengine - INFO - Epoch(train) [8][2100/5005] lr: 1.0000e-04 eta: 1:25:16 time: 0.2451 data_time: 0.0084 memory: 1685 loss: 1.3463 2023/04/14 03:08:37 - mmengine - INFO - Epoch(train) [8][2200/5005] lr: 1.0000e-04 eta: 1:24:31 time: 0.2362 data_time: 0.0085 memory: 1685 loss: 1.3715 2023/04/14 03:09:01 - mmengine - INFO - Epoch(train) [8][2300/5005] lr: 1.0000e-04 eta: 1:23:46 time: 0.2425 data_time: 0.0080 memory: 1685 loss: 1.3191 2023/04/14 03:09:25 - mmengine - INFO - Epoch(train) [8][2400/5005] lr: 1.0000e-04 eta: 1:23:02 time: 0.2295 data_time: 0.0082 memory: 1685 loss: 1.3733 2023/04/14 03:09:49 - mmengine - INFO - Epoch(train) [8][2500/5005] lr: 1.0000e-04 eta: 1:22:17 time: 0.2714 data_time: 0.0085 memory: 1685 loss: 1.2534 2023/04/14 03:10:13 - mmengine - INFO - Epoch(train) [8][2600/5005] lr: 1.0000e-04 eta: 1:21:32 time: 0.2264 data_time: 0.0079 memory: 1685 loss: 1.1392 2023/04/14 03:10:38 - mmengine - INFO - Epoch(train) [8][2700/5005] lr: 1.0000e-04 eta: 1:20:48 time: 0.2272 data_time: 0.0080 memory: 1685 loss: 1.1841 2023/04/14 03:11:03 - mmengine - INFO - Epoch(train) [8][2800/5005] lr: 1.0000e-04 eta: 1:20:04 time: 0.2379 data_time: 0.0081 memory: 1685 loss: 1.1673 2023/04/14 03:11:26 - mmengine - INFO - Epoch(train) [8][2900/5005] lr: 1.0000e-04 eta: 1:19:20 time: 0.2331 data_time: 0.0089 memory: 1685 loss: 1.3133 2023/04/14 03:11:42 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 03:11:50 - mmengine - INFO - Epoch(train) [8][3000/5005] lr: 1.0000e-04 eta: 1:18:36 time: 0.2311 data_time: 0.0080 memory: 1685 loss: 1.2742 2023/04/14 03:12:14 - mmengine - INFO - Epoch(train) [8][3100/5005] lr: 1.0000e-04 eta: 1:17:52 time: 0.2331 data_time: 0.0079 memory: 1685 loss: 1.1944 2023/04/14 03:12:38 - mmengine - INFO - Epoch(train) [8][3200/5005] lr: 1.0000e-04 eta: 1:17:08 time: 0.2366 data_time: 0.0080 memory: 1685 loss: 1.3251 2023/04/14 03:13:02 - mmengine - INFO - Epoch(train) [8][3300/5005] lr: 1.0000e-04 eta: 1:16:24 time: 0.2352 data_time: 0.0089 memory: 1685 loss: 1.1564 2023/04/14 03:13:26 - mmengine - INFO - Epoch(train) [8][3400/5005] lr: 1.0000e-04 eta: 1:15:40 time: 0.2233 data_time: 0.0085 memory: 1685 loss: 1.4123 2023/04/14 03:13:49 - mmengine - INFO - Epoch(train) [8][3500/5005] lr: 1.0000e-04 eta: 1:14:56 time: 0.2429 data_time: 0.0086 memory: 1685 loss: 1.3044 2023/04/14 03:14:14 - mmengine - INFO - Epoch(train) [8][3600/5005] lr: 1.0000e-04 eta: 1:14:13 time: 0.2321 data_time: 0.0087 memory: 1685 loss: 1.2018 2023/04/14 03:14:37 - mmengine - INFO - Epoch(train) [8][3700/5005] lr: 1.0000e-04 eta: 1:13:29 time: 0.2429 data_time: 0.0085 memory: 1685 loss: 1.2740 2023/04/14 03:15:01 - mmengine - INFO - Epoch(train) [8][3800/5005] lr: 1.0000e-04 eta: 1:12:46 time: 0.2383 data_time: 0.0086 memory: 1685 loss: 1.5691 2023/04/14 03:15:25 - mmengine - INFO - Epoch(train) [8][3900/5005] lr: 1.0000e-04 eta: 1:12:03 time: 0.2361 data_time: 0.0082 memory: 1685 loss: 1.2993 2023/04/14 03:15:41 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 03:15:49 - mmengine - INFO - Epoch(train) [8][4000/5005] lr: 1.0000e-04 eta: 1:11:20 time: 0.2306 data_time: 0.0094 memory: 1685 loss: 1.2630 2023/04/14 03:16:13 - mmengine - INFO - Epoch(train) [8][4100/5005] lr: 1.0000e-04 eta: 1:10:37 time: 0.2420 data_time: 0.0082 memory: 1685 loss: 1.3387 2023/04/14 03:16:37 - mmengine - INFO - Epoch(train) [8][4200/5005] lr: 1.0000e-04 eta: 1:09:54 time: 0.2551 data_time: 0.0081 memory: 1685 loss: 1.2296 2023/04/14 03:17:02 - mmengine - INFO - Epoch(train) [8][4300/5005] lr: 1.0000e-04 eta: 1:09:11 time: 0.2280 data_time: 0.0084 memory: 1685 loss: 1.3209 2023/04/14 03:17:25 - mmengine - INFO - Epoch(train) [8][4400/5005] lr: 1.0000e-04 eta: 1:08:28 time: 0.2355 data_time: 0.0090 memory: 1685 loss: 1.3625 2023/04/14 03:17:49 - mmengine - INFO - Epoch(train) [8][4500/5005] lr: 1.0000e-04 eta: 1:07:46 time: 0.2358 data_time: 0.0086 memory: 1685 loss: 1.3007 2023/04/14 03:18:12 - mmengine - INFO - Epoch(train) [8][4600/5005] lr: 1.0000e-04 eta: 1:07:03 time: 0.2274 data_time: 0.0085 memory: 1685 loss: 1.1734 2023/04/14 03:18:36 - mmengine - INFO - Epoch(train) [8][4700/5005] lr: 1.0000e-04 eta: 1:06:20 time: 0.2845 data_time: 0.0081 memory: 1685 loss: 1.2052 2023/04/14 03:19:00 - mmengine - INFO - Epoch(train) [8][4800/5005] lr: 1.0000e-04 eta: 1:05:38 time: 0.2293 data_time: 0.0097 memory: 1685 loss: 1.4795 2023/04/14 03:19:24 - mmengine - INFO - Epoch(train) [8][4900/5005] lr: 1.0000e-04 eta: 1:04:56 time: 0.2309 data_time: 0.0080 memory: 1685 loss: 1.2268 2023/04/14 03:19:39 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 03:19:47 - mmengine - INFO - Epoch(train) [8][5000/5005] lr: 1.0000e-04 eta: 1:04:13 time: 0.2338 data_time: 0.0090 memory: 1685 loss: 1.3747 2023/04/14 03:19:49 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 03:19:50 - mmengine - INFO - Saving checkpoint at 8 epochs 2023/04/14 03:20:06 - mmengine - INFO - Epoch(val) [8][100/196] eta: 0:00:14 time: 0.1423 data_time: 0.0078 memory: 1685 2023/04/14 03:21:15 - mmengine - INFO - Epoch(val) [8][196/196] qat.accuracy/top1: 69.3900 qat.accuracy/top5: 89.0340data_time: 0.0071 time: 0.2522 2023/04/14 03:21:19 - mmengine - INFO - Epoch(val) [8][100/196] eta: 0:00:03 time: 0.0264 data_time: 0.0125 memory: 969 2023/04/14 03:22:16 - mmengine - INFO - Epoch(val) [8][196/196] original.accuracy/top1: 69.3980 original.accuracy/top5: 89.0300data_time: 0.0270 time: 0.0410 2023/04/14 03:22:41 - mmengine - INFO - Epoch(train) [9][ 100/5005] lr: 1.0000e-04 eta: 1:03:30 time: 0.2489 data_time: 0.0081 memory: 1685 loss: 1.2702 2023/04/14 03:23:06 - mmengine - INFO - Epoch(train) [9][ 200/5005] lr: 1.0000e-04 eta: 1:02:48 time: 0.2360 data_time: 0.0082 memory: 1685 loss: 1.0557 2023/04/14 03:23:30 - mmengine - INFO - Epoch(train) [9][ 300/5005] lr: 1.0000e-04 eta: 1:02:06 time: 0.2376 data_time: 0.0083 memory: 1685 loss: 1.2405 2023/04/14 03:23:54 - mmengine - INFO - Epoch(train) [9][ 400/5005] lr: 1.0000e-04 eta: 1:01:24 time: 0.2570 data_time: 0.0084 memory: 1685 loss: 1.2567 2023/04/14 03:24:19 - mmengine - INFO - Epoch(train) [9][ 500/5005] lr: 1.0000e-04 eta: 1:00:43 time: 0.2390 data_time: 0.0081 memory: 1685 loss: 1.3326 2023/04/14 03:24:46 - mmengine - INFO - Epoch(train) [9][ 600/5005] lr: 1.0000e-04 eta: 1:00:02 time: 0.2412 data_time: 0.0081 memory: 1685 loss: 1.2403 2023/04/14 03:25:10 - mmengine - INFO - Epoch(train) [9][ 700/5005] lr: 1.0000e-04 eta: 0:59:20 time: 0.2384 data_time: 0.0089 memory: 1685 loss: 1.4857 2023/04/14 03:25:34 - mmengine - INFO - Epoch(train) [9][ 800/5005] lr: 1.0000e-04 eta: 0:58:39 time: 0.2419 data_time: 0.0081 memory: 1685 loss: 1.4466 2023/04/14 03:25:59 - mmengine - INFO - Epoch(train) [9][ 900/5005] lr: 1.0000e-04 eta: 0:57:58 time: 0.2353 data_time: 0.0083 memory: 1685 loss: 1.3458 2023/04/14 03:26:13 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 03:26:23 - mmengine - INFO - Epoch(train) [9][1000/5005] lr: 1.0000e-04 eta: 0:57:16 time: 0.2391 data_time: 0.0087 memory: 1685 loss: 1.2833 2023/04/14 03:26:49 - mmengine - INFO - Epoch(train) [9][1100/5005] lr: 1.0000e-04 eta: 0:56:36 time: 0.2828 data_time: 0.0075 memory: 1685 loss: 1.2999 2023/04/14 03:27:13 - mmengine - INFO - Epoch(train) [9][1200/5005] lr: 1.0000e-04 eta: 0:55:54 time: 0.2487 data_time: 0.0082 memory: 1685 loss: 1.1124 2023/04/14 03:27:38 - mmengine - INFO - Epoch(train) [9][1300/5005] lr: 1.0000e-04 eta: 0:55:13 time: 0.2573 data_time: 0.0087 memory: 1685 loss: 1.2740 2023/04/14 03:28:03 - mmengine - INFO - Epoch(train) [9][1400/5005] lr: 1.0000e-04 eta: 0:54:33 time: 0.2516 data_time: 0.0082 memory: 1685 loss: 1.1707 2023/04/14 03:28:27 - mmengine - INFO - Epoch(train) [9][1500/5005] lr: 1.0000e-04 eta: 0:53:52 time: 0.2444 data_time: 0.0082 memory: 1685 loss: 1.3027 2023/04/14 03:28:53 - mmengine - INFO - Epoch(train) [9][1600/5005] lr: 1.0000e-04 eta: 0:53:12 time: 0.2466 data_time: 0.0085 memory: 1685 loss: 1.2745 2023/04/14 03:29:17 - mmengine - INFO - Epoch(train) [9][1700/5005] lr: 1.0000e-04 eta: 0:52:31 time: 0.2336 data_time: 0.0082 memory: 1685 loss: 1.3819 2023/04/14 03:29:42 - mmengine - INFO - Epoch(train) [9][1800/5005] lr: 1.0000e-04 eta: 0:51:50 time: 0.2409 data_time: 0.0081 memory: 1685 loss: 1.1652 2023/04/14 03:30:06 - mmengine - INFO - Epoch(train) [9][1900/5005] lr: 1.0000e-04 eta: 0:51:10 time: 0.2316 data_time: 0.0082 memory: 1685 loss: 1.3074 2023/04/14 03:30:20 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 03:30:29 - mmengine - INFO - Epoch(train) [9][2000/5005] lr: 1.0000e-04 eta: 0:50:29 time: 0.2356 data_time: 0.0088 memory: 1685 loss: 1.2647 2023/04/14 03:30:54 - mmengine - INFO - Epoch(train) [9][2100/5005] lr: 1.0000e-04 eta: 0:49:49 time: 0.2334 data_time: 0.0086 memory: 1685 loss: 1.4097 2023/04/14 03:31:19 - mmengine - INFO - Epoch(train) [9][2200/5005] lr: 1.0000e-04 eta: 0:49:09 time: 0.2313 data_time: 0.0083 memory: 1685 loss: 1.2289 2023/04/14 03:31:43 - mmengine - INFO - Epoch(train) [9][2300/5005] lr: 1.0000e-04 eta: 0:48:28 time: 0.2534 data_time: 0.0083 memory: 1685 loss: 1.3046 2023/04/14 03:32:08 - mmengine - INFO - Epoch(train) [9][2400/5005] lr: 1.0000e-04 eta: 0:47:48 time: 0.2422 data_time: 0.0081 memory: 1685 loss: 1.1803 2023/04/14 03:32:32 - mmengine - INFO - Epoch(train) [9][2500/5005] lr: 1.0000e-04 eta: 0:47:08 time: 0.2492 data_time: 0.0084 memory: 1685 loss: 1.5523 2023/04/14 03:32:56 - mmengine - INFO - Epoch(train) [9][2600/5005] lr: 1.0000e-04 eta: 0:46:28 time: 0.2409 data_time: 0.0083 memory: 1685 loss: 1.4387 2023/04/14 03:33:19 - mmengine - INFO - Epoch(train) [9][2700/5005] lr: 1.0000e-04 eta: 0:45:48 time: 0.2334 data_time: 0.0080 memory: 1685 loss: 1.3176 2023/04/14 03:33:44 - mmengine - INFO - Epoch(train) [9][2800/5005] lr: 1.0000e-04 eta: 0:45:08 time: 0.2845 data_time: 0.0079 memory: 1685 loss: 1.1464 2023/04/14 03:34:08 - mmengine - INFO - Epoch(train) [9][2900/5005] lr: 1.0000e-04 eta: 0:44:29 time: 0.2432 data_time: 0.0081 memory: 1685 loss: 1.2485 2023/04/14 03:34:22 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 03:34:32 - mmengine - INFO - Epoch(train) [9][3000/5005] lr: 1.0000e-04 eta: 0:43:49 time: 0.2267 data_time: 0.0080 memory: 1685 loss: 1.1618 2023/04/14 03:34:52 - mmengine - INFO - Epoch(train) [9][3100/5005] lr: 1.0000e-04 eta: 0:43:08 time: 0.2979 data_time: 0.0091 memory: 1685 loss: 1.2052 2023/04/14 03:35:19 - mmengine - INFO - Epoch(train) [9][3200/5005] lr: 1.0000e-04 eta: 0:42:29 time: 0.2342 data_time: 0.0085 memory: 1685 loss: 1.3614 2023/04/14 03:35:43 - mmengine - INFO - Epoch(train) [9][3300/5005] lr: 1.0000e-04 eta: 0:41:50 time: 0.2304 data_time: 0.0091 memory: 1685 loss: 1.4380 2023/04/14 03:36:08 - mmengine - INFO - Epoch(train) [9][3400/5005] lr: 1.0000e-04 eta: 0:41:11 time: 0.2590 data_time: 0.0102 memory: 1685 loss: 1.0822 2023/04/14 03:36:32 - mmengine - INFO - Epoch(train) [9][3500/5005] lr: 1.0000e-04 eta: 0:40:31 time: 0.2256 data_time: 0.0091 memory: 1685 loss: 1.2287 2023/04/14 03:36:56 - mmengine - INFO - Epoch(train) [9][3600/5005] lr: 1.0000e-04 eta: 0:39:52 time: 0.2653 data_time: 0.0088 memory: 1685 loss: 1.3314 2023/04/14 03:37:20 - mmengine - INFO - Epoch(train) [9][3700/5005] lr: 1.0000e-04 eta: 0:39:13 time: 0.2510 data_time: 0.0093 memory: 1685 loss: 1.3794 2023/04/14 03:37:45 - mmengine - INFO - Epoch(train) [9][3800/5005] lr: 1.0000e-04 eta: 0:38:33 time: 0.2354 data_time: 0.0086 memory: 1685 loss: 1.3662 2023/04/14 03:38:10 - mmengine - INFO - Epoch(train) [9][3900/5005] lr: 1.0000e-04 eta: 0:37:55 time: 0.2682 data_time: 0.0090 memory: 1685 loss: 1.2292 2023/04/14 03:38:25 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 03:38:35 - mmengine - INFO - Epoch(train) [9][4000/5005] lr: 1.0000e-04 eta: 0:37:16 time: 0.2320 data_time: 0.0092 memory: 1685 loss: 1.3837 2023/04/14 03:38:59 - mmengine - INFO - Epoch(train) [9][4100/5005] lr: 1.0000e-04 eta: 0:36:37 time: 0.2346 data_time: 0.0087 memory: 1685 loss: 1.2843 2023/04/14 03:39:23 - mmengine - INFO - Epoch(train) [9][4200/5005] lr: 1.0000e-04 eta: 0:35:58 time: 0.2397 data_time: 0.0087 memory: 1685 loss: 1.2841 2023/04/14 03:39:47 - mmengine - INFO - Epoch(train) [9][4300/5005] lr: 1.0000e-04 eta: 0:35:19 time: 0.2420 data_time: 0.0087 memory: 1685 loss: 1.3304 2023/04/14 03:40:11 - mmengine - INFO - Epoch(train) [9][4400/5005] lr: 1.0000e-04 eta: 0:34:40 time: 0.2344 data_time: 0.0082 memory: 1685 loss: 1.2187 2023/04/14 03:40:36 - mmengine - INFO - Epoch(train) [9][4500/5005] lr: 1.0000e-04 eta: 0:34:02 time: 0.2325 data_time: 0.0085 memory: 1685 loss: 1.3233 2023/04/14 03:41:00 - mmengine - INFO - Epoch(train) [9][4600/5005] lr: 1.0000e-04 eta: 0:33:23 time: 0.2500 data_time: 0.0092 memory: 1685 loss: 1.5014 2023/04/14 03:41:24 - mmengine - INFO - Epoch(train) [9][4700/5005] lr: 1.0000e-04 eta: 0:32:44 time: 0.2698 data_time: 0.0085 memory: 1685 loss: 1.4152 2023/04/14 03:41:47 - mmengine - INFO - Epoch(train) [9][4800/5005] lr: 1.0000e-04 eta: 0:32:06 time: 0.2283 data_time: 0.0091 memory: 1685 loss: 1.1714 2023/04/14 03:42:12 - mmengine - INFO - Epoch(train) [9][4900/5005] lr: 1.0000e-04 eta: 0:31:27 time: 0.2329 data_time: 0.0081 memory: 1685 loss: 1.3166 2023/04/14 03:42:26 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 03:42:35 - mmengine - INFO - Epoch(train) [9][5000/5005] lr: 1.0000e-04 eta: 0:30:49 time: 0.2375 data_time: 0.0095 memory: 1685 loss: 1.3228 2023/04/14 03:42:36 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 03:42:36 - mmengine - INFO - Saving checkpoint at 9 epochs 2023/04/14 03:42:53 - mmengine - INFO - Epoch(val) [9][100/196] eta: 0:00:14 time: 0.1129 data_time: 0.0077 memory: 1685 2023/04/14 03:44:01 - mmengine - INFO - Epoch(val) [9][196/196] qat.accuracy/top1: 69.4340 qat.accuracy/top5: 89.0960data_time: 0.0070 time: 0.3009 2023/04/14 03:44:05 - mmengine - INFO - Epoch(val) [9][100/196] eta: 0:00:03 time: 0.0280 data_time: 0.0131 memory: 969 2023/04/14 03:45:02 - mmengine - INFO - Epoch(val) [9][196/196] original.accuracy/top1: 69.3140 original.accuracy/top5: 88.9700data_time: 0.0301 time: 0.0428 2023/04/14 03:45:29 - mmengine - INFO - Epoch(train) [10][ 100/5005] lr: 1.0000e-04 eta: 0:30:09 time: 0.3212 data_time: 0.0087 memory: 1685 loss: 1.2020 2023/04/14 03:45:53 - mmengine - INFO - Epoch(train) [10][ 200/5005] lr: 1.0000e-04 eta: 0:29:31 time: 0.2342 data_time: 0.0086 memory: 1685 loss: 1.1387 2023/04/14 03:46:17 - mmengine - INFO - Epoch(train) [10][ 300/5005] lr: 1.0000e-04 eta: 0:28:52 time: 0.2380 data_time: 0.0085 memory: 1685 loss: 1.3234 2023/04/14 03:46:42 - mmengine - INFO - Epoch(train) [10][ 400/5005] lr: 1.0000e-04 eta: 0:28:14 time: 0.2430 data_time: 0.0092 memory: 1685 loss: 1.3358 2023/04/14 03:47:06 - mmengine - INFO - Epoch(train) [10][ 500/5005] lr: 1.0000e-04 eta: 0:27:36 time: 0.3069 data_time: 0.0084 memory: 1685 loss: 1.3252 2023/04/14 03:47:30 - mmengine - INFO - Epoch(train) [10][ 600/5005] lr: 1.0000e-04 eta: 0:26:58 time: 0.2359 data_time: 0.0083 memory: 1685 loss: 1.2151 2023/04/14 03:47:54 - mmengine - INFO - Epoch(train) [10][ 700/5005] lr: 1.0000e-04 eta: 0:26:20 time: 0.2452 data_time: 0.0082 memory: 1685 loss: 1.3736 2023/04/14 03:48:18 - mmengine - INFO - Epoch(train) [10][ 800/5005] lr: 1.0000e-04 eta: 0:25:43 time: 0.2362 data_time: 0.0082 memory: 1685 loss: 1.2845 2023/04/14 03:48:42 - mmengine - INFO - Epoch(train) [10][ 900/5005] lr: 1.0000e-04 eta: 0:25:05 time: 0.2285 data_time: 0.0082 memory: 1685 loss: 1.3703 2023/04/14 03:48:56 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 03:49:07 - mmengine - INFO - Epoch(train) [10][1000/5005] lr: 1.0000e-04 eta: 0:24:27 time: 0.2369 data_time: 0.0086 memory: 1685 loss: 1.4507 2023/04/14 03:49:32 - mmengine - INFO - Epoch(train) [10][1100/5005] lr: 1.0000e-04 eta: 0:23:49 time: 0.2395 data_time: 0.0197 memory: 1685 loss: 1.2184 2023/04/14 03:49:56 - mmengine - INFO - Epoch(train) [10][1200/5005] lr: 1.0000e-04 eta: 0:23:12 time: 0.2322 data_time: 0.0085 memory: 1685 loss: 0.9990 2023/04/14 03:50:19 - mmengine - INFO - Epoch(train) [10][1300/5005] lr: 1.0000e-04 eta: 0:22:34 time: 0.2309 data_time: 0.0090 memory: 1685 loss: 1.2896 2023/04/14 03:50:44 - mmengine - INFO - Epoch(train) [10][1400/5005] lr: 1.0000e-04 eta: 0:21:57 time: 0.2995 data_time: 0.0086 memory: 1685 loss: 1.1162 2023/04/14 03:51:08 - mmengine - INFO - Epoch(train) [10][1500/5005] lr: 1.0000e-04 eta: 0:21:19 time: 0.2318 data_time: 0.0083 memory: 1685 loss: 1.4340 2023/04/14 03:51:32 - mmengine - INFO - Epoch(train) [10][1600/5005] lr: 1.0000e-04 eta: 0:20:42 time: 0.2299 data_time: 0.0085 memory: 1685 loss: 1.2450 2023/04/14 03:51:55 - mmengine - INFO - Epoch(train) [10][1700/5005] lr: 1.0000e-04 eta: 0:20:04 time: 0.2496 data_time: 0.0083 memory: 1685 loss: 1.3703 2023/04/14 03:52:19 - mmengine - INFO - Epoch(train) [10][1800/5005] lr: 1.0000e-04 eta: 0:19:27 time: 0.2362 data_time: 0.0086 memory: 1685 loss: 1.1424 2023/04/14 03:52:43 - mmengine - INFO - Epoch(train) [10][1900/5005] lr: 1.0000e-04 eta: 0:18:50 time: 0.2312 data_time: 0.0083 memory: 1685 loss: 1.4619 2023/04/14 03:52:56 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 03:53:08 - mmengine - INFO - Epoch(train) [10][2000/5005] lr: 1.0000e-04 eta: 0:18:13 time: 0.2341 data_time: 0.0086 memory: 1685 loss: 1.3034 2023/04/14 03:53:32 - mmengine - INFO - Epoch(train) [10][2100/5005] lr: 1.0000e-04 eta: 0:17:35 time: 0.2367 data_time: 0.0083 memory: 1685 loss: 1.1693 2023/04/14 03:53:56 - mmengine - INFO - Epoch(train) [10][2200/5005] lr: 1.0000e-04 eta: 0:16:58 time: 0.2855 data_time: 0.0094 memory: 1685 loss: 1.1886 2023/04/14 03:54:20 - mmengine - INFO - Epoch(train) [10][2300/5005] lr: 1.0000e-04 eta: 0:16:21 time: 0.2250 data_time: 0.0081 memory: 1685 loss: 1.2931 2023/04/14 03:54:44 - mmengine - INFO - Epoch(train) [10][2400/5005] lr: 1.0000e-04 eta: 0:15:44 time: 0.2331 data_time: 0.0084 memory: 1685 loss: 1.3577 2023/04/14 03:55:08 - mmengine - INFO - Epoch(train) [10][2500/5005] lr: 1.0000e-04 eta: 0:15:07 time: 0.2860 data_time: 0.0088 memory: 1685 loss: 1.1454 2023/04/14 03:55:31 - mmengine - INFO - Epoch(train) [10][2600/5005] lr: 1.0000e-04 eta: 0:14:31 time: 0.2407 data_time: 0.0089 memory: 1685 loss: 1.2980 2023/04/14 03:55:55 - mmengine - INFO - Epoch(train) [10][2700/5005] lr: 1.0000e-04 eta: 0:13:54 time: 0.2331 data_time: 0.0084 memory: 1685 loss: 1.2980 2023/04/14 03:56:19 - mmengine - INFO - Epoch(train) [10][2800/5005] lr: 1.0000e-04 eta: 0:13:17 time: 0.2418 data_time: 0.0082 memory: 1685 loss: 1.1507 2023/04/14 03:56:42 - mmengine - INFO - Epoch(train) [10][2900/5005] lr: 1.0000e-04 eta: 0:12:40 time: 0.2362 data_time: 0.0080 memory: 1685 loss: 1.2387 2023/04/14 03:56:55 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 03:57:06 - mmengine - INFO - Epoch(train) [10][3000/5005] lr: 1.0000e-04 eta: 0:12:04 time: 0.2274 data_time: 0.0089 memory: 1685 loss: 1.3646 2023/04/14 03:57:30 - mmengine - INFO - Epoch(train) [10][3100/5005] lr: 1.0000e-04 eta: 0:11:27 time: 0.2381 data_time: 0.0093 memory: 1685 loss: 1.3006 2023/04/14 03:57:54 - mmengine - INFO - Epoch(train) [10][3200/5005] lr: 1.0000e-04 eta: 0:10:50 time: 0.2535 data_time: 0.0085 memory: 1685 loss: 1.4139 2023/04/14 03:58:18 - mmengine - INFO - Epoch(train) [10][3300/5005] lr: 1.0000e-04 eta: 0:10:14 time: 0.2395 data_time: 0.0085 memory: 1685 loss: 1.0950 2023/04/14 03:58:42 - mmengine - INFO - Epoch(train) [10][3400/5005] lr: 1.0000e-04 eta: 0:09:37 time: 0.2449 data_time: 0.0082 memory: 1685 loss: 1.1673 2023/04/14 03:59:05 - mmengine - INFO - Epoch(train) [10][3500/5005] lr: 1.0000e-04 eta: 0:09:01 time: 0.2544 data_time: 0.0082 memory: 1685 loss: 1.3557 2023/04/14 03:59:29 - mmengine - INFO - Epoch(train) [10][3600/5005] lr: 1.0000e-04 eta: 0:08:25 time: 0.2347 data_time: 0.0085 memory: 1685 loss: 1.4671 2023/04/14 03:59:53 - mmengine - INFO - Epoch(train) [10][3700/5005] lr: 1.0000e-04 eta: 0:07:48 time: 0.2593 data_time: 0.0083 memory: 1685 loss: 1.0446 2023/04/14 04:00:17 - mmengine - INFO - Epoch(train) [10][3800/5005] lr: 1.0000e-04 eta: 0:07:12 time: 0.2350 data_time: 0.0082 memory: 1685 loss: 1.2064 2023/04/14 04:00:41 - mmengine - INFO - Epoch(train) [10][3900/5005] lr: 1.0000e-04 eta: 0:06:36 time: 0.2319 data_time: 0.0082 memory: 1685 loss: 1.0286 2023/04/14 04:00:54 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 04:01:05 - mmengine - INFO - Epoch(train) [10][4000/5005] lr: 1.0000e-04 eta: 0:06:00 time: 0.2242 data_time: 0.0090 memory: 1685 loss: 1.3729 2023/04/14 04:01:28 - mmengine - INFO - Epoch(train) [10][4100/5005] lr: 1.0000e-04 eta: 0:05:24 time: 0.2408 data_time: 0.0085 memory: 1685 loss: 1.3022 2023/04/14 04:01:52 - mmengine - INFO - Epoch(train) [10][4200/5005] lr: 1.0000e-04 eta: 0:04:48 time: 0.2317 data_time: 0.0083 memory: 1685 loss: 1.3019 2023/04/14 04:02:16 - mmengine - INFO - Epoch(train) [10][4300/5005] lr: 1.0000e-04 eta: 0:04:12 time: 0.2369 data_time: 0.0082 memory: 1685 loss: 1.1631 2023/04/14 04:02:41 - mmengine - INFO - Epoch(train) [10][4400/5005] lr: 1.0000e-04 eta: 0:03:36 time: 0.2434 data_time: 0.0089 memory: 1685 loss: 1.3754 2023/04/14 04:03:05 - mmengine - INFO - Epoch(train) [10][4500/5005] lr: 1.0000e-04 eta: 0:03:00 time: 0.2307 data_time: 0.0086 memory: 1685 loss: 1.3687 2023/04/14 04:03:30 - mmengine - INFO - Epoch(train) [10][4600/5005] lr: 1.0000e-04 eta: 0:02:24 time: 0.2343 data_time: 0.0082 memory: 1685 loss: 1.2422 2023/04/14 04:03:54 - mmengine - INFO - Epoch(train) [10][4700/5005] lr: 1.0000e-04 eta: 0:01:48 time: 0.2387 data_time: 0.0081 memory: 1685 loss: 1.2568 2023/04/14 04:04:19 - mmengine - INFO - Epoch(train) [10][4800/5005] lr: 1.0000e-04 eta: 0:01:13 time: 0.2536 data_time: 0.0080 memory: 1685 loss: 1.4088 2023/04/14 04:04:45 - mmengine - INFO - Epoch(train) [10][4900/5005] lr: 1.0000e-04 eta: 0:00:37 time: 0.2345 data_time: 0.0082 memory: 1685 loss: 1.2439 2023/04/14 04:04:57 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 04:05:08 - mmengine - INFO - Epoch(train) [10][5000/5005] lr: 1.0000e-04 eta: 0:00:01 time: 0.2339 data_time: 0.0091 memory: 1685 loss: 1.1030 2023/04/14 04:05:09 - mmengine - INFO - Exp name: lsq_openvino_resnet18_8xb32_in1k_20230413_224237 2023/04/14 04:05:10 - mmengine - INFO - Saving checkpoint at 10 epochs 2023/04/14 04:05:26 - mmengine - INFO - Epoch(val) [10][100/196] eta: 0:00:13 time: 0.0994 data_time: 0.0082 memory: 1685 2023/04/14 04:06:33 - mmengine - INFO - Epoch(val) [10][196/196] qat.accuracy/top1: 69.3400 qat.accuracy/top5: 89.0460data_time: 0.0071 time: 0.1986 2023/04/14 04:06:37 - mmengine - INFO - Epoch(val) [10][100/196] eta: 0:00:03 time: 0.0236 data_time: 0.0111 memory: 969 2023/04/14 04:07:33 - mmengine - INFO - Epoch(val) [10][196/196] original.accuracy/top1: 69.1080 original.accuracy/top5: 88.9460data_time: 0.0340 time: 0.0481