Replies: 2 comments 4 replies
-
|
Everything seems to be right, and the loss keeps decreasing over time but way too slowly. A general advice would be to try a larger learning rate. |
Beta Was this translation helpful? Give feedback.
-
|
thanks. |
Beta Was this translation helpful? Give feedback.
-
|
If I want to train for 100 epochs, is this learning rate change appropriate? _base_.optim_wrapper.optimizer.lr = 0.002
param_scheduler = [
# learning rate scheduler
# During the first 10 epochs, learning rate increases from 0 to lr * 10
# during the next 90 epochs, learning rate decreases from lr * 10 to lr * 1e-4
dict(
type='CosineAnnealingLR',
T_max=10,
eta_min=_base_.optim_wrapper.optimizer.lr * 10,
begin=0,
end=10,
by_epoch=True,
convert_to_iter_based=True),
dict(
type='CosineAnnealingLR',
T_max=90,
eta_min=_base_.optim_wrapper.optimizer.lr * 1e-4,
begin=10,
end=100,
by_epoch=True,
convert_to_iter_based=True),
# momentum scheduler
# During the first 10 epochs, momentum increases from 0 to 0.85 / 0.95
# during the next 90 epochs, momentum increases from 0.85 / 0.95 to 1
dict(
type='CosineAnnealingMomentum',
T_max=10,
eta_min=0.85 / 0.95,
begin=0,
end=10,
by_epoch=True,
convert_to_iter_based=True),
dict(
type='CosineAnnealingMomentum',
T_max=90,
eta_min=1,
begin=10,
end=100,
by_epoch=True,
convert_to_iter_based=True)
] |
Beta Was this translation helpful? Give feedback.
-
|
Nobody can tell whether a hyperparameter is "appropriate" without experiment results... But if you want to know whether such a schedule works as expected, you can adapt this script to mmocr for the visualization. |
Beta Was this translation helpful? Give feedback.
-
|
I have another question |
Beta Was this translation helpful? Give feedback.
-
|
The second one |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
Hello, Community Developers.
I'm having some issues training my own text detection model. I would be very grateful if you could bring some advice.
The following is my process:
First, I changed the dataset to the form of the sample in the official document
{'metainfo': {'dataset_type': 'BenetechMakingGraphsAccessible', 'task_name': 'textdet', 'category': [{'id': 0, 'name': 'text'}]}, 'data_list': [{'img_path': '/home/.../**.jpg', 'height': 523, 'width': 320, 'instances': [{'bbox': [45, 5, 411, 34], 'bbox_label': 0, 'polygon': [45, 5, 411, 5, 411, 31, 60, 34], 'text': 'Rural population long-run future projection in Wallis and Futuna', 'ignore': False}, ...], ...], }Then I modified the following configuration:
Modify my data address very name
In this configuration, I set epoch to 120
Finally, I ran the following command
The training result I got
2023/04/03 12:39:41 - mmengine - INFO - ------------------------------------------------------------ System environment: sys.platform: linux Python: 3.7.10 (default, Feb 26 2021, 18:47:35) [GCC 7.3.0] CUDA available: True numpy_random_seed: 443960725 GPU 0,1,2,3,4,5: NVIDIA GeForce RTX 3080 CUDA_HOME: /usr/local/cuda NVCC: Cuda compilation tools, release 11.1, V11.1.105 GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 PyTorch: 1.9.0 PyTorch compiling details: PyTorch built with: - GCC 7.3 - C++ Version: 201402 - Intel(R) oneAPI Math Kernel Library Version 2021.2-Product Build 20210312 for Intel(R) 64 architecture applications - Intel(R) MKL-DNN v2.1.2 (Git Hash 98be7e8afa711dc9b66c8ff3504129cb82013cdb) - OpenMP 201511 (a.k.a. OpenMP 4.5) - NNPACK is enabled - CPU capability usage: AVX2 - CUDA Runtime 11.1 - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 - CuDNN 8.0.5 - Magma 2.5.2 - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.1, CUDNN_VERSION=8.0.5, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-variable -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.9.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, TorchVision: 0.10.0 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 Distributed launcher: pytorch Distributed training: True GPU number: 6 ------------------------------------------------------------ 2023/04/03 12:39:41 - mmengine - INFO - Config: model = dict( type='DBNet', backbone=dict( type='CLIPResNet', init_cfg=dict( type='Pretrained', checkpoint= 'https://download.openmmlab.com/mmocr/backbone/resnet50-oclip-7ba0c533.pth' )), neck=dict( type='FPNC', in_channels=[256, 512, 1024, 2048], lateral_channels=256, asf_cfg=dict(attention_type='ScaleChannelSpatial')), det_head=dict( type='DBHead', in_channels=256, module_loss=dict(type='DBModuleLoss'), postprocessor=dict( type='DBPostprocessor', text_repr_type='quad', epsilon_ratio=0.002)), data_preprocessor=dict( type='TextDetDataPreprocessor', mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], bgr_to_rgb=True, pad_size_divisor=32)) train_pipeline = [ dict(type='LoadImageFromFile', color_type='color_ignore_orientation'), dict( type='LoadOCRAnnotations', with_bbox=True, with_polygon=True, with_label=True), dict( type='TorchVisionWrapper', op='ColorJitter', brightness=0.12549019607843137, saturation=0.5), dict( type='ImgAugWrapper', args=[['Fliplr', 0.5], { 'cls': 'Affine', 'rotate': [-10, 10] }, ['Resize', [0.5, 3.0]]]), dict(type='RandomCrop', min_side_ratio=0.1), dict(type='Resize', scale=(640, 640), keep_ratio=True), dict(type='Pad', size=(640, 640)), dict( type='PackTextDetInputs', meta_keys=('img_path', 'ori_shape', 'img_shape')) ] test_pipeline = [ dict(type='LoadImageFromFile', color_type='color_ignore_orientation'), dict(type='Resize', scale=(4068, 1024), keep_ratio=True), dict( type='LoadOCRAnnotations', with_polygon=True, with_bbox=True, with_label=True), dict( type='PackTextDetInputs', meta_keys=('img_path', 'ori_shape', 'img_shape', 'scale_factor', 'instances')) ] default_scope = 'mmocr' env_cfg = dict( cudnn_benchmark=False, mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0), dist_cfg=dict(backend='nccl')) randomness = dict(seed=None) default_hooks = dict( timer=dict(type='IterTimerHook'), logger=dict(type='LoggerHook', interval=1), param_scheduler=dict(type='ParamSchedulerHook'), checkpoint=dict(type='CheckpointHook', interval=2), sampler_seed=dict(type='DistSamplerSeedHook'), sync_buffer=dict(type='SyncBuffersHook'), visualization=dict( type='VisualizationHook', interval=1, enable=False, show=False, draw_gt=False, draw_pred=False)) log_level = 'INFO' log_processor = dict(type='LogProcessor', window_size=10, by_epoch=True) load_from = None resume = False val_evaluator = dict(type='HmeanIOUMetric') test_evaluator = dict(type='HmeanIOUMetric') vis_backends = [dict(type='LocalVisBackend')] visualizer = dict( type='TextDetLocalVisualizer', name='visualizer', vis_backends=[dict(type='LocalVisBackend')]) icdar2015_textdet_data_root = '/home/mydata/input/' icdar2015_textdet_train = dict( type='OCRDataset', data_root='/home/mydata/input/', ann_file='ocr_train.json', filter_cfg=dict(filter_empty_gt=True, min_size=16), pipeline=None) icdar2015_textdet_test = dict( type='OCRDataset', data_root='/home/mydata/input/', ann_file='ocr_test.json', test_mode=True, pipeline=None) optim_wrapper = dict( type='OptimWrapper', optimizer=dict(type='SGD', lr=0.002, momentum=0.9, weight_decay=0.0001)) train_cfg = dict(type='EpochBasedTrainLoop', max_epochs=120, val_interval=1) val_cfg = dict(type='ValLoop') test_cfg = dict(type='TestLoop') param_scheduler = [ dict(type='LinearLR', end=200, start_factor=0.001), dict(type='PolyLR', power=0.9, eta_min=1e-07, begin=200, end=1200) ] train_list = [ dict( type='OCRDataset', data_root='/home/mydata/input/', ann_file='ocr_train.json', filter_cfg=dict(filter_empty_gt=True, min_size=16), pipeline=None) ] test_list = [ dict( type='OCRDataset', data_root='/home/mydata/input/', ann_file='ocr_test.json', test_mode=True, pipeline=None) ] train_dataloader = dict( batch_size=4, num_workers=16, persistent_workers=True, sampler=dict(type='DefaultSampler', shuffle=True), dataset=dict( type='ConcatDataset', datasets=[ dict( type='OCRDataset', data_root='/home/mydata/input/', ann_file='ocr_train.json', filter_cfg=dict(filter_empty_gt=True, min_size=16), pipeline=None) ], pipeline=[ dict( type='LoadImageFromFile', color_type='color_ignore_orientation'), dict( type='LoadOCRAnnotations', with_bbox=True, with_polygon=True, with_label=True), dict( type='TorchVisionWrapper', op='ColorJitter', brightness=0.12549019607843137, saturation=0.5), dict( type='ImgAugWrapper', args=[['Fliplr', 0.5], { 'cls': 'Affine', 'rotate': [-10, 10] }, ['Resize', [0.5, 3.0]]]), dict(type='RandomCrop', min_side_ratio=0.1), dict(type='Resize', scale=(640, 640), keep_ratio=True), dict(type='Pad', size=(640, 640)), dict( type='PackTextDetInputs', meta_keys=('img_path', 'ori_shape', 'img_shape')) ])) val_dataloader = dict( batch_size=4, num_workers=8, persistent_workers=True, sampler=dict(type='DefaultSampler', shuffle=False), dataset=dict( type='ConcatDataset', datasets=[ dict( type='OCRDataset', data_root='/home/mydata/input/', ann_file='ocr_test.json', test_mode=True, pipeline=None) ], pipeline=[ dict( type='LoadImageFromFile', color_type='color_ignore_orientation'), dict(type='Resize', scale=(4068, 1024), keep_ratio=True), dict( type='LoadOCRAnnotations', with_polygon=True, with_bbox=True, with_label=True), dict( type='PackTextDetInputs', meta_keys=('img_path', 'ori_shape', 'img_shape', 'scale_factor', 'instances')) ])) test_dataloader = dict( batch_size=4, num_workers=8, persistent_workers=True, sampler=dict(type='DefaultSampler', shuffle=False), dataset=dict( type='ConcatDataset', datasets=[ dict( type='OCRDataset', data_root='/home/mydata/input/', ann_file='ocr_test.json', test_mode=True, pipeline=None) ], pipeline=[ dict( type='LoadImageFromFile', color_type='color_ignore_orientation'), dict(type='Resize', scale=(4068, 1024), keep_ratio=True), dict( type='LoadOCRAnnotations', with_polygon=True, with_bbox=True, with_label=True), dict( type='PackTextDetInputs', meta_keys=('img_path', 'ori_shape', 'img_shape', 'scale_factor', 'instances')) ])) auto_scale_lr = dict(base_batch_size=4) launcher = 'pytorch' work_dir = './work_dirs/dbnetpp_resnet50-oclip_fpnc_1200e_icdar2015' 2023/04/03 12:39:43 - 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 -------------------- 2023/04/03 12:39:43 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io 2023/04/03 12:39:43 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. 2023/04/03 12:39:54 - mmengine - INFO - load model from: https://download.openmmlab.com/mmocr/backbone/resnet50-oclip-7ba0c533.pth 2023/04/03 12:39:54 - mmengine - INFO - Loads checkpoint by http backend from path: https://download.openmmlab.com/mmocr/backbone/resnet50-oclip-7ba0c533.pth Name of parameter - Initialization information .... The value is the same before and after calling `init_weights` of DBNet 2023/04/03 12:39:56 - mmengine - INFO - Checkpoints will be saved to /mmocr/work_dirs/dbnetpp_resnet50-oclip_fpnc_1200e_icdar2015. 2023/04/03 12:40:00 - mmengine - INFO - Epoch(train) [1][ 1/2146] lr: 2.0000e-06 eta: 11 days, 21:32:08 time: 3.9917 data_time: 2.1671 memory: 4524 loss: 17.5933 loss_prob: 13.7704 loss_thr: 2.8410 loss_db: 0.9819 2023/04/03 12:40:00 - mmengine - INFO - Epoch(train) [1][ 2/2146] lr: 2.0000e-06 eta: 6 days, 14:32:17 time: 2.2163 data_time: 1.1019 memory: 4729 loss: 17.3339 loss_prob: 13.4220 loss_thr: 2.9319 loss_db: 0.9800 2023/04/03 12:40:01 - mmengine - INFO - Epoch(train) [1][ 3/2146] lr: 2.0000e-06 eta: 4 days, 19:42:40 time: 1.6176 data_time: 0.7355 memory: 4729 loss: 16.4382 loss_prob: 12.5648 loss_thr: 2.9031 loss_db: 0.9703 2023/04/03 12:40:01 - mmengine - INFO - Epoch(train) [1][ 4/2146] lr: 2.0000e-06 eta: 3 days, 21:24:28 time: 1.3058 data_time: 0.5555 memory: 4729 loss: 16.3932 loss_prob: 12.5339 loss_thr: 2.8904 loss_db: 0.9688 2023/04/03 12:40:02 - mmengine - INFO - Epoch(train) [1][ 5/2146] lr: 2.0000e-06 eta: 3 days, 9:00:22 time: 1.1324 data_time: 0.4641 memory: 4729 loss: 16.7636 loss_prob: 12.8642 loss_thr: 2.9285 loss_db: 0.9709 2023/04/03 12:40:02 - mmengine - INFO - Epoch(train) [1][ 6/2146] lr: 2.0000e-06 eta: 3 days, 0:18:45 time: 1.0109 data_time: 0.3872 memory: 4729 loss: 16.8796 loss_prob: 12.9825 loss_thr: 2.9252 loss_db: 0.9719 2023/04/03 12:40:02 - mmengine - INFO - Epoch(train) [1][ 7/2146] lr: 2.0000e-06 eta: 2 days, 18:47:17 time: 0.9337 data_time: 0.3323 memory: 4729 loss: 16.9751 loss_prob: 13.0699 loss_thr: 2.9334 loss_db: 0.9718 2023/04/03 12:40:03 - mmengine - INFO - Epoch(train) [1][ 8/2146] lr: 2.0000e-06 eta: 2 days, 13:50:06 time: 0.8645 data_time: 0.2911 memory: 4729 loss: 16.9655 loss_prob: 13.0505 loss_thr: 2.9429 loss_db: 0.9721 2023/04/03 12:40:03 - mmengine - INFO - Epoch(train) [1][ 9/2146] lr: 2.0000e-06 eta: 2 days, 10:02:16 time: 0.8114 data_time: 0.2591 memory: 4729 loss: 17.0130 loss_prob: 13.0781 loss_thr: 2.9625 loss_db: 0.9724 2023/04/03 12:40:04 - mmengine - INFO - Epoch(train) [1][ 10/2146] lr: 2.0000e-06 eta: 2 days, 7:17:47 time: 0.7730 data_time: 0.2334 memory: 4729 loss: 17.2032 loss_prob: 13.2335 loss_thr: 2.9959 loss_db: 0.9738 2023/04/03 12:40:04 - mmengine - INFO - Epoch(train) [1][ 11/2146] lr: 2.0000e-06 eta: 2 days, 5:00:43 time: 0.4161 data_time: 0.0170 memory: 4729 loss: 17.2217 loss_prob: 13.2229 loss_thr: 3.0248 loss_db: 0.9740 2023/04/03 12:40:04 - mmengine - INFO - Epoch(train) [1][ 12/2146] lr: 2.0000e-06 eta: 2 days, 2:47:12 time: 0.4087 data_time: 0.0136 memory: 4729 loss: 17.2021 loss_prob: 13.2003 loss_thr: 3.0279 loss_db: 0.9739 2023/04/03 12:40:05 - mmengine - INFO - Epoch(train) [1][ 13/2146] lr: 2.0000e-06 eta: 2 days, 1:12:29 time: 0.4090 data_time: 0.0136 memory: 4729 loss: 17.3482 loss_prob: 13.3327 loss_thr: 3.0398 loss_db: 0.9757 2023/04/03 12:40:05 - mmengine - INFO - Epoch(train) [1][ 14/2146] lr: 2.0000e-06 eta: 1 day, 23:34:30 time: 0.4088 data_time: 0.0124 memory: 4729 loss: 17.2537 loss_prob: 13.2340 loss_thr: 3.0445 loss_db: 0.9752 2023/04/03 12:40:06 - mmengine - INFO - Epoch(train) [1][ 15/2146] lr: 2.0000e-06 eta: 1 day, 22:14:03 time: 0.4033 data_time: 0.0028 memory: 4729 loss: 17.0010 loss_prob: 13.0152 loss_thr: 3.0114 loss_db: 0.9743 2023/04/03 12:40:06 - mmengine - INFO - Epoch(train) [1][ 16/2146] lr: 2.0000e-06 eta: 1 day, 21:15:37 time: 0.4059 data_time: 0.0064 memory: 4729 loss: 16.7903 loss_prob: 12.8051 loss_thr: 3.0123 loss_db: 0.9729 2023/04/03 12:40:06 - mmengine - INFO - Epoch(train) [1][ 17/2146] lr: 2.0000e-06 eta: 1 day, 20:37:03 time: 0.4068 data_time: 0.0064 memory: 4729 loss: 16.4949 loss_prob: 12.5287 loss_thr: 2.9957 loss_db: 0.9704 2023/04/03 12:40:07 - mmengine - INFO - Epoch(train) [1][ 18/2146] lr: 2.0000e-06 eta: 1 day, 19:50:29 time: 0.4117 data_time: 0.0101 memory: 4729 loss: 16.4251 loss_prob: 12.4622 loss_thr: 2.9924 loss_db: 0.9706 2023/04/03 12:40:07 - mmengine - INFO - Epoch(train) [1][ 19/2146] lr: 2.0000e-06 eta: 1 day, 19:19:25 time: 0.4206 data_time: 0.0101 memory: 4729 loss: 16.1116 loss_prob: 12.2047 loss_thr: 2.9379 loss_db: 0.9690 2023/04/03 12:40:08 - mmengine - INFO - Epoch(train) [1][ 20/2146] lr: 2.0000e-06 eta: 1 day, 18:34:51 time: 0.4176 data_time: 0.0112 memory: 4729 loss: 15.8800 loss_prob: 12.0030 loss_thr: 2.9089 loss_db: 0.9681 2023/04/03 12:40:08 - mmengine - INFO - Epoch(train) [1][ 21/2146] lr: 2.0000e-06 eta: 1 day, 18:02:13 time: 0.4190 data_time: 0.0211 memory: 4729 loss: 15.5596 loss_prob: 11.7250 loss_thr: 2.8684 loss_db: 0.9662 2023/04/03 12:40:09 - mmengine - INFO - Epoch(train) [1][ 22/2146] lr: 2.0000e-06 eta: 1 day, 17:26:42 time: 0.4227 data_time: 0.0211 memory: 4729 loss: 15.4810 loss_prob: 11.6443 loss_thr: 2.8705 loss_db: 0.9662 2023/04/03 12:40:09 - mmengine - INFO - Epoch(train) [1][ 23/2146] lr: 2.0000e-06 eta: 1 day, 17:01:04 time: 0.4246 data_time: 0.0211 memory: 4729 loss: 15.4489 loss_prob: 11.6112 loss_thr: 2.8707 loss_db: 0.9670 2023/04/03 12:40:09 - mmengine - INFO - Epoch(train) [1][ 24/2146] lr: 2.0000e-06 eta: 1 day, 16:27:45 time: 0.4265 data_time: 0.0211 memory: 4729 loss: 15.4141 loss_prob: 11.5909 loss_thr: 2.8550 loss_db: 0.9683 2023/04/03 12:40:10 - mmengine - INFO - Epoch(train) [1][ 25/2146] lr: 2.0000e-06 eta: 1 day, 15:59:53 time: 0.4285 data_time: 0.0211 memory: 4729 loss: 15.3092 loss_prob: 11.4734 loss_thr: 2.8674 loss_db: 0.9684 2023/04/03 12:40:10 - mmengine - INFO - Epoch(train) [1][ 26/2146] lr: 2.0000e-06 eta: 1 day, 15:38:22 time: 0.4285 data_time: 0.0175 memory: 4729 loss: 15.2319 loss_prob: 11.4158 loss_thr: 2.8467 loss_db: 0.9694 2023/04/03 12:40:11 - mmengine - INFO - Epoch(train) [1][ 27/2146] lr: 2.0000e-06 eta: 1 day, 15:14:28 time: 0.4209 data_time: 0.0175 memory: 4729 loss: 15.2968 loss_prob: 11.4776 loss_thr: 2.8462 loss_db: 0.9730 2023/04/03 12:40:11 - mmengine - INFO - Epoch(train) [1][ 28/2146] lr: 2.0000e-06 eta: 1 day, 14:50:42 time: 0.4174 data_time: 0.0138 memory: 4729 loss: 15.1300 loss_prob: 11.3416 loss_thr: 2.8156 loss_db: 0.9728 2023/04/03 12:40:11 - mmengine - INFO - Epoch(train) [1][ 29/2146] lr: 2.0000e-06 eta: 1 day, 14:27:54 time: 0.4088 data_time: 0.0138 memory: 4729 loss: 15.0930 loss_prob: 11.3071 loss_thr: 2.8124 loss_db: 0.9735 2023/04/03 12:40:12 - mmengine - INFO - Epoch(train) [1][ 30/2146] lr: 2.0000e-06 eta: 1 day, 14:09:02 time: 0.4096 data_time: 0.0128 memory: 4729 loss: 14.9176 loss_prob: 11.1351 loss_thr: 2.8091 loss_db: 0.9734 2023/04/03 12:40:12 - mmengine - INFO - Epoch(train) [1][ 31/2146] lr: 2.0000e-06 eta: 1 day, 13:52:44 time: 0.4076 data_time: 0.0028 memory: 4729 loss: 14.8606 loss_prob: 11.0955 loss_thr: 2.7906 loss_db: 0.9746 2023/04/03 12:40:13 - mmengine - INFO - Epoch(train) [1][ 32/2146] lr: 2.0000e-06 eta: 1 day, 13:41:27 time: 0.4115 data_time: 0.0061 memory: 4729 loss: 14.6614 loss_prob: 10.9282 loss_thr: 2.7597 loss_db: 0.9735 2023/04/03 12:40:13 - mmengine - INFO - Epoch(train) [1][ 33/2146] lr: 2.0000e-06 eta: 1 day, 13:29:28 time: 0.4108 data_time: 0.0061 memory: 4729 loss: 14.4315 loss_prob: 10.7155 loss_thr: 2.7444 loss_db: 0.9716 2023/04/03 12:40:14 - mmengine - INFO - Epoch(train) [1][ 34/2146] lr: 2.0000e-06 eta: 1 day, 13:23:06 time: 0.4195 data_time: 0.0092 memory: 4729 loss: 14.0829 loss_prob: 10.4022 loss_thr: 2.7145 loss_db: 0.9662 2023/04/03 12:40:14 - mmengine - INFO - Epoch(train) [1][ 35/2146] lr: 2.0000e-06 eta: 1 day, 13:11:39 time: 0.4221 data_time: 0.0092 memory: 4729 loss: 14.0465 loss_prob: 10.3682 loss_thr: 2.7116 loss_db: 0.9667 2023/04/03 12:40:14 - mmengine - INFO - Epoch(train) [1][ 36/2146] lr: 2.0000e-06 eta: 1 day, 12:54:07 time: 0.4165 data_time: 0.0101 memory: 4729 loss: 13.9724 loss_prob: 10.2957 loss_thr: 2.7099 loss_db: 0.9668 2023/04/03 12:40:15 - mmengine - INFO - Epoch(train) [1][ 37/2146] lr: 2.0000e-06 eta: 1 day, 12:46:41 time: 0.4213 data_time: 0.0189 memory: 4729 loss: 13.8134 loss_prob: 10.1524 loss_thr: 2.6946 loss_db: 0.9664 2023/04/03 12:40:15 - mmengine - INFO - Epoch(train) [1][ 38/2146] lr: 2.0000e-06 eta: 1 day, 12:36:29 time: 0.4243 data_time: 0.0188 memory: 4729 loss: 13.7975 loss_prob: 10.1175 loss_thr: 2.7131 loss_db: 0.9670 2023/04/03 12:40:16 - mmengine - INFO - Epoch(train) [1][ 39/2146] lr: 2.0000e-06 eta: 1 day, 12:27:43 time: 0.4286 data_time: 0.0188 memory: 4729 loss: 13.6821 loss_prob: 10.0080 loss_thr: 2.7076 loss_db: 0.9665 2023/04/03 12:40:16 - mmengine - INFO - Epoch(train) [1][ 40/2146] lr: 2.0000e-06 eta: 1 day, 12:20:35 time: 0.4324 data_time: 0.0189 memory: 4729 loss: 13.5460 loss_prob: 9.8949 loss_thr: 2.6852 loss_db: 0.9660 2023/04/03 12:40:17 - mmengine - INFO - Epoch(train) [1][ 41/2146] lr: 2.0000e-06 eta: 1 day, 12:07:59 time: 0.4296 data_time: 0.0189 memory: 4729 loss: 13.4187 loss_prob: 9.7584 loss_thr: 2.6958 loss_db: 0.9645 2023/04/03 12:40:17 - mmengine - INFO - Epoch(train) [1][ 42/2146] lr: 2.0000e-06 eta: 1 day, 11:56:19 time: 0.4242 data_time: 0.0156 memory: 4729 loss: 13.2822 loss_prob: 9.6402 loss_thr: 2.6779 loss_db: 0.9642 2023/04/03 12:40:17 - mmengine - INFO - Epoch(train) [1][ 43/2146] lr: 2.0000e-06 eta: 1 day, 11:45:44 time: 0.4203 data_time: 0.0157 memory: 4729 loss: 13.2092 loss_prob: 9.5797 loss_thr: 2.6648 loss_db: 0.9646 2023/04/03 12:40:18 - mmengine - INFO - Epoch(train) [1][ 44/2146] lr: 2.0000e-06 eta: 1 day, 11:36:35 time: 0.4136 data_time: 0.0126 memory: 4729 loss: 13.3232 loss_prob: 9.6949 loss_thr: 2.6588 loss_db: 0.9696 2023/04/03 12:40:18 - mmengine - INFO - Epoch(train) [1][ 45/2146] lr: 2.0000e-06 eta: 1 day, 11:28:35 time: 0.4120 data_time: 0.0127 memory: 4729 loss: 13.1664 loss_prob: 9.5651 loss_thr: 2.6333 loss_db: 0.9680 2023/04/03 12:40:19 - mmengine - INFO - Epoch(train) [1][ 46/2146] lr: 2.0000e-06 eta: 1 day, 11:23:56 time: 0.4194 data_time: 0.0117 memory: 4729 loss: 12.9890 loss_prob: 9.4148 loss_thr: 2.6083 loss_db: 0.9658 2023/04/03 12:40:19 - mmengine - INFO - Epoch(train) [1][ 47/2146] lr: 2.0000e-06 eta: 1 day, 11:19:57 time: 0.4193 data_time: 0.0030 memory: 4729 loss: 12.9129 loss_prob: 9.3325 loss_thr: 2.6157 loss_db: 0.9647 2023/04/03 12:40:20 - mmengine - INFO - Epoch(train) [1][ 48/2146] lr: 2.0000e-06 eta: 1 day, 11:15:23 time: 0.4212 data_time: 0.0069 memory: 4729 loss: 12.8005 loss_prob: 9.2502 loss_thr: 2.5854 loss_db: 0.9649 2023/04/03 12:40:20 - mmengine - INFO - Epoch(train) [1][ 49/2146] lr: 2.0000e-06 eta: 1 day, 11:07:15 time: 0.4180 data_time: 0.0070 memory: 4729 loss: 12.9727 loss_prob: 9.3810 loss_thr: 2.6244 loss_db: 0.9672 2023/04/03 12:40:20 - mmengine - INFO - Epoch(train) [1][ 50/2146] lr: 2.0000e-06 eta: 1 day, 11:01:37 time: 0.4162 data_time: 0.0103 memory: 4729 loss: 12.8787 loss_prob: 9.2965 loss_thr: 2.6164 loss_db: 0.9658 2023/04/03 12:40:21 - mmengine - INFO - Epoch(train) [1][ 51/2146] lr: 2.0000e-06 eta: 1 day, 11:00:43 time: 0.4254 data_time: 0.0104 memory: 4729 loss: 12.8984 loss_prob: 9.3177 loss_thr: 2.6130 loss_db: 0.9678 2023/04/03 12:40:21 - mmengine - INFO - Epoch(train) [1][ 52/2146] lr: 2.0000e-06 eta: 1 day, 10:52:33 time: 0.4253 data_time: 0.0113 memory: 4729 loss: 12.9383 loss_prob: 9.3330 loss_thr: 2.6362 loss_db: 0.9690 2023/04/03 12:40:22 - mmengine - INFO - Epoch(train) [1][ 53/2146] lr: 2.0000e-06 eta: 1 day, 10:48:16 time: 0.4292 data_time: 0.0200 memory: 4729 loss: 12.8859 loss_prob: 9.3131 loss_thr: 2.6031 loss_db: 0.9698 2023/04/03 12:40:22 - mmengine - INFO - Epoch(train) [1][ 54/2146] lr: 2.0000e-06 eta: 1 day, 10:40:57 time: 0.4280 data_time: 0.0199 memory: 4729 loss: 12.8684 loss_prob: 9.2851 loss_thr: 2.6124 loss_db: 0.9708 2023/04/03 12:40:22 - mmengine - INFO - Epoch(train) [1][ 55/2146] lr: 2.0000e-06 eta: 1 day, 10:33:16 time: 0.4252 data_time: 0.0199 memory: 4729 loss: 12.7432 loss_prob: 9.1805 loss_thr: 2.5938 loss_db: 0.9689 2023/04/03 12:40:23 - mmengine - INFO - Epoch(train) [1][ 56/2146] lr: 2.0000e-06 eta: 1 day, 10:31:06 time: 0.4261 data_time: 0.0199 memory: 4729 loss: 12.7754 loss_prob: 9.2021 loss_thr: 2.6025 loss_db: 0.9708 2023/04/03 12:40:23 - mmengine - INFO - Epoch(train) [1][ 57/2146] lr: 2.0000e-06 eta: 1 day, 10:24:11 time: 0.4200 data_time: 0.0198 memory: 4729 loss: 12.6641 loss_prob: 9.1277 loss_thr: 2.5653 loss_db: 0.9711 2023/04/03 12:40:24 - mmengine - INFO - Epoch(train) [1][ 58/2146] lr: 2.0000e-06 eta: 1 day, 10:15:21 time: 0.4119 data_time: 0.0159 memory: 4729 loss: 12.6623 loss_prob: 9.0972 loss_thr: 2.5939 loss_db: 0.9712 2023/04/03 12:40:24 - mmengine - INFO - Epoch(train) [1][ 59/2146] lr: 2.0000e-06 eta: 1 day, 10:09:20 time: 0.4115 data_time: 0.0158 memory: 4729 loss: 12.4441 loss_prob: 8.9363 loss_thr: 2.5378 loss_db: 0.9699 2023/04/03 12:40:24 - mmengine - INFO - Epoch(train) [1][ 60/2146] lr: 2.0000e-06 eta: 1 day, 10:04:18 time: 0.4097 data_time: 0.0124 memory: 4729 loss: 12.2178 loss_prob: 8.7688 loss_thr: 2.4842 loss_db: 0.9647 2023/04/03 12:40:25 - mmengine - INFO - Epoch(train) [1][ 61/2146] lr: 2.0000e-06 eta: 1 day, 10:00:40 time: 0.4043 data_time: 0.0123 memory: 4729 loss: 12.1250 loss_prob: 8.6989 loss_thr: 2.4620 loss_db: 0.9640 2023/04/03 12:40:25 - mmengine - INFO - Epoch(train) [1][ 62/2146] lr: 2.0000e-06 eta: 1 day, 9:56:36 time: 0.4069 data_time: 0.0113 memory: 4729 loss: 11.9947 loss_prob: 8.5927 loss_thr: 2.4392 loss_db: 0.9628 2023/04/03 12:40:26 - mmengine - INFO - Epoch(train) [1][ 63/2146] lr: 2.0000e-06 eta: 1 day, 9:49:33 time: 0.4006 data_time: 0.0025 memory: 4729 loss: 11.9564 loss_prob: 8.5315 loss_thr: 2.4628 loss_db: 0.9621 2023/04/03 12:40:26 - mmengine - INFO - Epoch(train) [1][ 64/2146] lr: 2.0000e-06 eta: 1 day, 9:46:49 time: 0.4043 data_time: 0.0055 memory: 4729 loss: 11.8296 loss_prob: 8.4270 loss_thr: 2.4445 loss_db: 0.9581 2023/04/03 12:40:27 - mmengine - INFO - Epoch(train) [1][ 65/2146] lr: 2.0000e-06 eta: 1 day, 9:43:10 time: 0.4074 data_time: 0.0055 memory: 4729 loss: 11.7791 loss_prob: 8.3991 loss_thr: 2.4231 loss_db: 0.9569 2023/04/03 12:40:27 - mmengine - INFO - Epoch(train) [1][ 66/2146] lr: 2.0000e-06 eta: 1 day, 9:39:06 time: 0.4028 data_time: 0.0089 memory: 4729 loss: 11.6379 loss_prob: 8.3104 loss_thr: 2.3714 loss_db: 0.9562 2023/04/03 12:40:27 - mmengine - INFO - Epoch(train) [1][ 67/2146] lr: 2.0000e-06 eta: 1 day, 9:36:01 time: 0.4060 data_time: 0.0088 memory: 4729 loss: 11.5585 loss_prob: 8.2572 loss_thr: 2.3486 loss_db: 0.9527 2023/04/03 12:40:28 - mmengine - INFO - Epoch(train) [1][ 68/2146] lr: 2.0000e-06 eta: 1 day, 9:31:32 time: 0.4097 data_time: 0.0098 memory: 4729 loss: 11.3556 loss_prob: 8.1077 loss_thr: 2.2979 loss_db: 0.9499 2023/04/03 12:40:28 - mmengine - INFO - Epoch(train) [1][ 69/2146] lr: 2.0000e-06 eta: 1 day, 9:30:49 time: 0.4158 data_time: 0.0184 memory: 4729 loss: 11.2567 loss_prob: 8.0119 loss_thr: 2.2958 loss_db: 0.9489 2023/04/03 12:40:29 - mmengine - INFO - Epoch(train) [1][ 70/2146] lr: 2.0000e-06 eta: 1 day, 9:27:18 time: 0.4162 data_time: 0.0184 memory: 4729 loss: 11.3519 loss_prob: 8.1033 loss_thr: 2.2933 loss_db: 0.9553 2023/04/03 12:40:29 - mmengine - INFO - Epoch(train) [1][ 71/2146] lr: 2.0000e-06 eta: 1 day, 9:24:33 time: 0.4159 data_time: 0.0183 memory: 4729 loss: 11.2773 loss_prob: 8.0407 loss_thr: 2.2814 loss_db: 0.9552 2023/04/03 12:40:29 - mmengine - INFO - Epoch(train) [1][ 72/2146] lr: 2.0000e-06 eta: 1 day, 9:21:26 time: 0.4158 data_time: 0.0185 memory: 4729 loss: 11.1929 loss_prob: 8.0003 loss_thr: 2.2385 loss_db: 0.9540 2023/04/03 12:40:30 - mmengine - INFO - Epoch(train) [1][ 73/2146] lr: 2.0000e-06 eta: 1 day, 9:17:41 time: 0.4189 data_time: 0.0186 memory: 4729 loss: 11.1063 loss_prob: 7.9512 loss_thr: 2.2014 loss_db: 0.9536 2023/04/03 12:40:30 - mmengine - INFO - Epoch(train) [1][ 74/2146] lr: 2.0000e-06 eta: 1 day, 9:13:02 time: 0.4142 data_time: 0.0155 memory: 4729 loss: 11.1547 loss_prob: 8.0131 loss_thr: 2.1848 loss_db: 0.9568 2023/04/03 12:40:31 - mmengine - INFO - Epoch(train) [1][ 75/2146] lr: 2.0000e-06 eta: 1 day, 9:10:37 time: 0.4148 data_time: 0.0155 memory: 4729 loss: 11.1381 loss_prob: 7.9965 loss_thr: 2.1822 loss_db: 0.9595 2023/04/03 12:40:31 - mmengine - INFO - Epoch(train) [1][ 76/2146] lr: 2.0000e-06 eta: 1 day, 9:08:01 time: 0.4156 data_time: 0.0122 memory: 4729 loss: 11.1209 loss_prob: 7.9685 loss_thr: 2.1923 loss_db: 0.9601 2023/04/03 12:40:31 - mmengine - INFO - Epoch(train) [1][ 77/2146] lr: 2.0000e-06 eta: 1 day, 9:04:51 time: 0.4141 data_time: 0.0122 memory: 4729 loss: 11.0592 loss_prob: 7.9276 loss_thr: 2.1688 loss_db: 0.9627 2023/04/03 12:40:32 - mmengine - INFO - Epoch(train) [1][ 78/2146] lr: 2.0000e-06 eta: 1 day, 9:02:50 time: 0.4167 data_time: 0.0113 memory: 4729 loss: 11.0494 loss_prob: 7.9486 loss_thr: 2.1376 loss_db: 0.9633 2023/04/03 12:40:32 - mmengine - INFO - Epoch(train) [1][ 79/2146] lr: 2.0000e-06 eta: 1 day, 9:00:14 time: 0.4125 data_time: 0.0028 memory: 4729 loss: 10.9998 loss_prob: 7.8973 loss_thr: 2.1390 loss_db: 0.9635 2023/04/03 12:40:33 - mmengine - INFO - Epoch(train) [1][ 80/2146] lr: 2.0000e-06 eta: 1 day, 9:00:31 time: 0.4180 data_time: 0.0059 memory: 4729 loss: 11.0148 loss_prob: 7.8958 loss_thr: 2.1544 loss_db: 0.9646 2023/04/03 12:40:33 - mmengine - INFO - Epoch(train) [1][ 81/2146] lr: 2.0000e-06 eta: 1 day, 8:56:30 time: 0.4144 data_time: 0.0059 memory: 4729 loss: 10.9803 loss_prob: 7.8710 loss_thr: 2.1453 loss_db: 0.9640 2023/04/03 12:40:34 - mmengine - INFO - Epoch(train) [1][ 82/2146] lr: 2.0000e-06 eta: 1 day, 8:54:06 time: 0.4143 data_time: 0.0087 memory: 4729 loss: 10.9563 loss_prob: 7.8802 loss_thr: 2.1103 loss_db: 0.9658 2023/04/03 12:40:34 - mmengine - INFO - Epoch(train) [1][ 83/2146] lr: 2.0000e-06 eta: 1 day, 8:53:12 time: 0.4184 data_time: 0.0086 memory: 4729 loss: 10.9564 loss_prob: 7.8956 loss_thr: 2.0945 loss_db: 0.9663 2023/04/03 12:40:34 - mmengine - INFO - Epoch(train) [1][ 84/2146] lr: 2.0000e-06 eta: 1 day, 8:51:02 time: 0.4215 data_time: 0.0097 memory: 4729 loss: 10.9008 loss_prob: 7.8309 loss_thr: 2.1034 loss_db: 0.9665 2023/04/03 12:40:35 - mmengine - INFO - Epoch(train) [1][ 85/2146] lr: 2.0000e-06 eta: 1 day, 8:49:16 time: 0.4218 data_time: 0.0188 memory: 4729 loss: 10.9355 loss_prob: 7.8729 loss_thr: 2.0936 loss_db: 0.9689 2023/04/03 12:40:35 - mmengine - INFO - Epoch(train) [1][ 86/2146] lr: 2.0000e-06 eta: 1 day, 8:46:08 time: 0.4196 data_time: 0.0189 memory: 4729 loss: 10.9433 loss_prob: 7.8846 loss_thr: 2.0896 loss_db: 0.9690 2023/04/03 12:40:36 - mmengine - INFO - Epoch(train) [1][ 87/2146] lr: 2.0000e-06 eta: 1 day, 8:46:10 time: 0.4248 data_time: 0.0190 memory: 4729 loss: 10.9661 loss_prob: 7.8864 loss_thr: 2.1098 loss_db: 0.9699 2023/04/03 12:40:36 - mmengine - INFO - Epoch(train) [1][ 88/2146] lr: 2.0000e-06 eta: 1 day, 8:42:54 time: 0.4214 data_time: 0.0189 memory: 4729 loss: 10.9301 loss_prob: 7.8477 loss_thr: 2.1111 loss_db: 0.9714 2023/04/03 12:40:37 - mmengine - INFO - Epoch(train) [1][ 89/2146] lr: 2.0000e-06 eta: 1 day, 8:39:10 time: 0.4180 data_time: 0.0190 memory: 4729 loss: 10.9093 loss_prob: 7.8581 loss_thr: 2.0784 loss_db: 0.9728 2023/04/03 12:40:37 - mmengine - INFO - Epoch(train) [1][ 90/2146] lr: 2.0000e-06 eta: 1 day, 8:35:38 time: 0.4096 data_time: 0.0160 memory: 4729 loss: 10.8432 loss_prob: 7.8419 loss_thr: 2.0309 loss_db: 0.9704 2023/04/03 12:40:37 - mmengine - INFO - Epoch(train) [1][ 91/2146] lr: 2.0000e-06 eta: 1 day, 8:33:05 time: 0.4112 data_time: 0.0160 memory: 4729 loss: 10.7790 loss_prob: 7.8067 loss_thr: 2.0023 loss_db: 0.9700 2023/04/03 12:40:38 - mmengine - INFO - Epoch(train) [1][ 92/2146] lr: 2.0000e-06 eta: 1 day, 8:32:03 time: 0.4130 data_time: 0.0131 memory: 4729 loss: 10.7232 loss_prob: 7.7358 loss_thr: 2.0183 loss_db: 0.9691 2023/04/03 12:40:38 - mmengine - INFO - Epoch(train) [1][ 93/2146] lr: 2.0000e-06 eta: 1 day, 8:28:42 time: 0.4069 data_time: 0.0132 memory: 4729 loss: 10.6728 loss_prob: 7.6975 loss_thr: 2.0050 loss_db: 0.9703 2023/04/03 12:40:39 - mmengine - INFO - Epoch(train) [1][ 94/2146] lr: 2.0000e-06 eta: 1 day, 8:27:29 time: 0.4080 data_time: 0.0123 memory: 4729 loss: 10.5795 loss_prob: 7.6398 loss_thr: 1.9703 loss_db: 0.9695 2023/04/03 12:40:39 - mmengine - INFO - Epoch(train) [1][ 95/2146] lr: 2.0000e-06 eta: 1 day, 8:25:54 time: 0.4074 data_time: 0.0032 memory: 4729 loss: 10.5540 loss_prob: 7.6395 loss_thr: 1.9465 loss_db: 0.9680 2023/04/03 12:40:39 - mmengine - INFO - Epoch(train) [1][ 96/2146] lr: 2.0000e-06 eta: 1 day, 8:26:36 time: 0.4147 data_time: 0.0063 memory: 4729 loss: 10.4830 loss_prob: 7.5942 loss_thr: 1.9205 loss_db: 0.9683 2023/04/03 12:40:40 - mmengine - INFO - Epoch(train) [1][ 97/2146] lr: 2.0000e-06 eta: 1 day, 8:23:56 time: 0.4082 data_time: 0.0063 memory: 4729 loss: 10.4198 loss_prob: 7.5419 loss_thr: 1.9110 loss_db: 0.9669 2023/04/03 12:40:40 - mmengine - INFO - Epoch(train) [1][ 98/2146] lr: 2.0000e-06 eta: 1 day, 8:24:11 time: 0.4149 data_time: 0.0094 memory: 4729 loss: 10.3274 loss_prob: 7.4778 loss_thr: 1.8859 loss_db: 0.9637 2023/04/03 12:40:41 - mmengine - INFO - Epoch(train) [1][ 99/2146] lr: 2.0000e-06 eta: 1 day, 8:23:05 time: 0.4197 data_time: 0.0094 memory: 4729 loss: 10.2970 loss_prob: 7.4469 loss_thr: 1.8870 loss_db: 0.9631 2023/04/03 12:40:41 - mmengine - INFO - Epoch(train) [1][ 100/2146] lr: 2.0000e-06 eta: 1 day, 8:20:46 time: 0.4213 data_time: 0.0106 memory: 4729 loss: 10.1804 loss_prob: 7.3332 loss_thr: 1.8839 loss_db: 0.9633 2023/04/03 12:40:42 - mmengine - INFO - Epoch(train) [1][ 101/2146] lr: 2.0000e-06 eta: 1 day, 8:20:59 time: 0.4269 data_time: 0.0193 memory: 4729 loss: 10.2295 loss_prob: 7.3564 loss_thr: 1.9093 loss_db: 0.9638 2023/04/03 12:40:42 - mmengine - INFO - Epoch(train) [1][ 102/2146] lr: 2.0000e-06 eta: 1 day, 8:19:57 time: 0.4264 data_time: 0.0194 memory: 4729 loss: 10.2731 loss_prob: 7.3994 loss_thr: 1.9082 loss_db: 0.9655 2023/04/03 12:40:42 - mmengine - INFO - Epoch(train) [1][ 103/2146] lr: 2.0000e-06 eta: 1 day, 8:19:04 time: 0.4313 data_time: 0.0193 memory: 4729 loss: 10.2373 loss_prob: 7.3787 loss_thr: 1.8930 loss_db: 0.9657 2023/04/03 12:40:43 - mmengine - INFO - Epoch(train) [1][ 104/2146] lr: 2.0000e-06 eta: 1 day, 8:17:52 time: 0.4308 data_time: 0.0192 memory: 4729 loss: 10.2699 loss_prob: 7.3944 loss_thr: 1.9083 loss_db: 0.9671 2023/04/03 12:40:43 - mmengine - INFO - Epoch(train) [1][ 105/2146] lr: 2.0000e-06 eta: 1 day, 8:15:03 time: 0.4272 data_time: 0.0193 memory: 4729 loss: 10.1564 loss_prob: 7.2926 loss_thr: 1.8974 loss_db: 0.9663 2023/04/03 12:40:44 - mmengine - INFO - Epoch(train) [1][ 106/2146] lr: 2.0000e-06 eta: 1 day, 8:11:59 time: 0.4178 data_time: 0.0162 memory: 4729 loss: 10.1113 loss_prob: 7.2528 loss_thr: 1.8934 loss_db: 0.9651 2023/04/03 12:40:44 - mmengine - INFO - Epoch(train) [1][ 107/2146] lr: 2.0000e-06 eta: 1 day, 8:09:40 time: 0.4177 data_time: 0.0163 memory: 4729 loss: 10.1137 loss_prob: 7.2795 loss_thr: 1.8670 loss_db: 0.9672 2023/04/03 12:40:44 - mmengine - INFO - Epoch(train) [1][ 108/2146] lr: 2.0000e-06 eta: 1 day, 8:09:22 time: 0.4160 data_time: 0.0131 memory: 4729 loss: 10.1562 loss_prob: 7.2960 loss_thr: 1.8900 loss_db: 0.9702 2023/04/03 12:40:45 - mmengine - INFO - Epoch(train) [1][ 109/2146] lr: 2.0000e-06 eta: 1 day, 8:08:25 time: 0.4158 data_time: 0.0131 memory: 4729 loss: 10.1491 loss_prob: 7.2953 loss_thr: 1.8825 loss_db: 0.9713 2023/04/03 12:40:45 - mmengine - INFO - Epoch(train) [1][ 110/2146] lr: 2.0000e-06 eta: 1 day, 8:06:57 time: 0.4171 data_time: 0.0120 memory: 4729 loss: 10.1426 loss_prob: 7.2886 loss_thr: 1.8824 loss_db: 0.9716 2023/04/03 12:40:46 - mmengine - INFO - Epoch(train) [1][ 111/2146] lr: 2.0000e-06 eta: 1 day, 8:06:16 time: 0.4145 data_time: 0.0032 memory: 4729 loss: 10.0196 loss_prob: 7.1944 loss_thr: 1.8541 loss_db: 0.9712 2023/04/03 12:40:46 - mmengine - INFO - Epoch(train) [1][ 112/2146] lr: 2.0000e-06 eta: 1 day, 8:05:03 time: 0.4135 data_time: 0.0064 memory: 4729 loss: 9.8994 loss_prob: 7.1003 loss_thr: 1.8299 loss_db: 0.9692 2023/04/03 12:40:47 - mmengine - INFO - Epoch(train) [1][ 113/2146] lr: 2.0000e-06 eta: 1 day, 8:03:43 time: 0.4117 data_time: 0.0064 memory: 4729 loss: 9.8417 loss_prob: 7.0491 loss_thr: 1.8278 loss_db: 0.9649 2023/04/03 12:40:47 - mmengine - INFO - Epoch(train) [1][ 114/2146] lr: 2.0000e-06 eta: 1 day, 8:03:58 time: 0.4150 data_time: 0.0098 memory: 4729 loss: 9.7635 loss_prob: 6.9991 loss_thr: 1.8012 loss_db: 0.9632 2023/04/03 12:40:47 - mmengine - INFO - Epoch(train) [1][ 115/2146] lr: 2.0000e-06 eta: 1 day, 8:04:47 time: 0.4237 data_time: 0.0097 memory: 4729 loss: 9.7603 loss_prob: 6.9842 loss_thr: 1.8111 loss_db: 0.9649 2023/04/03 12:40:48 - mmengine - INFO - Epoch(train) [1][ 116/2146] lr: 2.0000e-06 eta: 1 day, 8:03:47 time: 0.4283 data_time: 0.0108 memory: 4729 loss: 9.7419 loss_prob: 6.9702 loss_thr: 1.8059 loss_db: 0.9658 2023/04/03 12:40:48 - mmengine - INFO - Epoch(train) [1][ 117/2146] lr: 2.0000e-06 eta: 1 day, 8:04:12 time: 0.4351 data_time: 0.0203 memory: 4729 loss: 9.6580 loss_prob: 6.8948 loss_thr: 1.7998 loss_db: 0.9634 2023/04/03 12:40:49 - mmengine - INFO - Epoch(train) [1][ 118/2146] lr: 2.0000e-06 eta: 1 day, 8:02:41 time: 0.4316 data_time: 0.0205 memory: 4729 loss: 9.5742 loss_prob: 6.8412 loss_thr: 1.7708 loss_db: 0.9622 2023/04/03 12:40:49 - mmengine - INFO - Epoch(train) [1][ 119/2146] lr: 2.0000e-06 eta: 1 day, 8:00:46 time: 0.4285 data_time: 0.0207 memory: 4729 loss: 9.5284 loss_prob: 6.8022 loss_thr: 1.7642 loss_db: 0.9621 2023/04/03 12:40:50 - mmengine - INFO - Epoch(train) [1][ 120/2146] lr: 2.0000e-06 eta: 1 day, 8:00:34 time: 0.4315 data_time: 0.0206 memory: 4729 loss: 9.5000 loss_prob: 6.7777 loss_thr: 1.7604 loss_db: 0.9619 2023/04/03 12:40:50 - mmengine - INFO - Epoch(train) [1][ 121/2146] lr: 2.0000e-06 eta: 1 day, 8:00:42 time: 0.4335 data_time: 0.0206 memory: 4729 loss: 9.4826 loss_prob: 6.7794 loss_thr: 1.7405 loss_db: 0.9626 2023/04/03 12:40:50 - mmengine - INFO - Epoch(train) [1][ 122/2146] lr: 2.0000e-06 eta: 1 day, 7:59:44 time: 0.4338 data_time: 0.0174 memory: 4729 loss: 9.4809 loss_prob: 6.7689 loss_thr: 1.7487 loss_db: 0.9634 2023/04/03 12:40:51 - mmengine - INFO - Epoch(train) [1][ 123/2146] lr: 2.0000e-06 eta: 1 day, 7:58:29 time: 0.4336 data_time: 0.0174 memory: 4729 loss: 9.4759 loss_prob: 6.7664 loss_thr: 1.7433 loss_db: 0.9662 2023/04/03 12:40:51 - mmengine - INFO - Epoch(train) [1][ 124/2146] lr: 2.0000e-06 eta: 1 day, 7:59:01 time: 0.4344 data_time: 0.0141 memory: 4729 loss: 9.4177 loss_prob: 6.7158 loss_thr: 1.7353 loss_db: 0.9666 2023/04/03 12:40:52 - mmengine - INFO - Epoch(train) [1][ 125/2146] lr: 2.0000e-06 eta: 1 day, 7:57:02 time: 0.4263 data_time: 0.0141 memory: 4729 loss: 9.3776 loss_prob: 6.6948 loss_thr: 1.7181 loss_db: 0.9647 2023/04/03 12:40:52 - mmengine - INFO - Epoch(train) [1][ 126/2146] lr: 2.0000e-06 eta: 1 day, 7:54:57 time: 0.4227 data_time: 0.0129 memory: 4729 loss: 9.4035 loss_prob: 6.7016 loss_thr: 1.7361 loss_db: 0.9659 2023/04/03 12:40:53 - mmengine - INFO - Epoch(train) [1][ 127/2146] lr: 2.0000e-06 eta: 1 day, 7:53:22 time: 0.4167 data_time: 0.0034 memory: 4729 loss: 9.4632 loss_prob: 6.7387 loss_thr: 1.7557 loss_db: 0.9687 2023/04/03 12:40:53 - mmengine - INFO - Epoch(train) [1][ 128/2146] lr: 2.0000e-06 eta: 1 day, 7:53:15 time: 0.4202 data_time: 0.0062 memory: 4729 loss: 9.4293 loss_prob: 6.7043 loss_thr: 1.7567 loss_db: 0.9683 2023/04/03 12:40:53 - mmengine - INFO - Epoch(train) [1][ 129/2146] lr: 2.0000e-06 eta: 1 day, 7:51:32 time: 0.4202 data_time: 0.0060 memory: 4729 loss: 9.3665 loss_prob: 6.6585 loss_thr: 1.7415 loss_db: 0.9664 2023/04/03 12:40:54 - mmengine - INFO - Epoch(train) [1][ 130/2146] lr: 2.0000e-06 eta: 1 day, 7:51:44 time: 0.4211 data_time: 0.0094 memory: 4729 loss: 9.4149 loss_prob: 6.6903 loss_thr: 1.7557 loss_db: 0.9688 2023/04/03 12:40:54 - mmengine - INFO - Epoch(train) [1][ 131/2146] lr: 2.0000e-06 eta: 1 day, 7:51:25 time: 0.4196 data_time: 0.0094 memory: 4729 loss: 9.4259 loss_prob: 6.6916 loss_thr: 1.7645 loss_db: 0.9699 2023/04/03 12:40:55 - mmengine - INFO - Epoch(train) [1][ 132/2146] lr: 2.0000e-06 eta: 1 day, 7:49:30 time: 0.4162 data_time: 0.0104 memory: 4729 loss: 9.4076 loss_prob: 6.6850 loss_thr: 1.7520 loss_db: 0.9706 2023/04/03 12:40:55 - mmengine - INFO - Epoch(train) [1][ 133/2146] lr: 2.0000e-06 eta: 1 day, 7:48:45 time: 0.4173 data_time: 0.0193 memory: 4729 loss: 9.4262 loss_prob: 6.6918 loss_thr: 1.7619 loss_db: 0.9725 2023/04/03 12:40:55 - mmengine - INFO - Epoch(train) [1][ 134/2146] lr: 2.0000e-06 eta: 1 day, 7:46:43 time: 0.4092 data_time: 0.0192 memory: 4729 loss: 9.4225 loss_prob: 6.6883 loss_thr: 1.7619 loss_db: 0.9723 2023/04/03 12:40:56 - mmengine - INFO - Epoch(train) [1][ 135/2146] lr: 2.0000e-06 eta: 1 day, 7:44:53 time: 0.4089 data_time: 0.0193 memory: 4729 loss: 9.4250 loss_prob: 6.6858 loss_thr: 1.7646 loss_db: 0.9745 2023/04/03 12:40:56 - mmengine - INFO - Epoch(train) [1][ 136/2146] lr: 2.0000e-06 eta: 1 day, 7:43:58 time: 0.4118 data_time: 0.0194 memory: 4729 loss: 9.3692 loss_prob: 6.6500 loss_thr: 1.7452 loss_db: 0.9741 2023/04/03 12:40:57 - mmengine - INFO - Epoch(train) [1][ 137/2146] lr: 2.0000e-06 eta: 1 day, 7:42:00 time: 0.4100 data_time: 0.0194 memory: 4729 loss: 9.3628 loss_prob: 6.6459 loss_thr: 1.7436 loss_db: 0.9734 2023/04/03 12:40:57 - mmengine - INFO - Epoch(train) [1][ 138/2146] lr: 2.0000e-06 eta: 1 day, 7:39:51 time: 0.4031 data_time: 0.0163 memory: 4729 loss: 9.3663 loss_prob: 6.6669 loss_thr: 1.7259 loss_db: 0.9736 2023/04/03 12:40:57 - mmengine - INFO - Epoch(train) [1][ 139/2146] lr: 2.0000e-06 eta: 1 day, 7:38:07 time: 0.4023 data_time: 0.0163 memory: 4729 loss: 9.3507 loss_prob: 6.6666 loss_thr: 1.7125 loss_db: 0.9716 2023/04/03 12:40:58 - mmengine - INFO - Epoch(train) [1][ 140/2146] lr: 2.0000e-06 eta: 1 day, 7:39:08 time: 0.4048 data_time: 0.0130 memory: 4729 loss: 9.2661 loss_prob: 6.6038 loss_thr: 1.6946 loss_db: 0.9676 2023/04/03 12:40:58 - mmengine - INFO - Epoch(train) [1][ 141/2146] lr: 2.0000e-06 eta: 1 day, 7:37:24 time: 0.3997 data_time: 0.0130 memory: 4729 loss: 9.2862 loss_prob: 6.6207 loss_thr: 1.6981 loss_db: 0.9674 2023/04/03 12:40:59 - mmengine - INFO - Epoch(train) [1][ 142/2146] lr: 2.0000e-06 eta: 1 day, 7:36:48 time: 0.4033 data_time: 0.0121 memory: 4729 loss: 9.3188 loss_prob: 6.6611 loss_thr: 1.6891 loss_db: 0.9686 2023/04/03 12:40:59 - mmengine - INFO - Epoch(train) [1][ 143/2146] lr: 2.0000e-06 eta: 1 day, 7:37:52 time: 0.4089 data_time: 0.0031 memory: 4729 loss: 9.2572 loss_prob: 6.6089 loss_thr: 1.6805 loss_db: 0.9679 2023/04/03 12:41:00 - mmengine - INFO - Epoch(train) [1][ 144/2146] lr: 2.0000e-06 eta: 1 day, 7:38:16 time: 0.4163 data_time: 0.0063 memory: 4729 loss: 9.2416 loss_prob: 6.6038 loss_thr: 1.6699 loss_db: 0.9679 2023/04/03 12:41:00 - mmengine - INFO - Epoch(train) [1][ 145/2146] lr: 2.0000e-06 eta: 1 day, 7:37:20 time: 0.4187 data_time: 0.0063 memory: 4729 loss: 9.1938 loss_prob: 6.5624 loss_thr: 1.6649 loss_db: 0.9665 2023/04/03 12:41:00 - mmengine - INFO - Epoch(train) [1][ 146/2146] lr: 2.0000e-06 eta: 1 day, 7:36:58 time: 0.4203 data_time: 0.0090 memory: 4729 loss: 9.1709 loss_prob: 6.5440 loss_thr: 1.6605 loss_db: 0.9664 2023/04/03 12:41:01 - mmengine - INFO - Epoch(train) [1][ 147/2146] lr: 2.0000e-06 eta: 1 day, 7:37:09 time: 0.4270 data_time: 0.0090 memory: 4729 loss: 9.1146 loss_prob: 6.5122 loss_thr: 1.6356 loss_db: 0.9669 2023/04/03 12:41:01 - mmengine - INFO - Epoch(train) [1][ 148/2146] lr: 2.0000e-06 eta: 1 day, 7:36:15 time: 0.4307 data_time: 0.0103 memory: 4729 loss: 9.0995 loss_prob: 6.4870 loss_thr: 1.6447 loss_db: 0.9679 2023/04/03 12:41:02 - mmengine - INFO - Epoch(train) [1][ 149/2146] lr: 2.0000e-06 eta: 1 day, 7:37:09 time: 0.4394 data_time: 0.0200 memory: 4729 loss: 9.1005 loss_prob: 6.4714 loss_thr: 1.6586 loss_db: 0.9706 2023/04/03 12:41:02 - mmengine - INFO - Epoch(train) [1][ 150/2146] lr: 2.0000e-06 eta: 1 day, 7:35:58 time: 0.4319 data_time: 0.0201 memory: 4729 loss: 9.1144 loss_prob: 6.4851 loss_thr: 1.6555 loss_db: 0.9737 2023/04/03 12:41:03 - mmengine - INFO - Epoch(train) [1][ 151/2146] lr: 2.0000e-06 eta: 1 day, 7:34:56 time: 0.4339 data_time: 0.0200 memory: 4729 loss: 9.0403 loss_prob: 6.4323 loss_thr: 1.6362 loss_db: 0.9717 2023/04/03 12:41:03 - mmengine - INFO - Epoch(train) [1][ 152/2146] lr: 2.0000e-06 eta: 1 day, 7:34:14 time: 0.4333 data_time: 0.0199 memory: 4729 loss: 8.9559 loss_prob: 6.3408 loss_thr: 1.6461 loss_db: 0.9690 2023/04/03 12:41:03 - mmengine - INFO - Epoch(train) [1][ 153/2146] lr: 2.0000e-06 eta: 1 day, 7:33:02 time: 0.4254 data_time: 0.0199 memory: 4729 loss: 8.9449 loss_prob: 6.3393 loss_thr: 1.6385 loss_db: 0.9670 2023/04/03 12:41:04 - mmengine - INFO - Epoch(train) [1][ 154/2146] lr: 2.0000e-06 eta: 1 day, 7:31:07 time: 0.4172 data_time: 0.0167 memory: 4729 loss: 8.9735 loss_prob: 6.3574 loss_thr: 1.6484 loss_db: 0.9677 2023/04/03 12:41:04 - mmengine - INFO - Epoch(train) [1][ 155/2146] lr: 2.0000e-06 eta: 1 day, 7:28:39 time: 0.4113 data_time: 0.0166 memory: 4729 loss: 8.9884 loss_prob: 6.3638 loss_thr: 1.6559 loss_db: 0.9687 2023/04/03 12:41:05 - mmengine - INFO - Epoch(train) [1][ 156/2146] lr: 2.0000e-06 eta: 1 day, 7:28:04 time: 0.4101 data_time: 0.0138 memory: 4729 loss: 8.9511 loss_prob: 6.3352 loss_thr: 1.6487 loss_db: 0.9672 2023/04/03 12:41:05 - mmengine - INFO - Epoch(train) [1][ 157/2146] lr: 2.0000e-06 eta: 1 day, 7:26:08 time: 0.4022 data_time: 0.0139 memory: 4729 loss: 8.9695 loss_prob: 6.3537 loss_thr: 1.6480 loss_db: 0.9678 2023/04/03 12:41:05 - mmengine - INFO - Epoch(train) [1][ 158/2146] lr: 2.0000e-06 eta: 1 day, 7:24:53 time: 0.4005 data_time: 0.0126 memory: 4729 loss: 8.9181 loss_prob: 6.3210 loss_thr: 1.6324 loss_db: 0.9647 2023/04/03 12:41:06 - mmengine - INFO - Epoch(train) [1][ 159/2146] lr: 2.0000e-06 eta: 1 day, 7:25:04 time: 0.3978 data_time: 0.0029 memory: 4729 loss: 8.8569 loss_prob: 6.2827 loss_thr: 1.6111 loss_db: 0.9631 2023/04/03 12:41:06 - mmengine - INFO - Epoch(train) [1][ 160/2146] lr: 2.0000e-06 eta: 1 day, 7:24:36 time: 0.3999 data_time: 0.0062 memory: 4729 loss: 8.8093 loss_prob: 6.2442 loss_thr: 1.6040 loss_db: 0.9612 2023/04/03 12:41:07 - mmengine - INFO - Epoch(train) [1][ 161/2146] lr: 2.0000e-06 eta: 1 day, 7:24:05 time: 0.4013 data_time: 0.0062 memory: 4729 loss: 8.7648 loss_prob: 6.2092 loss_thr: 1.5942 loss_db: 0.9614 2023/04/03 12:41:07 - mmengine - INFO - Epoch(train) [1][ 162/2146] lr: 2.0000e-06 eta: 1 day, 7:22:58 time: 0.3993 data_time: 0.0096 memory: 4729 loss: 8.7570 loss_prob: 6.2151 loss_thr: 1.5791 loss_db: 0.9627 2023/04/03 12:41:07 - mmengine - INFO - Epoch(train) [1][ 163/2146] lr: 2.0000e-06 eta: 1 day, 7:22:46 time: 0.4026 data_time: 0.0095 memory: 4729 loss: 8.7302 loss_prob: 6.1945 loss_thr: 1.5707 loss_db: 0.9650 2023/04/03 12:41:08 - mmengine - INFO - Epoch(train) [1][ 164/2146] lr: 2.0000e-06 eta: 1 day, 7:23:09 time: 0.4107 data_time: 0.0109 memory: 4729 loss: 8.7112 loss_prob: 6.1786 loss_thr: 1.5661 loss_db: 0.9665 2023/04/03 12:41:08 - mmengine - INFO - Epoch(train) [1][ 165/2146] lr: 2.0000e-06 eta: 1 day, 7:22:57 time: 0.4186 data_time: 0.0206 memory: 4729 loss: 8.7201 loss_prob: 6.1832 loss_thr: 1.5706 loss_db: 0.9662 2023/04/03 12:41:09 - mmengine - INFO - Epoch(train) [1][ 166/2146] lr: 2.0000e-06 eta: 1 day, 7:21:48 time: 0.4162 data_time: 0.0205 memory: 4729 loss: 8.7041 loss_prob: 6.1803 loss_thr: 1.5592 loss_db: 0.9646 2023/04/03 12:41:09 - mmengine - INFO - Epoch(train) [1][ 167/2146] lr: 2.0000e-06 eta: 1 day, 7:21:23 time: 0.4215 data_time: 0.0205 memory: 4729 loss: 8.6293 loss_prob: 6.1121 loss_thr: 1.5547 loss_db: 0.9625 2023/04/03 12:41:10 - mmengine - INFO - Epoch(train) [1][ 168/2146] lr: 2.0000e-06 eta: 1 day, 7:21:48 time: 0.4276 data_time: 0.0206 memory: 4729 loss: 8.6508 loss_prob: 6.1242 loss_thr: 1.5605 loss_db: 0.9661 2023/04/03 12:41:10 - mmengine - INFO - Epoch(train) [1][ 169/2146] lr: 2.0000e-06 eta: 1 day, 7:20:59 time: 0.4236 data_time: 0.0206 memory: 4729 loss: 8.6461 loss_prob: 6.1175 loss_thr: 1.5623 loss_db: 0.9663 2023/04/03 12:41:10 - mmengine - INFO - Epoch(train) [1][ 170/2146] lr: 2.0000e-06 eta: 1 day, 7:20:09 time: 0.4220 data_time: 0.0172 memory: 4729 loss: 8.6377 loss_prob: 6.1142 loss_thr: 1.5576 loss_db: 0.9659 2023/04/03 12:41:11 - mmengine - INFO - Epoch(train) [1][ 171/2146] lr: 2.0000e-06 eta: 1 day, 7:18:51 time: 0.4187 data_time: 0.0172 memory: 4729 loss: 8.6535 loss_prob: 6.1173 loss_thr: 1.5693 loss_db: 0.9670 2023/04/03 12:41:11 - mmengine - INFO - Epoch(train) [1][ 172/2146] lr: 2.0000e-06 eta: 1 day, 7:18:01 time: 0.4195 data_time: 0.0138 memory: 4729 loss: 8.6183 loss_prob: 6.0906 loss_thr: 1.5640 loss_db: 0.9638 2023/04/03 12:41:12 - mmengine - INFO - Epoch(train) [1][ 173/2146] lr: 2.0000e-06 eta: 1 day, 7:17:05 time: 0.4163 data_time: 0.0138 memory: 4729 loss: 8.5988 loss_prob: 6.0791 loss_thr: 1.5567 loss_db: 0.9631 2023/04/03 12:41:12 - mmengine - INFO - Epoch(train) [1][ 174/2146] lr: 2.0000e-06 eta: 1 day, 7:17:13 time: 0.4152 data_time: 0.0125 memory: 4729 loss: 8.5560 loss_prob: 6.0533 loss_thr: 1.5400 loss_db: 0.9628 2023/04/03 12:41:12 - mmengine - INFO - Epoch(train) [1][ 175/2146] lr: 2.0000e-06 eta: 1 day, 7:16:27 time: 0.4128 data_time: 0.0028 memory: 4729 loss: 8.5022 loss_prob: 6.0315 loss_thr: 1.5068 loss_db: 0.9639 2023/04/03 12:41:13 - mmengine - INFO - Epoch(train) [1][ 176/2146] lr: 2.0000e-06 eta: 1 day, 7:16:58 time: 0.4192 data_time: 0.0065 memory: 4729 loss: 8.6022 loss_prob: 6.1161 loss_thr: 1.5184 loss_db: 0.9677 2023/04/03 12:41:13 - mmengine - INFO - Epoch(train) [1][ 177/2146] lr: 2.0000e-06 eta: 1 day, 7:15:50 time: 0.4160 data_time: 0.0065 memory: 4729 loss: 8.5943 loss_prob: 6.1160 loss_thr: 1.5116 loss_db: 0.9668 2023/04/03 12:41:14 - mmengine - INFO - Epoch(train) [1][ 178/2146] lr: 2.0000e-06 eta: 1 day, 7:15:57 time: 0.4148 data_time: 0.0099 memory: 4729 loss: 8.6682 loss_prob: 6.1755 loss_thr: 1.5253 loss_db: 0.9674 2023/04/03 12:41:14 - mmengine - INFO - Epoch(train) [1][ 179/2146] lr: 2.0000e-06 eta: 1 day, 7:14:40 time: 0.4125 data_time: 0.0099 memory: 4729 loss: 8.6627 loss_prob: 6.1783 loss_thr: 1.5159 loss_db: 0.9685 2023/04/03 12:41:14 - mmengine - INFO - Epoch(train) [1][ 180/2146] lr: 2.0000e-06 eta: 1 day, 7:13:06 time: 0.4091 data_time: 0.0110 memory: 4729 loss: 8.6732 loss_prob: 6.1890 loss_thr: 1.5153 loss_db: 0.9689 2023/04/03 12:41:15 - mmengine - INFO - Epoch(train) [1][ 181/2146] lr: 2.0000e-06 eta: 1 day, 7:13:36 time: 0.4161 data_time: 0.0201 memory: 4729 loss: 8.6802 loss_prob: 6.1989 loss_thr: 1.5117 loss_db: 0.9696 2023/04/03 12:41:15 - mmengine - INFO - Epoch(train) [1][ 182/2146] lr: 2.0000e-06 eta: 1 day, 7:12:47 time: 0.4159 data_time: 0.0201 memory: 4729 loss: 8.7047 loss_prob: 6.2095 loss_thr: 1.5223 loss_db: 0.9729 2023/04/03 12:41:16 - mmengine - INFO - Epoch(train) [1][ 183/2146] lr: 2.0000e-06 eta: 1 day, 7:12:55 time: 0.4202 data_time: 0.0202 memory: 4729 loss: 8.6953 loss_prob: 6.2105 loss_thr: 1.5139 loss_db: 0.9709 2023/04/03 12:4 ... ... 2023/04/03 13:18:42 - mmengine - INFO - Epoch(val) [1][379/379] eta: 0:00:00 time: 3.6606 data_time: 0.0070 memory: 3077 2023/04/03 13:19:05 - mmengine - INFO - Evaluating hmean-iou... 2023/04/03 13:19:06 - mmengine - INFO - prediction score threshold: 0.30, recall: 0.0000, precision: 0.0000, hmean: 0.0000 2023/04/03 13:19:06 - mmengine - INFO - prediction score threshold: 0.40, recall: 0.0000, precision: 0.0000, hmean: 0.0000 2023/04/03 13:19:06 - mmengine - INFO - prediction score threshold: 0.50, recall: 0.0000, precision: 0.0000, hmean: 0.0000 2023/04/03 13:19:06 - mmengine - INFO - prediction score threshold: 0.60, recall: 0.0000, precision: 0.0000, hmean: 0.0000 2023/04/03 13:19:06 - mmengine - INFO - prediction score threshold: 0.70, recall: 0.0000, precision: 0.0000, hmean: 0.0000 2023/04/03 13:19:06 - mmengine - INFO - prediction score threshold: 0.80, recall: 0.0000, precision: 0.0000, hmean: 0.0000 2023/04/03 13:19:06 - mmengine - INFO - prediction score threshold: 0.90, recall: 0.0000, precision: 0.0000, hmean: 0.0000 2023/04/03 13:19:06 - mmengine - INFO - Epoch(val) [1][379/379] icdar/precision: 0.0000 icdar/recall: 0.0000 icdar/hmean: 0.0000data_time: 0.0070 time: 3.6606 2023/04/03 13:19:07 - mmengine - INFO - Epoch(train) [2][ 1/2146] lr: 1.2040e-05 eta: 1 day, 5:20:09 time: 0.4551 data_time: 0.0813 memory: 4713 loss: 5.4965 loss_prob: 3.3455 loss_thr: 1.1780 loss_db: 0.9730 2023/04/03 13:19:07 - mmengine - INFO - Epoch(train) [2][ 2/2146] lr: 1.2040e-05 eta: 1 day, 5:20:09 time: 0.4568 data_time: 0.0834 memory: 4714 loss: 5.4838 loss_prob: 3.3388 loss_thr: 1.1753 loss_db: 0.9698 2023/04/03 13:19:08 - mmengine - INFO - Epoch(train) [2][ 3/2146] lr: 1.2040e-05 eta: 1 day, 5:20:07 time: 0.4579 data_time: 0.0834 memory: 4714 loss: 5.4796 loss_prob: 3.3375 loss_thr: 1.1724 loss_db: 0.9698 2023/04/03 13:19:08 - mmengine - INFO - Epoch(train) [2][ 4/2146] lr: 1.2040e-05 eta: 1 day, 5:20:01 time: 0.4574 data_time: 0.0834 memory: 4714 loss: 5.5357 loss_prob: 3.3867 loss_thr: 1.1779 loss_db: 0.9711 2023/04/03 13:19:09 - mmengine - INFO - Epoch(train) [2][ 5/2146] lr: 1.2040e-05 eta: 1 day, 5:20:07 time: 0.4657 data_time: 0.0858 memory: 4714 loss: 5.5565 loss_prob: 3.4017 loss_thr: 1.1798 ... ... 2023/04/04 00:41:52 - mmengine - INFO - Epoch(val) [42][376/379] eta: 0:00:00 time: 0.2575 data_time: 0.0168 memory: 3969 2023/04/04 00:41:53 - mmengine - INFO - Epoch(val) [42][377/379] eta: 0:00:00 time: 0.2586 data_time: 0.0180 memory: 3969 2023/04/04 00:41:53 - mmengine - INFO - Epoch(val) [42][378/379] eta: 0:00:00 time: 0.2558 data_time: 0.0153 memory: 3969 2023/04/04 00:41:53 - mmengine - INFO - Epoch(val) [42][379/379] eta: 0:00:00 time: 0.2424 data_time: 0.0071 memory: 3077 2023/04/04 00:41:53 - mmengine - INFO - Evaluating hmean-iou... 2023/04/04 00:41:54 - mmengine - INFO - prediction score threshold: 0.30, recall: 0.0000, precision: 0.0000, hmean: 0.0000 2023/04/04 00:41:54 - mmengine - INFO - prediction score threshold: 0.40, recall: 0.0000, precision: 0.0000, hmean: 0.0000 2023/04/04 00:41:54 - mmengine - INFO - prediction score threshold: 0.50, recall: 0.0000, precision: 0.0000, hmean: 0.0000 2023/04/04 00:41:54 - mmengine - INFO - prediction score threshold: 0.60, recall: 0.0000, precision: 0.0000, hmean: 0.0000 2023/04/04 00:41:54 - mmengine - INFO - prediction score threshold: 0.70, recall: 0.0000, precision: 0.0000, hmean: 0.0000 2023/04/04 00:41:54 - mmengine - INFO - prediction score threshold: 0.80, recall: 0.0000, precision: 0.0000, hmean: 0.0000 2023/04/04 00:41:54 - mmengine - INFO - prediction score threshold: 0.90, recall: 0.0000, precision: 0.0000, hmean: 0.0000 2023/04/04 00:41:54 - mmengine - INFO - Epoch(val) [42][379/379] icdar/precision: 0.0000 icdar/recall: 0.0000 icdar/hmean: 0.0000data_time: 0.0071 time: 0.2424 2023/04/04 00:41:55 - mmengine - INFO - Epoch(train) [43][ 1/2146] lr: 4.2369e-04 eta: 19:21:58 time: 0.4715 data_time: 0.0841 memory: 4714 loss: 4.8211 loss_prob: 2.8363 loss_thr: 1.1119 loss_db: 0.8729 2023/04/04 00:41:56 - mmengine - INFO - Epoch(train) [43][ 2/2146] lr: 4.2369e-04 eta: 19:21:58 time: 0.4790 data_time: 0.0841 memory: 4714 loss: 4.8279 loss_prob: 2.8348 loss_thr: 1.1182 loss_db: 0.8748 2023/04/04 00:41:56 - mmengine - INFO - Epoch(train) [43][ 3/2146] lr: 4.2369e-04 eta: 19:21:57 time: 0.4796 data_time: 0.0839 memory: 4714 loss: 4.8142 loss_prob: 2.8237 loss_thr: 1.1149 loss_db: 0.8755 2023/04/04 00:41:57 - mmengine - INFO - Epoch(train) [43][ 4/2146] lr: 4.2369e-04 eta: 19:21:57 time: 0.4845 data_time: 0.0839 memory: 4714 loss: 4.8804 loss_prob: 2.8405 loss_thr: 1.1279 loss_db: 0.9120 2023/04/04 00:41:57 - mmengine - INFO - Epoch(train) [43][ 5/2146] lr: 4.2369e-04 eta: 19:21:57 time: 0.4907 data_time: 0.0839 memory: 4714 loss: 4.9374 loss_prob: 2.8516 loss_thr: 1.1418 loss_db: 0.9440Why my recall, precision and hmean are always 0, my training data and test data are in the same format.
Beta Was this translation helpful? Give feedback.
All reactions