Change resnet size

main.py

@@ -1,11 +1,14 @@
 from mmocr.apis import MMOCRInferencer
-
+from mmdet.apis import init_detector, inference_detector
 import gradio as gr
 import cv2
 import sys
 import torch
 
+print('Loading model...')
 device = 'gpu' if torch.cuda.is_available() else 'cpu'
+table_det = init_detector('model/table-det/config.py',
+                          'model/table-det/model.pth', device=device)
 
 ocr = MMOCRInferencer(det='model/det/config.py',
             det_weights='model/det/model.pth',

model/det/20230224_051330.log

@@ -1,755 +0,0 @@
-2023/02/24 05:13:32 - mmengine - INFO - 
-------------------------------------------------------------
-System environment:
-    sys.platform: linux
-    Python: 3.8.10 (default, Nov 14 2022, 12:59:47) [GCC 9.4.0]
-    CUDA available: True
-    numpy_random_seed: 1569491978
-    GPU 0: Tesla T4
-    CUDA_HOME: /usr/local/cuda
-    NVCC: Cuda compilation tools, release 11.6, V11.6.124
-    GCC: x86_64-linux-gnu-gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0
-    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: 4.6.0
-    MMEngine: 0.5.0
-
-Runtime environment:
-    cudnn_benchmark: True
-    mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0}
-    dist_cfg: {'backend': 'nccl'}
-    seed: None
-    Distributed launcher: none
-    Distributed training: False
-    GPU number: 1
-------------------------------------------------------------
-
-2023/02/24 05:13:33 - mmengine - INFO - Config:
-file_client_args = dict(backend='disk')
-model = dict(
-    type='DBNet',
-    backbone=dict(
-        type='mmdet.ResNet',
-        depth=18,
-        num_stages=4,
-        out_indices=(0, 1, 2, 3),
-        frozen_stages=-1,
-        norm_cfg=dict(type='BN', requires_grad=True),
-        init_cfg=dict(type='Pretrained', checkpoint='torchvision://resnet18'),
-        norm_eval=False,
-        style='caffe'),
-    neck=dict(
-        type='FPNC', in_channels=[64, 128, 256, 512], lateral_channels=256),
-    det_head=dict(
-        type='DBHead',
-        in_channels=256,
-        module_loss=dict(type='DBModuleLoss'),
-        postprocessor=dict(type='DBPostprocessor', text_repr_type='quad')),
-    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',
-        file_client_args=dict(backend='disk'),
-        color_type='color_ignore_orientation'),
-    dict(
-        type='LoadOCRAnnotations',
-        with_polygon=True,
-        with_bbox=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',
-        file_client_args=dict(backend='disk'),
-        color_type='color_ignore_orientation'),
-    dict(type='Resize', scale=(1333, 736), 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'))
-]
-icdar2015_textdet_data_root = 'data/det/textdet-thvote'
-icdar2015_textdet_train = dict(
-    type='OCRDataset',
-    data_root='data/det/textdet-thvote',
-    ann_file='textdet_train.json',
-    data_prefix=dict(img_path='imgs/'),
-    filter_cfg=dict(filter_empty_gt=True, min_size=32),
-    pipeline=[
-        dict(
-            type='LoadImageFromFile',
-            file_client_args=dict(backend='disk'),
-            color_type='color_ignore_orientation'),
-        dict(
-            type='LoadOCRAnnotations',
-            with_polygon=True,
-            with_bbox=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'))
-    ])
-icdar2015_textdet_test = dict(
-    type='OCRDataset',
-    data_root='data/det/textdet-thvote',
-    ann_file='textdet_test.json',
-    data_prefix=dict(img_path='imgs/'),
-    test_mode=True,
-    pipeline=[
-        dict(
-            type='LoadImageFromFile',
-            file_client_args=dict(backend='disk'),
-            color_type='color_ignore_orientation'),
-        dict(type='Resize', scale=(1333, 736), 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'))
-    ])
-default_scope = 'mmocr'
-env_cfg = dict(
-    cudnn_benchmark=True,
-    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=5),
-    param_scheduler=dict(type='ParamSchedulerHook'),
-    checkpoint=dict(type='CheckpointHook', interval=20),
-    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')])
-optim_wrapper = dict(
-    type='OptimWrapper',
-    optimizer=dict(type='SGD', lr=0.007, momentum=0.9, weight_decay=0.0001))
-train_cfg = dict(type='EpochBasedTrainLoop', max_epochs=1200, val_interval=20)
-val_cfg = dict(type='ValLoop')
-test_cfg = dict(type='TestLoop')
-param_scheduler = [dict(type='PolyLR', power=0.9, eta_min=1e-07, end=1200)]
-train_dataloader = dict(
-    batch_size=16,
-    num_workers=8,
-    persistent_workers=True,
-    sampler=dict(type='DefaultSampler', shuffle=True),
-    dataset=dict(
-        type='OCRDataset',
-        data_root='data/det/textdet-thvote',
-        ann_file='textdet_train.json',
-        data_prefix=dict(img_path='imgs/'),
-        filter_cfg=dict(filter_empty_gt=True, min_size=32),
-        pipeline=[
-            dict(
-                type='LoadImageFromFile',
-                file_client_args=dict(backend='disk'),
-                color_type='color_ignore_orientation'),
-            dict(
-                type='LoadOCRAnnotations',
-                with_polygon=True,
-                with_bbox=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=1,
-    num_workers=4,
-    persistent_workers=True,
-    sampler=dict(type='DefaultSampler', shuffle=False),
-    dataset=dict(
-        type='OCRDataset',
-        data_root='data/det/textdet-thvote',
-        ann_file='textdet_test.json',
-        data_prefix=dict(img_path='imgs/'),
-        test_mode=True,
-        pipeline=[
-            dict(
-                type='LoadImageFromFile',
-                file_client_args=dict(backend='disk'),
-                color_type='color_ignore_orientation'),
-            dict(type='Resize', scale=(1333, 736), 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'))
-        ]))
-test_dataloader = dict(
-    batch_size=1,
-    num_workers=4,
-    persistent_workers=True,
-    sampler=dict(type='DefaultSampler', shuffle=False),
-    dataset=dict(
-        type='OCRDataset',
-        data_root='data/det/textdet-thvote',
-        ann_file='textdet_test.json',
-        data_prefix=dict(img_path='imgs/'),
-        test_mode=True,
-        pipeline=[
-            dict(
-                type='LoadImageFromFile',
-                file_client_args=dict(backend='disk'),
-                color_type='color_ignore_orientation'),
-            dict(type='Resize', scale=(1333, 736), 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'))
-        ]))
-auto_scale_lr = dict(base_batch_size=16)
-launcher = 'none'
-work_dir = './work_dirs/dbnet_resnet18_fpnc_1200e_icdar2015'
-
-2023/02/24 05:13:33 - mmengine - WARNING - The "visualizer" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead.
-2023/02/24 05:13:33 - mmengine - WARNING - The "vis_backend" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead.
-2023/02/24 05:13:34 - mmengine - WARNING - The "model" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead.
-2023/02/24 05:13:34 - mmengine - WARNING - The "model" registry in mmdet did not set import location. Fallback to call `mmdet.utils.register_all_modules` instead.
-2023/02/24 05:13:38 - mmengine - INFO - Distributed training is not used, all SyncBatchNorm (SyncBN) layers in the model will be automatically reverted to BatchNormXd layers if they are used.
-2023/02/24 05:13:38 - mmengine - WARNING - The "hook" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead.
-2023/02/24 05:13:38 - 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                         
- -------------------- 
-2023/02/24 05:13:39 - mmengine - WARNING - The "loop" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead.
-2023/02/24 05:13:39 - mmengine - WARNING - The "dataset" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead.
-2023/02/24 05:13:39 - mmengine - WARNING - The "transform" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead.
-2023/02/24 05:13:39 - mmengine - WARNING - The "data sampler" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead.
-2023/02/24 05:13:39 - mmengine - WARNING - The "optimizer constructor" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead.
-2023/02/24 05:13:39 - mmengine - WARNING - The "optimizer" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead.
-2023/02/24 05:13:39 - mmengine - WARNING - The "optim wrapper" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead.
-2023/02/24 05:13:39 - mmengine - WARNING - The "parameter scheduler" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead.
-2023/02/24 05:13:40 - mmengine - WARNING - The "metric" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead.
-2023/02/24 05:13:40 - mmengine - WARNING - The "weight initializer" registry in mmocr did not set import location. Fallback to call `mmocr.utils.register_all_modules` instead.
-2023/02/24 05:13:40 - mmengine - INFO - load model from: torchvision://resnet18
-2023/02/24 05:13:40 - mmengine - INFO - Loads checkpoint by torchvision backend from path: torchvision://resnet18
-2023/02/24 05:13:40 - mmengine - WARNING - The model and loaded state dict do not match exactly
-
-unexpected key in source state_dict: fc.weight, fc.bias
-
-Name of parameter - Initialization information
-
-backbone.conv1.weight - torch.Size([64, 3, 7, 7]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.bn1.weight - torch.Size([64]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.bn1.bias - torch.Size([64]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer1.0.conv1.weight - torch.Size([64, 64, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer1.0.bn1.weight - torch.Size([64]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer1.0.bn1.bias - torch.Size([64]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer1.0.conv2.weight - torch.Size([64, 64, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer1.0.bn2.weight - torch.Size([64]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer1.0.bn2.bias - torch.Size([64]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer1.1.conv1.weight - torch.Size([64, 64, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer1.1.bn1.weight - torch.Size([64]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer1.1.bn1.bias - torch.Size([64]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer1.1.conv2.weight - torch.Size([64, 64, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer1.1.bn2.weight - torch.Size([64]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer1.1.bn2.bias - torch.Size([64]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer2.0.conv1.weight - torch.Size([128, 64, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer2.0.bn1.weight - torch.Size([128]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer2.0.bn1.bias - torch.Size([128]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer2.0.conv2.weight - torch.Size([128, 128, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer2.0.bn2.weight - torch.Size([128]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer2.0.bn2.bias - torch.Size([128]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer2.0.downsample.0.weight - torch.Size([128, 64, 1, 1]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer2.0.downsample.1.weight - torch.Size([128]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer2.0.downsample.1.bias - torch.Size([128]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer2.1.conv1.weight - torch.Size([128, 128, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer2.1.bn1.weight - torch.Size([128]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer2.1.bn1.bias - torch.Size([128]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer2.1.conv2.weight - torch.Size([128, 128, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer2.1.bn2.weight - torch.Size([128]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer2.1.bn2.bias - torch.Size([128]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer3.0.conv1.weight - torch.Size([256, 128, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer3.0.bn1.weight - torch.Size([256]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer3.0.bn1.bias - torch.Size([256]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer3.0.conv2.weight - torch.Size([256, 256, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer3.0.bn2.weight - torch.Size([256]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer3.0.bn2.bias - torch.Size([256]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer3.0.downsample.0.weight - torch.Size([256, 128, 1, 1]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer3.0.downsample.1.weight - torch.Size([256]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer3.0.downsample.1.bias - torch.Size([256]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer3.1.conv1.weight - torch.Size([256, 256, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer3.1.bn1.weight - torch.Size([256]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer3.1.bn1.bias - torch.Size([256]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer3.1.conv2.weight - torch.Size([256, 256, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer3.1.bn2.weight - torch.Size([256]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer3.1.bn2.bias - torch.Size([256]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer4.0.conv1.weight - torch.Size([512, 256, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer4.0.bn1.weight - torch.Size([512]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer4.0.bn1.bias - torch.Size([512]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer4.0.conv2.weight - torch.Size([512, 512, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer4.0.bn2.weight - torch.Size([512]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer4.0.bn2.bias - torch.Size([512]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer4.0.downsample.0.weight - torch.Size([512, 256, 1, 1]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer4.0.downsample.1.weight - torch.Size([512]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer4.0.downsample.1.bias - torch.Size([512]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer4.1.conv1.weight - torch.Size([512, 512, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer4.1.bn1.weight - torch.Size([512]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer4.1.bn1.bias - torch.Size([512]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer4.1.conv2.weight - torch.Size([512, 512, 3, 3]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer4.1.bn2.weight - torch.Size([512]): 
-PretrainedInit: load from torchvision://resnet18 
-
-backbone.layer4.1.bn2.bias - torch.Size([512]): 
-PretrainedInit: load from torchvision://resnet18 
-
-neck.lateral_convs.0.conv.weight - torch.Size([256, 64, 1, 1]): 
-Initialized by user-defined `init_weights` in ConvModule  
-
-neck.lateral_convs.1.conv.weight - torch.Size([256, 128, 1, 1]): 
-Initialized by user-defined `init_weights` in ConvModule  
-
-neck.lateral_convs.2.conv.weight - torch.Size([256, 256, 1, 1]): 
-Initialized by user-defined `init_weights` in ConvModule  
-
-neck.lateral_convs.3.conv.weight - torch.Size([256, 512, 1, 1]): 
-Initialized by user-defined `init_weights` in ConvModule  
-
-neck.smooth_convs.0.conv.weight - torch.Size([64, 256, 3, 3]): 
-Initialized by user-defined `init_weights` in ConvModule  
-
-neck.smooth_convs.1.conv.weight - torch.Size([64, 256, 3, 3]): 
-Initialized by user-defined `init_weights` in ConvModule  
-
-neck.smooth_convs.2.conv.weight - torch.Size([64, 256, 3, 3]): 
-Initialized by user-defined `init_weights` in ConvModule  
-
-neck.smooth_convs.3.conv.weight - torch.Size([64, 256, 3, 3]): 
-Initialized by user-defined `init_weights` in ConvModule  
-
-det_head.binarize.0.weight - torch.Size([64, 256, 3, 3]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.binarize.1.weight - torch.Size([64]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.binarize.1.bias - torch.Size([64]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.binarize.3.weight - torch.Size([64, 64, 2, 2]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.binarize.3.bias - torch.Size([64]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.binarize.4.weight - torch.Size([64]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.binarize.4.bias - torch.Size([64]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.binarize.6.weight - torch.Size([64, 1, 2, 2]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.binarize.6.bias - torch.Size([1]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.threshold.0.weight - torch.Size([64, 256, 3, 3]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.threshold.1.weight - torch.Size([64]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.threshold.1.bias - torch.Size([64]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.threshold.3.weight - torch.Size([64, 64, 2, 2]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.threshold.3.bias - torch.Size([64]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.threshold.4.weight - torch.Size([64]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.threshold.4.bias - torch.Size([64]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.threshold.6.weight - torch.Size([64, 1, 2, 2]): 
-The value is the same before and after calling `init_weights` of DBNet  
-
-det_head.threshold.6.bias - torch.Size([1]): 
-The value is the same before and after calling `init_weights` of DBNet  
-2023/02/24 05:13:40 - mmengine - INFO - Checkpoints will be saved to /content/mmocr/work_dirs/dbnet_resnet18_fpnc_1200e_icdar2015.
-2023/02/24 05:16:48 - mmengine - INFO - Epoch(train)    [1][ 5/22]  lr: 7.0000e-03  eta: 11 days, 10:56:37  time: 37.4994  data_time: 13.3666  memory: 12058  loss: 10.5798  loss_prob: 7.3334  loss_thr: 2.3504  loss_db: 0.8960
-2023/02/24 05:17:25 - mmengine - INFO - Epoch(train)    [1][10/22]  lr: 7.0000e-03  eta: 6 days, 20:37:40  time: 22.4578  data_time: 6.7581  memory: 6713  loss: 8.0422  loss_prob: 5.2998  loss_thr: 1.8354  loss_db: 0.9071
-2023/02/24 05:17:49 - mmengine - INFO - Epoch(train)    [1][15/22]  lr: 7.0000e-03  eta: 5 days, 1:36:06  time: 6.1375  data_time: 0.0814  memory: 6713  loss: 5.2709  loss_prob: 3.0675  loss_thr: 1.2472  loss_db: 0.9562
-2023/02/24 05:18:13 - mmengine - INFO - Epoch(train)    [1][20/22]  lr: 7.0000e-03  eta: 4 days, 3:52:43  time: 4.8026  data_time: 0.0312  memory: 6713  loss: 4.9844  loss_prob: 2.8490  loss_thr: 1.1389  loss_db: 0.9965
-2023/02/24 05:18:25 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 05:21:34 - mmengine - INFO - Epoch(train)    [2][ 5/22]  lr: 6.9947e-03  eta: 5 days, 8:31:25  time: 21.5618  data_time: 7.1003  memory: 11447  loss: 4.8425  loss_prob: 2.8106  loss_thr: 1.0607  loss_db: 0.9712
-2023/02/24 05:22:09 - mmengine - INFO - Epoch(train)    [2][10/22]  lr: 6.9947e-03  eta: 4 days, 20:24:29  time: 22.4338  data_time: 7.1646  memory: 6712  loss: 4.7001  loss_prob: 2.7874  loss_thr: 1.1015  loss_db: 0.8112
-2023/02/24 05:22:33 - mmengine - INFO - Epoch(train)    [2][15/22]  lr: 6.9947e-03  eta: 4 days, 9:30:51  time: 5.9429  data_time: 0.0877  memory: 6712  loss: 4.4307  loss_prob: 2.7478  loss_thr: 1.1405  loss_db: 0.5424
-2023/02/24 05:22:56 - mmengine - INFO - Epoch(train)    [2][20/22]  lr: 6.9947e-03  eta: 4 days, 0:51:26  time: 4.7033  data_time: 0.0489  memory: 6712  loss: 4.1205  loss_prob: 2.6747  loss_thr: 1.0579  loss_db: 0.3879
-2023/02/24 05:23:05 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 05:25:58 - mmengine - INFO - Epoch(train)    [3][ 5/22]  lr: 6.9895e-03  eta: 4 days, 14:13:27  time: 19.7292  data_time: 6.3200  memory: 6712  loss: 3.7028  loss_prob: 2.4246  loss_thr: 0.9721  loss_db: 0.3061
-2023/02/24 05:26:33 - mmengine - INFO - Epoch(train)    [3][10/22]  lr: 6.9895e-03  eta: 4 days, 8:44:41  time: 20.8299  data_time: 6.3501  memory: 6712  loss: 3.4052  loss_prob: 2.1909  loss_thr: 0.9435  loss_db: 0.2709
-2023/02/24 05:26:53 - mmengine - INFO - Epoch(train)    [3][15/22]  lr: 6.9895e-03  eta: 4 days, 2:14:03  time: 5.4242  data_time: 0.0758  memory: 6712  loss: 3.1914  loss_prob: 2.0126  loss_thr: 0.9125  loss_db: 0.2664
-2023/02/24 05:27:15 - mmengine - INFO - Epoch(train)    [3][20/22]  lr: 6.9895e-03  eta: 3 days, 21:04:03  time: 4.1317  data_time: 0.0486  memory: 6712  loss: 2.9899  loss_prob: 1.8336  loss_thr: 0.8950  loss_db: 0.2613
-2023/02/24 05:27:23 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 05:30:21 - mmengine - INFO - Epoch(train)    [4][ 5/22]  lr: 6.9842e-03  eta: 4 days, 7:06:23  time: 19.9728  data_time: 6.5625  memory: 6712  loss: 2.7135  loss_prob: 1.6040  loss_thr: 0.8757  loss_db: 0.2338
-2023/02/24 05:30:55 - mmengine - INFO - Epoch(train)    [4][10/22]  lr: 6.9842e-03  eta: 4 days, 3:31:24  time: 21.1335  data_time: 6.5916  memory: 6712  loss: 2.5669  loss_prob: 1.4807  loss_thr: 0.8647  loss_db: 0.2215
-2023/02/24 05:31:16 - mmengine - INFO - Epoch(train)    [4][15/22]  lr: 6.9842e-03  eta: 3 days, 23:16:49  time: 5.4703  data_time: 0.0655  memory: 6712  loss: 2.5318  loss_prob: 1.4490  loss_thr: 0.8641  loss_db: 0.2187
-2023/02/24 05:31:37 - mmengine - INFO - Epoch(train)    [4][20/22]  lr: 6.9842e-03  eta: 3 days, 19:28:30  time: 4.1855  data_time: 0.0463  memory: 6712  loss: 2.4536  loss_prob: 1.3779  loss_thr: 0.8595  loss_db: 0.2161
-2023/02/24 05:31:43 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 05:34:41 - mmengine - INFO - Epoch(train)    [5][ 5/22]  lr: 6.9790e-03  eta: 4 days, 3:03:02  time: 19.6819  data_time: 6.5648  memory: 6712  loss: 2.2837  loss_prob: 1.2531  loss_thr: 0.8280  loss_db: 0.2027
-2023/02/24 05:35:13 - mmengine - INFO - Epoch(train)    [5][10/22]  lr: 6.9790e-03  eta: 4 days, 0:23:14  time: 20.9855  data_time: 6.6279  memory: 6712  loss: 2.2122  loss_prob: 1.1990  loss_thr: 0.8168  loss_db: 0.1964
-2023/02/24 05:35:36 - mmengine - INFO - Epoch(train)    [5][15/22]  lr: 6.9790e-03  eta: 3 days, 21:16:29  time: 5.4636  data_time: 0.0946  memory: 6712  loss: 2.1482  loss_prob: 1.1455  loss_thr: 0.8120  loss_db: 0.1906
-2023/02/24 05:35:57 - mmengine - INFO - Epoch(train)    [5][20/22]  lr: 6.9790e-03  eta: 3 days, 18:24:00  time: 4.3929  data_time: 0.0363  memory: 6712  loss: 2.2215  loss_prob: 1.2052  loss_thr: 0.8195  loss_db: 0.1968
-2023/02/24 05:36:05 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 05:39:01 - mmengine - INFO - Epoch(train)    [6][ 5/22]  lr: 6.9737e-03  eta: 4 days, 0:33:26  time: 19.8343  data_time: 6.6865  memory: 6712  loss: 2.2092  loss_prob: 1.1873  loss_thr: 0.8270  loss_db: 0.1949
-2023/02/24 05:39:35 - mmengine - INFO - Epoch(train)    [6][10/22]  lr: 6.9737e-03  eta: 3 days, 22:34:41  time: 21.0220  data_time: 6.7316  memory: 6712  loss: 2.0882  loss_prob: 1.0934  loss_thr: 0.8093  loss_db: 0.1856
-2023/02/24 05:39:55 - mmengine - INFO - Epoch(train)    [6][15/22]  lr: 6.9737e-03  eta: 3 days, 19:56:56  time: 5.3949  data_time: 0.0639  memory: 6712  loss: 2.0953  loss_prob: 1.1014  loss_thr: 0.8072  loss_db: 0.1867
-2023/02/24 05:40:15 - mmengine - INFO - Epoch(train)    [6][20/22]  lr: 6.9737e-03  eta: 3 days, 17:30:13  time: 3.9802  data_time: 0.0307  memory: 6712  loss: 2.1803  loss_prob: 1.1807  loss_thr: 0.8064  loss_db: 0.1932
-2023/02/24 05:40:24 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 05:43:18 - mmengine - INFO - Epoch(train)    [7][ 5/22]  lr: 6.9685e-03  eta: 3 days, 22:38:09  time: 19.3378  data_time: 6.0656  memory: 6712  loss: 2.1125  loss_prob: 1.1454  loss_thr: 0.7801  loss_db: 0.1870
-2023/02/24 05:43:52 - mmengine - INFO - Epoch(train)    [7][10/22]  lr: 6.9685e-03  eta: 3 days, 21:03:26  time: 20.8409  data_time: 6.1127  memory: 6712  loss: 2.1082  loss_prob: 1.1444  loss_thr: 0.7752  loss_db: 0.1886
-2023/02/24 05:44:14 - mmengine - INFO - Epoch(train)    [7][15/22]  lr: 6.9685e-03  eta: 3 days, 18:57:55  time: 5.6460  data_time: 0.0896  memory: 6712  loss: 2.0828  loss_prob: 1.1309  loss_thr: 0.7652  loss_db: 0.1867
-2023/02/24 05:44:35 - mmengine - INFO - Epoch(train)    [7][20/22]  lr: 6.9685e-03  eta: 3 days, 16:56:45  time: 4.2613  data_time: 0.0588  memory: 6712  loss: 1.9454  loss_prob: 1.0347  loss_thr: 0.7355  loss_db: 0.1752
-2023/02/24 05:44:42 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 05:47:42 - mmengine - INFO - Epoch(train)    [8][ 5/22]  lr: 6.9632e-03  eta: 3 days, 21:35:37  time: 20.0738  data_time: 7.0659  memory: 6712  loss: 1.9103  loss_prob: 1.0182  loss_thr: 0.7198  loss_db: 0.1723
-2023/02/24 05:48:18 - mmengine - INFO - Epoch(train)    [8][10/22]  lr: 6.9632e-03  eta: 3 days, 20:19:25  time: 21.6464  data_time: 7.0947  memory: 6712  loss: 1.9593  loss_prob: 1.0665  loss_thr: 0.7176  loss_db: 0.1751
-2023/02/24 05:48:41 - mmengine - INFO - Epoch(train)    [8][15/22]  lr: 6.9632e-03  eta: 3 days, 18:33:12  time: 5.8713  data_time: 0.0769  memory: 6712  loss: 1.9544  loss_prob: 1.0733  loss_thr: 0.7049  loss_db: 0.1762
-2023/02/24 05:49:01 - mmengine - INFO - Epoch(train)    [8][20/22]  lr: 6.9632e-03  eta: 3 days, 16:48:01  time: 4.3373  data_time: 0.0467  memory: 6712  loss: 1.8306  loss_prob: 0.9863  loss_thr: 0.6770  loss_db: 0.1673
-2023/02/24 05:49:08 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 05:52:08 - mmengine - INFO - Epoch(train)    [9][ 5/22]  lr: 6.9580e-03  eta: 3 days, 20:51:50  time: 20.0004  data_time: 6.3228  memory: 6712  loss: 1.9089  loss_prob: 1.0586  loss_thr: 0.6772  loss_db: 0.1731
-2023/02/24 05:52:41 - mmengine - INFO - Epoch(train)    [9][10/22]  lr: 6.9580e-03  eta: 3 days, 19:38:00  time: 21.3337  data_time: 6.3790  memory: 6712  loss: 1.8955  loss_prob: 1.0480  loss_thr: 0.6761  loss_db: 0.1714
-2023/02/24 05:53:02 - mmengine - INFO - Epoch(train)    [9][15/22]  lr: 6.9580e-03  eta: 3 days, 17:59:55  time: 5.3263  data_time: 0.0722  memory: 6712  loss: 1.7788  loss_prob: 0.9520  loss_thr: 0.6654  loss_db: 0.1614
-2023/02/24 05:53:21 - mmengine - INFO - Epoch(train)    [9][20/22]  lr: 6.9580e-03  eta: 3 days, 16:25:34  time: 4.0420  data_time: 0.0361  memory: 6712  loss: 1.8003  loss_prob: 0.9682  loss_thr: 0.6678  loss_db: 0.1643
-2023/02/24 05:53:31 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 05:56:27 - mmengine - INFO - Epoch(train)   [10][ 5/22]  lr: 6.9527e-03  eta: 3 days, 20:00:04  time: 19.6905  data_time: 6.3250  memory: 6712  loss: 1.8357  loss_prob: 0.9859  loss_thr: 0.6834  loss_db: 0.1663
-2023/02/24 05:57:04 - mmengine - INFO - Epoch(train)   [10][10/22]  lr: 6.9527e-03  eta: 3 days, 19:04:40  time: 21.3082  data_time: 6.3498  memory: 6712  loss: 1.8376  loss_prob: 0.9889  loss_thr: 0.6809  loss_db: 0.1677
-2023/02/24 05:57:27 - mmengine - INFO - Epoch(train)   [10][15/22]  lr: 6.9527e-03  eta: 3 days, 17:43:00  time: 6.0559  data_time: 0.0570  memory: 6712  loss: 1.7998  loss_prob: 0.9688  loss_thr: 0.6660  loss_db: 0.1651
-2023/02/24 05:57:48 - mmengine - INFO - Epoch(train)   [10][20/22]  lr: 6.9527e-03  eta: 3 days, 16:20:24  time: 4.4162  data_time: 0.0306  memory: 6712  loss: 1.8812  loss_prob: 1.0357  loss_thr: 0.6779  loss_db: 0.1676
-2023/02/24 05:57:55 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 06:01:01 - mmengine - INFO - Epoch(train)   [11][ 5/22]  lr: 6.9474e-03  eta: 3 days, 19:46:57  time: 20.5701  data_time: 7.2858  memory: 6712  loss: 1.8385  loss_prob: 1.0164  loss_thr: 0.6580  loss_db: 0.1641
-2023/02/24 06:01:42 - mmengine - INFO - Epoch(train)   [11][10/22]  lr: 6.9474e-03  eta: 3 days, 19:04:25  time: 22.6877  data_time: 7.3177  memory: 6712  loss: 1.7372  loss_prob: 0.9383  loss_thr: 0.6403  loss_db: 0.1586
-2023/02/24 06:02:04 - mmengine - INFO - Epoch(train)   [11][15/22]  lr: 6.9474e-03  eta: 3 days, 17:48:05  time: 6.3309  data_time: 0.0664  memory: 6712  loss: 1.8261  loss_prob: 1.0116  loss_thr: 0.6501  loss_db: 0.1644
-2023/02/24 06:02:27 - mmengine - INFO - Epoch(train)   [11][20/22]  lr: 6.9474e-03  eta: 3 days, 16:36:23  time: 4.4944  data_time: 0.0463  memory: 6712  loss: 1.8030  loss_prob: 0.9974  loss_thr: 0.6439  loss_db: 0.1618
-2023/02/24 06:02:35 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 06:05:43 - mmengine - INFO - Epoch(train)   [12][ 5/22]  lr: 6.9422e-03  eta: 3 days, 19:50:54  time: 21.0802  data_time: 6.7792  memory: 6712  loss: 1.7311  loss_prob: 0.9384  loss_thr: 0.6339  loss_db: 0.1588
-2023/02/24 06:06:16 - mmengine - INFO - Epoch(train)   [12][10/22]  lr: 6.9422e-03  eta: 3 days, 18:57:51  time: 22.1269  data_time: 6.7959  memory: 6712  loss: 1.7188  loss_prob: 0.9327  loss_thr: 0.6281  loss_db: 0.1580
-2023/02/24 06:06:38 - mmengine - INFO - Epoch(train)   [12][15/22]  lr: 6.9422e-03  eta: 3 days, 17:47:34  time: 5.4922  data_time: 0.0768  memory: 6712  loss: 1.7922  loss_prob: 0.9895  loss_thr: 0.6431  loss_db: 0.1596
-2023/02/24 06:06:59 - mmengine - INFO - Epoch(train)   [12][20/22]  lr: 6.9422e-03  eta: 3 days, 16:38:54  time: 4.2930  data_time: 0.0734  memory: 6712  loss: 1.8091  loss_prob: 1.0073  loss_thr: 0.6390  loss_db: 0.1628
-2023/02/24 06:07:07 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 06:10:14 - mmengine - INFO - Epoch(train)   [13][ 5/22]  lr: 6.9369e-03  eta: 3 days, 19:34:46  time: 20.8475  data_time: 6.4862  memory: 6712  loss: 1.7225  loss_prob: 0.9462  loss_thr: 0.6158  loss_db: 0.1605
-2023/02/24 06:10:45 - mmengine - INFO - Epoch(train)   [13][10/22]  lr: 6.9369e-03  eta: 3 days, 18:42:55  time: 21.8411  data_time: 6.5244  memory: 6712  loss: 1.6861  loss_prob: 0.9208  loss_thr: 0.6085  loss_db: 0.1568
-2023/02/24 06:11:08 - mmengine - INFO - Epoch(train)   [13][15/22]  lr: 6.9369e-03  eta: 3 days, 17:39:18  time: 5.3518  data_time: 0.0755  memory: 6712  loss: 1.6869  loss_prob: 0.9212  loss_thr: 0.6091  loss_db: 0.1566
-2023/02/24 06:11:29 - mmengine - INFO - Epoch(train)   [13][20/22]  lr: 6.9369e-03  eta: 3 days, 16:36:44  time: 4.4030  data_time: 0.0427  memory: 6712  loss: 1.6707  loss_prob: 0.9171  loss_thr: 0.5988  loss_db: 0.1549
-2023/02/24 06:11:39 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 06:14:33 - mmengine - INFO - Epoch(train)   [14][ 5/22]  lr: 6.9317e-03  eta: 3 days, 19:01:21  time: 19.7550  data_time: 6.6183  memory: 6712  loss: 1.7619  loss_prob: 1.0020  loss_thr: 0.6010  loss_db: 0.1589
-2023/02/24 06:15:10 - mmengine - INFO - Epoch(train)   [14][10/22]  lr: 6.9317e-03  eta: 3 days, 18:23:36  time: 21.1018  data_time: 6.6633  memory: 6712  loss: 1.7161  loss_prob: 0.9654  loss_thr: 0.5944  loss_db: 0.1563
-2023/02/24 06:15:32 - mmengine - INFO - Epoch(train)   [14][15/22]  lr: 6.9317e-03  eta: 3 days, 17:22:58  time: 5.8873  data_time: 0.0648  memory: 6712  loss: 1.7192  loss_prob: 0.9679  loss_thr: 0.5954  loss_db: 0.1559
-2023/02/24 06:15:54 - mmengine - INFO - Epoch(train)   [14][20/22]  lr: 6.9317e-03  eta: 3 days, 16:25:59  time: 4.3364  data_time: 0.0274  memory: 6712  loss: 1.6298  loss_prob: 0.8869  loss_thr: 0.5926  loss_db: 0.1503
-2023/02/24 06:16:02 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 06:19:00 - mmengine - INFO - Epoch(train)   [15][ 5/22]  lr: 6.9264e-03  eta: 3 days, 18:44:04  time: 19.8166  data_time: 6.2719  memory: 6712  loss: 1.6233  loss_prob: 0.8843  loss_thr: 0.5895  loss_db: 0.1495
-2023/02/24 06:19:35 - mmengine - INFO - Epoch(train)   [15][10/22]  lr: 6.9264e-03  eta: 3 days, 18:05:44  time: 21.3018  data_time: 6.3262  memory: 6712  loss: 1.6084  loss_prob: 0.8760  loss_thr: 0.5845  loss_db: 0.1478
-2023/02/24 06:19:56 - mmengine - INFO - Epoch(train)   [15][15/22]  lr: 6.9264e-03  eta: 3 days, 17:09:18  time: 5.6332  data_time: 0.0798  memory: 6712  loss: 1.5740  loss_prob: 0.8612  loss_thr: 0.5668  loss_db: 0.1460
-2023/02/24 06:20:17 - mmengine - INFO - Epoch(train)   [15][20/22]  lr: 6.9264e-03  eta: 3 days, 16:14:55  time: 4.2267  data_time: 0.0394  memory: 6712  loss: 1.6627  loss_prob: 0.9368  loss_thr: 0.5743  loss_db: 0.1516
-2023/02/24 06:20:27 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 06:23:29 - mmengine - INFO - Epoch(train)   [16][ 5/22]  lr: 6.9211e-03  eta: 3 days, 18:31:35  time: 20.4682  data_time: 6.4376  memory: 6712  loss: 1.6751  loss_prob: 0.9439  loss_thr: 0.5791  loss_db: 0.1521
-2023/02/24 06:24:03 - mmengine - INFO - Epoch(train)   [16][10/22]  lr: 6.9211e-03  eta: 3 days, 17:53:41  time: 21.5603  data_time: 6.4834  memory: 6712  loss: 1.5881  loss_prob: 0.8699  loss_thr: 0.5714  loss_db: 0.1468
-2023/02/24 06:24:24 - mmengine - INFO - Epoch(train)   [16][15/22]  lr: 6.9211e-03  eta: 3 days, 17:01:54  time: 5.5439  data_time: 0.0674  memory: 6712  loss: 1.5751  loss_prob: 0.8581  loss_thr: 0.5713  loss_db: 0.1457
-2023/02/24 06:24:46 - mmengine - INFO - Epoch(train)   [16][20/22]  lr: 6.9211e-03  eta: 3 days, 16:11:19  time: 4.3338  data_time: 0.0365  memory: 6712  loss: 1.6895  loss_prob: 0.9474  loss_thr: 0.5892  loss_db: 0.1528
-2023/02/24 06:24:54 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 06:27:53 - mmengine - INFO - Epoch(train)   [17][ 5/22]  lr: 6.9159e-03  eta: 3 days, 18:13:54  time: 20.1179  data_time: 7.1602  memory: 6712  loss: 1.5890  loss_prob: 0.8658  loss_thr: 0.5758  loss_db: 0.1473
-2023/02/24 06:28:29 - mmengine - INFO - Epoch(train)   [17][10/22]  lr: 6.9159e-03  eta: 3 days, 17:40:34  time: 21.5151  data_time: 7.1956  memory: 6712  loss: 1.5827  loss_prob: 0.8728  loss_thr: 0.5623  loss_db: 0.1476
-2023/02/24 06:28:52 - mmengine - INFO - Epoch(train)   [17][15/22]  lr: 6.9159e-03  eta: 3 days, 16:53:08  time: 5.8176  data_time: 0.0500  memory: 6712  loss: 1.5498  loss_prob: 0.8583  loss_thr: 0.5468  loss_db: 0.1447
-2023/02/24 06:29:14 - mmengine - INFO - Epoch(train)   [17][20/22]  lr: 6.9159e-03  eta: 3 days, 16:06:35  time: 4.5159  data_time: 0.0371  memory: 6712  loss: 1.5092  loss_prob: 0.8323  loss_thr: 0.5363  loss_db: 0.1406
-2023/02/24 06:29:21 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 06:32:11 - mmengine - INFO - Epoch(train)   [18][ 5/22]  lr: 6.9106e-03  eta: 3 days, 17:49:46  time: 19.0947  data_time: 6.1514  memory: 6712  loss: 1.6499  loss_prob: 0.9514  loss_thr: 0.5495  loss_db: 0.1489
-2023/02/24 06:32:41 - mmengine - INFO - Epoch(train)   [18][10/22]  lr: 6.9106e-03  eta: 3 days, 17:13:04  time: 19.9988  data_time: 6.1911  memory: 6712  loss: 1.5199  loss_prob: 0.8403  loss_thr: 0.5375  loss_db: 0.1421
-2023/02/24 06:33:02 - mmengine - INFO - Epoch(train)   [18][15/22]  lr: 6.9106e-03  eta: 3 days, 16:26:56  time: 5.1812  data_time: 0.0669  memory: 6712  loss: 1.6338  loss_prob: 0.9296  loss_thr: 0.5540  loss_db: 0.1502
-2023/02/24 06:33:23 - mmengine - INFO - Epoch(train)   [18][20/22]  lr: 6.9106e-03  eta: 3 days, 15:41:20  time: 4.1944  data_time: 0.0443  memory: 6712  loss: 1.6014  loss_prob: 0.9109  loss_thr: 0.5433  loss_db: 0.1472
-2023/02/24 06:33:30 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 06:36:13 - mmengine - INFO - Epoch(train)   [19][ 5/22]  lr: 6.9054e-03  eta: 3 days, 17:11:46  time: 18.3289  data_time: 6.0725  memory: 6712  loss: 1.5865  loss_prob: 0.9040  loss_thr: 0.5374  loss_db: 0.1451
-2023/02/24 06:36:44 - mmengine - INFO - Epoch(train)   [19][10/22]  lr: 6.9054e-03  eta: 3 days, 16:38:02  time: 19.4111  data_time: 6.1041  memory: 6712  loss: 1.5683  loss_prob: 0.8961  loss_thr: 0.5286  loss_db: 0.1436
-2023/02/24 06:37:03 - mmengine - INFO - Epoch(train)   [19][15/22]  lr: 6.9054e-03  eta: 3 days, 15:52:19  time: 5.0075  data_time: 0.0461  memory: 6712  loss: 1.4666  loss_prob: 0.8143  loss_thr: 0.5145  loss_db: 0.1378
-2023/02/24 06:37:22 - mmengine - INFO - Epoch(train)   [19][20/22]  lr: 6.9054e-03  eta: 3 days, 15:07:51  time: 3.8092  data_time: 0.0239  memory: 6712  loss: 1.4776  loss_prob: 0.8202  loss_thr: 0.5185  loss_db: 0.1389
-2023/02/24 06:37:30 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 06:40:10 - mmengine - INFO - Epoch(train)   [20][ 5/22]  lr: 6.9001e-03  eta: 3 days, 16:31:37  time: 18.0048  data_time: 5.7458  memory: 6712  loss: 1.5888  loss_prob: 0.9128  loss_thr: 0.5328  loss_db: 0.1432
-2023/02/24 06:40:37 - mmengine - INFO - Epoch(train)   [20][10/22]  lr: 6.9001e-03  eta: 3 days, 15:56:20  time: 18.7845  data_time: 5.7862  memory: 6712  loss: 1.6013  loss_prob: 0.9205  loss_thr: 0.5363  loss_db: 0.1445
-2023/02/24 06:40:58 - mmengine - INFO - Epoch(train)   [20][15/22]  lr: 6.9001e-03  eta: 3 days, 15:14:41  time: 4.7755  data_time: 0.0685  memory: 6712  loss: 1.4801  loss_prob: 0.8185  loss_thr: 0.5228  loss_db: 0.1389
-2023/02/24 06:41:17 - mmengine - INFO - Epoch(train)   [20][20/22]  lr: 6.9001e-03  eta: 3 days, 14:32:56  time: 3.9535  data_time: 0.0450  memory: 6712  loss: 1.4580  loss_prob: 0.8092  loss_thr: 0.5116  loss_db: 0.1372
-2023/02/24 06:41:23 - mmengine - INFO - Exp name: dbnet_resnet18_fpnc_1200e_icdar2015_20230224_051330
-2023/02/24 06:41:23 - mmengine - INFO - Saving checkpoint at 20 epochs
-2023/02/24 06:43:59 - mmengine - INFO - Epoch(val)   [20][ 5/88]    eta: 0:42:55  time: 31.0259  data_time: 0.0756  memory: 8651  

model/det/config.py

@@ -3,16 +3,16 @@ model = dict(
     type='DBNet',
     backbone=dict(
         type='mmdet.ResNet',
-        depth=18,
+        depth=50,
         num_stages=4,
         out_indices=(0, 1, 2, 3),
         frozen_stages=-1,
         norm_cfg=dict(type='BN', requires_grad=True),
-        init_cfg=dict(type='Pretrained', checkpoint='torchvision://resnet18'),
+        init_cfg=dict(type='Pretrained', checkpoint='torchvision://resnet50'),
         norm_eval=False,
         style='caffe'),
     neck=dict(
-        type='FPNC', in_channels=[64, 128, 256, 512], lateral_channels=256),
+        type='FPNC', in_channels=[256, 512, 1024, 2048], lateral_channels=256),
     det_head=dict(
         type='DBHead',
         in_channels=256,
@@ -170,7 +170,7 @@ default_hooks = dict(
     timer=dict(type='IterTimerHook'),
     logger=dict(type='LoggerHook', interval=5),
     param_scheduler=dict(type='ParamSchedulerHook'),
-    checkpoint=dict(type='CheckpointHook', interval=5),
+    checkpoint=dict(type='CheckpointHook', interval=10),
     sampler_seed=dict(type='DistSamplerSeedHook'),
     sync_buffer=dict(type='SyncBuffersHook'),
     visualization=dict(
@@ -191,13 +191,13 @@ visualizer = dict(
     type='TextDetLocalVisualizer',
     name='visualizer',
     vis_backends=[dict(type='LocalVisBackend')])
-max_epochs = 30
+max_epochs = 100
 optim_wrapper = dict(
     type='OptimWrapper', optimizer=dict(type='Adam', lr=0.001))
-train_cfg = dict(type='EpochBasedTrainLoop', max_epochs=30, val_interval=10)
+train_cfg = dict(type='EpochBasedTrainLoop', max_epochs=100, val_interval=10)
 val_cfg = dict(type='ValLoop')
 test_cfg = dict(type='TestLoop')
-param_scheduler = [dict(type='PolyLR', power=0.9, end=30)]
+param_scheduler = [dict(type='PolyLR', power=0.9, end=100)]
 thvotecount_textdet_train = dict(
     type='OCRDataset',
     data_root='data/det/vl+vc-textdet',
@@ -265,7 +265,7 @@ thvotecount_textdet_test = dict(
             meta_keys=('img_path', 'ori_shape', 'img_shape', 'scale_factor'))
     ])
 train_dataloader = dict(
-    batch_size=40,
+    batch_size=16,
     num_workers=8,
     persistent_workers=True,
     sampler=dict(type='DefaultSampler', shuffle=True),
@@ -315,7 +315,7 @@ train_dataloader = dict(
                 meta_keys=('img_path', 'ori_shape', 'img_shape'))
         ]))
 val_dataloader = dict(
-    batch_size=8,
+    batch_size=2,
     num_workers=4,
     persistent_workers=True,
     sampler=dict(type='DefaultSampler', shuffle=False),
@@ -342,7 +342,7 @@ val_dataloader = dict(
                            'scale_factor'))
         ]))
 test_dataloader = dict(
-    batch_size=8,
+    batch_size=2,
     num_workers=4,
     persistent_workers=True,
     sampler=dict(type='DefaultSampler', shuffle=False),
@@ -370,4 +370,4 @@ test_dataloader = dict(
         ]))
 auto_scale_lr = dict(base_batch_size=100)
 launcher = 'none'
-work_dir = './work_dirs/dbnet_resnet18_fpnc_votecount'
+work_dir = './work_dirs/votecount'

model/det/model.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e4e79675b6bab2c627fbf19196dd0baf57c7016958892e75ffb0eed0bc3c467b
-size 148596609
+oid sha256:749e01b5c50326bce4c9659ce50938da4ce28e6e6fca788a68c7af97851bc562
+size 307332579