BizhuWu
commented
1 year ago Can you give me more details about the version of MMDetection you have installed?
Closed majintao1994 closed 10 months ago
BizhuWu
commented
1 year ago Can you give me more details about the version of MMDetection you have installed?
After I have extracted the weights of the pre training model, I need to ask the following questions when using this backone.pth to train the downstream detection tasks. What's wrong? Thank you
error： Traceback (most recent call last): File "/home/work/mmselftest/mmself/tools/train.py", line 187, in
main()
File "/home/work/mmselftest/mmself/tools/train.py", line 161, in main
test_cfg=cfg.get('test_cfg'))
File "/home/work/mmselftest/mmself/mmdet/models/builder.py", line 77, in build_detector
return build(cfg, DETECTORS, dict(train_cfg=train_cfg, test_cfg=test_cfg))
File "/home/work/mmselftest/mmself/mmdet/models/builder.py", line 34, in build
return build_from_cfg(cfg, registry, default_args)
File "/home/ty/anaconda3/envs/trt2/lib/python3.7/site-packages/mmcv/utils/registry.py", line 182, in build_from_cfg
raise type(e)(f'{obj_cls.name}: {e}')
TypeError: FasterRCNN: ResNet: init() got an unexpected keyword argument 'init_cfg'
my config file;
base = 'faster_rcnn_r50_fpn_1x_coco.py'
model = dict( backbone=dict( frozen_stages=-1, init_cfg=dict( type='Pretrained', checkpoint='checkpoints/backbone_200.pth') ) )
optimizer
optimizer = dict( lr=0.02 * (1 / 8) )
Runner type
runner = dict(delete=True, type='IterBasedRunner', max_iters=9000)
checkpoint_config = dict(interval=3000)
evaluation = dict(interval=3000)