Open Alexwong-1024 opened 11 months ago
Please check that the versions of your installed packages are consistent with those in the INSTALL instruction file. ---- Replied Message ---- | From | @.> | | Date | 11/06/2023 20:41 | | To | ShichengMiao16/MADet @.> | | Cc | Subscribed @.***> | | Subject | [ShichengMiao16/MADet] ConvModule' object has no attribute 'weight', (Issue #1) |
Please tell me, I use the be command python tools/train.py configs/madet/madet_r101_fpn_1x_coco.py for training, but an error like this occurs, nexpected key in source state_dict: fc.weight, fc.bias
Traceback (most recent call last): File "tools/train.py", line 185, in main() File "tools/train.py", line 158, in main model = build_detector( File "/root/autodl-tmp/MADet/mmdet/models/builder.py", line 67, in build_detector return build(cfg, DETECTORS, dict(train_cfg=train_cfg, test_cfg=test_cfg)) File "/root/autodl-tmp/MADet/mmdet/models/builder.py", line 32, in build return build_from_cfg(cfg, registry, default_args) File "/root/miniconda3/lib/python3.8/site-packages/mmcv/utils/registry.py", line 171, in build_from_cfg return obj_cls(**args) File "/root/autodl-tmp/MADet/mmdet/models/detectors/madet.py", line 18, in init super(MADet, self).init(backbone, neck, bbox_head, train_cfg, File "/root/autodl-tmp/MADet/mmdet/models/detectors/single_stage.py", line 33, in init self.init_weights(pretrained=pretrained) File "/root/autodl-tmp/MADet/mmdet/models/detectors/single_stage.py", line 50, in init_weights self.bbox_head.init_weights() File "/root/autodl-tmp/MADet/mmdet/models/dense_heads/madet_head.py", line 177, in init_weights normal_init(self.dcn, std=0.01) File "/root/miniconda3/lib/python3.8/site-packages/mmcv/cnn/utils/weight_init.py", line 24, in normalinit nn.init.normal(module.weight, mean, std) File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 771, in getattr raise ModuleAttributeError("'{}' object has no attribute '{}'".format( torch.nn.modules.module.ModuleAttributeError: 'ConvModule' object has no attribute 'weight', looking forward to your reply.
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you are subscribed to this thread.Message ID: @.***>
Thanks for the quick reply, I'll give it a try.
Thanks for the quick reply, I'll give it a try.
Hello, I encountered the same problem as you. How did you solve it? Can you provide specific methods or commands?
Thanks for the quick reply, I'll give it a try.
Hello, I encountered the same problem as you. How did you solve it? Can you provide specific methods or commands?
Hello,brother.I also encountered the same problem as you. Is it a problem with CUDA and Pytorch versions? Did you install CUDA9.2 and pytorch 1.3.1.
Please tell me, I use the be command python tools/train.py configs/madet/madet_r101_fpn_1x_coco.py for training, but an error like this occurs, nexpected key in source state_dict: fc.weight, fc.bias
Traceback (most recent call last): File "tools/train.py", line 185, in
main()
File "tools/train.py", line 158, in main
model = build_detector(
File "/root/autodl-tmp/MADet/mmdet/models/builder.py", line 67, in build_detector
return build(cfg, DETECTORS, dict(train_cfg=train_cfg, test_cfg=test_cfg))
File "/root/autodl-tmp/MADet/mmdet/models/builder.py", line 32, in build
return build_from_cfg(cfg, registry, default_args)
File "/root/miniconda3/lib/python3.8/site-packages/mmcv/utils/registry.py", line 171, in build_from_cfg
return obj_cls(**args)
File "/root/autodl-tmp/MADet/mmdet/models/detectors/madet.py", line 18, in init
super(MADet, self).init(backbone, neck, bbox_head, train_cfg,
File "/root/autodl-tmp/MADet/mmdet/models/detectors/single_stage.py", line 33, in init
self.init_weights(pretrained=pretrained)
File "/root/autodl-tmp/MADet/mmdet/models/detectors/single_stage.py", line 50, in init_weights
self.bbox_head.init_weights()
File "/root/autodl-tmp/MADet/mmdet/models/dense_heads/madet_head.py", line 177, in init_weights
normal_init(self.dcn, std=0.01)
File "/root/miniconda3/lib/python3.8/site-packages/mmcv/cnn/utils/weight_init.py", line 24, in normalinit
nn.init.normal(module.weight, mean, std)
File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 771, in getattr
raise ModuleAttributeError("'{}' object has no attribute '{}'".format(
torch.nn.modules.module.ModuleAttributeError: 'ConvModule' object has no attribute 'weight', looking forward to your reply.