search

SwinTransformer / Swin-Transformer-Object-Detection

This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows" on Object Detection and Instance Segmentation.
https://arxiv.org/abs/2103.14030
Apache License 2.0
1.79k stars 377 forks source link

Debugging with VS Code #94

Open LaCandela opened 2 years ago

LaCandela commented 2 years ago

When I try to debug with VS Code I get the following error:

Exception has occurred: AttributeError (note: full exception trace is shown but execution is paused at: getattr) 'ConfigDict' object has no attribute 'frozen' File "/mmcv/mmcv/utils/config.py", line 48, in getattr__ (Current frame) raise ex File "/mmcv/mmcv/utils/registry.py", line 34, in build_from_cfg args = cfg.copy() File "/Swin-Transformer-Object-Detection/mmdet/models/builder.py", line 34, in build return build_from_cfg(cfg, registry, default_args) File "/Swin-Transformer-Object-Detection/mmdet/models/builder.py", line 77, in build_detector return build(cfg, DETECTORS, dict(train_cfg=train_cfg, test_cfg=test_cfg)) File "/Swin-Transformer-Object-Detection/tools/train.py", line 158, in main model = build_detector( File "/Swin-Transformer-Object-Detection/tools/train.py", line 187, in main()

However, when I run the code from command line it works fine. Seems like a VS Code issue. Does anywone experienced the same and has probably a solution?

wtliao commented 2 years ago

Hi @LaCandela , I am encountering the same issue as yours. Have you solved it?

LaCandela commented 2 years ago

Hi @LaCandela , I am encountering the same issue as yours. Have you solved it?

Not yet, I made a workaround and replaced the .copy() function calls on the config object with copy.deepcopy() where VS code complains. Now debugging is possible in VS Code but the loss doesn't decreasing so there is still something wrong...