Installation and Compatibility Issues with mmcv-full and mmsegmentation #17
Description
I‘m pity that I can not run this project for the cuda version=11.7.
I'm encountering significant problems while trying to set up this project due to my CUDA version, which is 11.7. No matter what I try, I am unable to install mmcv-full==1.3.0. The root cause appears to be a mismatch between my CUDA 11.7 version and the compatible PyTorch version required by mmcv-full==1.3.0.
During the installation process of mmcv-full==1.3.0, I received the following error messages:
ERROR: Failed building wheel for mmcv-full
Running setup.py clean for mmcv-full
Failed to build mmcv-full
ERROR: Could not build wheels for mmcv-full, which is required to install pyproject.toml-based projects
After this unsuccessful attempt, I proceeded to install mmsegmentation==0.28.0 via pip install. However, this led to another issue. When I tried to run the project, I encountered an error stating that VIT_MLA is not registered, preventing the project from running properly.
I would greatly appreciate any guidance or suggestions on how to resolve these compatibility and installation issues, either by finding a way to make mmcv-full==1.3.0 work with my CUDA 11.7 setup, or by addressing the VIT_MLA registration problem in mmsegmentation==0.28.0.
[2025-05-09 10:28:12,899] [semantic.py:304] [INFO] running semantic seg model!
/home/anaconda3/envs/CC/lib/python3.9/site-packages/mmseg/models/builder.py:41: UserWarning: train_cfg and test_cfg is deprecated, please specify them in model
warnings.warn(
Traceback (most recent call last):
File "/home/anaconda3/envs/CC/lib/python3.9/site-packages/mmcv/utils/registry.py", line 69, in build_from_cfg
return obj_cls(**args)
File "/home/anaconda3/envs/CC/lib/python3.9/site-packages/mmseg/models/segmentors/encoder_decoder.py", line 36, in __init__
self.backbone = builder.build_backbone(backbone)
File "/home/anaconda3/envs/CC/lib/python3.9/site-packages/mmseg/models/builder.py", line 20, in build_backbone
return BACKBONES.build(cfg)
File "/home/anaconda3/envs/CC/lib/python3.9/site-packages/mmcv/utils/registry.py", line 237, in build
return self.build_func(*args, **kwargs, registry=self)
File "/home/anaconda3/envs/CC/lib/python3.9/site-packages/mmcv/cnn/builder.py", line 27, in build_model_from_cfg
return build_from_cfg(cfg, registry, default_args)
File "/home/anaconda3/envs/CC/lib/python3.9/site-packages/mmcv/utils/registry.py", line 61, in build_from_cfg
raise KeyError(
KeyError: 'VIT_MLA is not in the models registry'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/home/expr_SS/FoodSAM-main/FoodSAM/semantic.py", line 323, in <module>
main(args)
File "/home/expr_SS/FoodSAM-main/FoodSAM/semantic.py", line 305, in main
semantic_predict(args.data_root, args.img_dir, args.ann_dir, args.semantic_config, args.options, args.aug_test, args.semantic_checkpoint, args.eval_options, args.output, args.color_list_path, args.img_path)
File "/home/expr_SS/FoodSAM-main/FoodSAM/FoodSAM_tools/predict_semantic_mask.py", line 184, in semantic_predict
model = init_segmentor(config, checkpoint)
File "/home/anaconda3/envs/CC/lib/python3.9/site-packages/mmseg/apis/inference.py", line 32, in init_segmentor
model = build_segmentor(config.model, test_cfg=config.get('test_cfg'))
File "/home/anaconda3/envs/CC/lib/python3.9/site-packages/mmseg/models/builder.py", line 48, in build_segmentor
return SEGMENTORS.build(
File "/home/anaconda3/envs/CC/lib/python3.9/site-packages/mmcv/utils/registry.py", line 237, in build
return self.build_func(*args, **kwargs, registry=self)
File "/home/anaconda3/envs/CC/lib/python3.9/site-packages/mmcv/cnn/builder.py", line 27, in build_model_from_cfg
return build_from_cfg(cfg, registry, default_args)
File "/home/anaconda3/envs/CC/lib/python3.9/site-packages/mmcv/utils/registry.py", line 72, in build_from_cfg
raise type(e)(f'{obj_cls.__name__}: {e}')
KeyError: "EncoderDecoder: 'VIT_MLA is not in the models registry'"
Thank you in advance for your help!