TRAINING ERROR ON ADE20K

[EthanAbitbol3]

Hello everyone, and thank you for your work. I have just cloned the complete repository and installed all the necessary dependencies. I have tested DCNV3 :

and tried to initiate training using the InternImage configuration file to observe the learning process. However, it is not working as expected :

• Windows 10
• Package Version

---

addict 2.4.0 backcall 0.2.0 brotlipy 0.7.0 certifi 2022.12.7 cffi 1.15.1 charset-normalizer 2.0.4 click 8.1.3 colorama 0.4.6 cryptography 39.0.1 cycler 0.11.0 DCNv3 1.0 debugpy 1.5.1 decorator 5.1.1 entrypoints 0.4 filelock 3.12.0 fonttools 4.38.0 fsspec 2023.1.0 huggingface-hub 0.15.1 idna 3.4 importlib-metadata 6.6.0 ipykernel 6.15.2 ipython 7.31.1 jedi 0.18.1 jupyter_client 7.4.9 jupyter_core 4.11.2 kiwisolver 1.4.4 Markdown 3.4.3 markdown-it-py 2.2.0 matplotlib 3.5.3 matplotlib-inline 0.1.6 mdurl 0.1.2 mkl-fft 1.3.1 mkl-random 1.2.2 mkl-service 2.4.0 mmcls 0.25.0 mmcv-full 1.5.0 mmdet 2.28.1 mmsegmentation 0.27.0 model-index 0.1.11 nest-asyncio 1.5.6 numpy 1.21.5 opencv-python 4.7.0.72 openmim 0.3.7 ordered-set 4.1.0 packaging 22.0 pandas 1.3.5 parso 0.8.3 pickleshare 0.7.5 Pillow 9.4.0 pip 22.3.1 prettytable 3.7.0 prompt-toolkit 3.0.36 psutil 5.9.0 pycocotools 2.0.6 pycparser 2.21 Pygments 2.15.1 pyOpenSSL 23.0.0 pyparsing 3.0.9 PySocks 1.7.1 python-dateutil 2.8.2 pytz 2023.3 pywin32 305.1 PyYAML 6.0 pyzmq 23.2.0 regex 2023.6.3 requests 2.28.1 rich 13.4.1 scipy 1.7.3 setuptools 65.6.3 six 1.16.0 tabulate 0.9.0 termcolor 2.3.0 terminaltables 3.1.10 timm 0.6.11 tomli 2.0.1 torch 1.11.0+cu113 torchvision 0.12.0+cu113 tornado 6.2 tqdm 4.65.0 traitlets 5.7.1 typing_extensions 4.3.0 urllib3 1.26.14 wcwidth 0.2.5 wheel 0.38.4 win-inet-pton 1.1.0 wincertstore 0.2 yacs 0.1.8 yapf 0.33.0 zipp 3.15.0

[daniel-code]

Try to modify the path join methods in train.py

1. from pathlib import Path
2. https://github.com/OpenGVLab/InternImage/blob/master/segmentation/train.py#L132 original

   cfg.work_dir = osp.join('./work_dirs',
                                   osp.splitext(osp.basename(args.config))[0])
   

   to

   cfg.work_dir = Path(osp.join('./work_dirs',
                                   osp.splitext(osp.basename(args.config))[0])).as_posix()
   

3. https://github.com/OpenGVLab/InternImage/blob/master/segmentation/train.py#L167 original

   cfg.dump(osp.join(cfg.work_dir, osp.basename(args.config)))
   

   to

   cfg.dump(Path(osp.join(cfg.work_dir, osp.basename(args.config))).as_posix())