Su YR

icon indicating a website link https://www.linkedin.com/in/suyanru/

icon location Taiwan icon location Computer Vision, Deep Learning, Image Segmentation, Video Classification

Results 3 comments of Su YR

Is DCNV3 supported on Windows?

I have built DCNv3 from the source CUDA code on Windows (CUDA 11.3). It can inference, but there is an issue in the training script. ``` yapf.yapflib.verifier.InternalError: (unicode error) 'unicodeescape'...

TRAINING ERROR ON ADE20K

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

Number of Tokens different than papers

Thank you for reporting the issue. The function is fixed by #26 This function is only used for calculating the position embedding. The tokens are generated by [F.conv3d](https://github.com/daniel-code/TubeViT/blob/main/tubevit/model.py#L79) in `SparseTubesTokenizer`,...