yangwenjie

It is very unfriendly to directly pass the entire config as a argument to a function, e.g., https://github.com/ArrowLuo/CLIP4Clip/blob/508ffa3de39ba0563a03199c440ab602a72e9b6f/modules/modeling.py#L30

First of all, thanks for your open-source codes. Then, I want to make a request. Currently, the [evaluate_movingfashion.py](https://github.com/HumaticsLAB/SEAM-Match-RCNN/blob/f632e489c6c8ae68afd33f749add4e1a18a7bcb9/evaluate_movingfashion.py#L15) was custom made just for the SEAM-Match-RCNN model. For example, the [temporal_aggregator](https://github.com/HumaticsLAB/SEAM-Match-RCNN/blob/f632e489c6c8ae68afd33f749add4e1a18a7bcb9/evaluate_movingfashion.py#L42)...

## 🚀 Feature Thanks for your excellent work！ As mentioned in the paper， `we first train a discrete VAE as the image tokenizer on our collected fashion images with the...

### Branch main branch (mmpretrain version) ### Describe the bug This project wasted nearly 12 hours of my time solving the incompatibility issues between mmcv, mmdet, mmtrack, and numpy. I...

incompatibility between mmcv, mmdet, mmtrack.

3

This project wasted nearly 12 hours of my time solving the incompatibility issues between mmcv, mmdet, mmtrack, and numpy. I still haven't solved this problem.