Chuang Lin

Results 9 comments of Chuang Lin

Can you try "pip install -e . --user" instead?

Hi, Thanks for your interests. I add it here https://drive.google.com/file/d/1qqILBDC--N2bQsUZ06D-t4w3a03jVDFm/view?usp=share_link. By the way, "coco_65_concepts.txt" is not necessary for running the code

Hi, Thanks for your interest. The code now is not supporting it. We have a plan to release such a script later. Or. you can just generate the custom vocabulary...

Hi, Can you check timm version? Or try pip install timm==v0.5.4? --Chuang

Thanks for your interests! VLDet currently does not support image and text retrieval. You can try to solve retrieval problems as the CLIP way.

Hi, Thanks for your interest. The main results are trained on 8 GPUs. 3 GPUs and small batchsize may drop the performance.

Thank you for your interest. There is no need to do anything; just wait for a while. The fix AP will be computed by running [this line of code](https://github.com/FoundationVision/GenerateU/blob/main/projects/DDETRS/train_net.py#L229).

Thank you for your interest. There is no need to do anything; just wait for a while. The fix AP will be computed by running [this line of code](https://github.com/FoundationVision/GenerateU/blob/main/projects/DDETRS/train_net.py#L229).