search

muzairkhattak / ProText

[CVPRW 2024] Official repository of paper titled "Learning to Prompt with Text Only Supervision for Vision-Language Models".
https://muzairkhattak.github.io/ProText/
MIT License
86 stars 4 forks source link

Timm model #4

Closed TahaKoleilat closed 4 months ago

TahaKoleilat commented 5 months ago

How to perform this method on a timm clip model? What should be changed so that the code works?

muzairkhattak commented 5 months ago

Hi @TahaKoleilat,

Can you clarify if you need to use timm clip model just for the inference? or you also required to perform training?

For the inference, you will need to replace the clip model instantiations with the timm models in the protext.py.

Let me know if you have any confusion! Thank you!

muzairkhattak commented 4 months ago

Hi @TahaKoleilat,

I am closing this issue for now. Feel free to reopen if you still have any question.

Thank you and kind regards!