search

zqhang / AnomalyCLIP

Official implementation for AnomalyCLIP (ICLR 2024)
MIT License
257 stars 25 forks source link

The process for the last class token of vision is different from original CLIP. #14

Open XpracticeYSKM opened 4 months ago

XpracticeYSKM commented 4 months ago

In line 387 of AnomalyCLIP_lib/AnomalyCLIP.py , the global class token from vision is directly feed into projector and don't process it with self.ln_post, which is different from original CLIP. It is a mistake or some special settting？It will contribute a lot for AnomalyCLIP？

zqhang commented 4 months ago

We have tested both settings, and this discrepancy does not impact the performance of AnomalyCLIP.