Spaces:

---

imageomics

/

bioclip-demo

like 70

Running

model.logit_scale has already exp() before called, so no need to exp() twice. If double exp(), some values might become too large and overflow. Then, the ranking becomes meaningless and wrong. And it will affect probabilities/confidence, making the result seems over-confident.

This change is consistent with BioCLIP offical Github repo