-
Add model-specific JSON label files
duplicates of the existing JSON, required for pybioclip updated functionality
-
Update text embeddings from TOL-10M catalog
- regenerated `txt_emb_species.json` from #13 catalog.csv (resulted in
350,098 species)
- Added `txt_emb_bioclip.npy`: BioCLIP v1 txt embeddings
- Added `txt_emb_biocap.npy`: BioCap txt embeddings
- Updated `txt_emb_species.npy` with a copy of `txt_emb_bioclip.npy`,
supporting backward compat
-
Add BioCLIP embeddings (#9)
- Add BioCLIP embeddings (91501708dc8c2341d0818c1809570520698695c1)
- clarify text embeddings (1e1eb8ed8203ad9bb3ef17d006587f0924cd1110)
