metadata
license: apache-2.0
datasets:
- shiwk24/MathCanvas-Imagen
base_model:
- google/siglip2-base-patch16-224
library_name: transformers
SigLip2 Math
This version of siglip2 is fine tuned on shiwk24/MathCanvas-Imagen using the code_derived_captions split.
I trained for 1 epoch on 4M math images, with a random selection between the tikz code or caption using a batch size of 640.
This is not a classification model, since the loss function was pairwise contrastive loss. Use for embedding or downstream classifier training is recommended.
