metadata
license: mit
datasets:
- ILSVRC/imagenet-1k
- mlfoundations/datacomp_small
base_model:
- laion/CLIP-ViT-g-14-laion2B-s12B-b42K
Model Initialized from laion/CLIP-ViT-g-14-laion2B-s12B-b42K. The image encoder is finetuned with FARE at $\epsilon=2/255$. The text encoder is finetuned with LEAF at $k=1$ with $\rho=50$.
To load this model use:
from transformers import CLIPProcessor, CLIPModel
model_name = "LEAF-CLIP/OpenCLIP-ViT-g-rho50-k1-FARE2"
processor_name = "laion/CLIP-ViT-g-14-laion2B-s12B-b42K"
model = CLIPModel.from_pretrained(model_name)
processor = CLIPProcessor.from_pretrained(processor_name)