update

README.md

@@ -1,4 +1,12 @@
-# [ViTamin: Design Scalable Vision Models in the Vision-language Era](https://arxiv.org/pdf/2404.02132.pdf)
+---
+license: mit
+datasets:
+- mlfoundations/datacomp_1b
+pipeline_tag: feature-extraction
+---
+
+# Model card for ViTamin-XL-336px
+
 Official huggingface models of **ViTamin**, from the following paper:
 
 [ViTamin: Design Scalable Vision Models in the Vision-language Era](https://arxiv.org/pdf/2404.02132.pdf).\
@@ -6,7 +14,7 @@ Official huggingface models of **ViTamin**, from the following paper:
 🏠  Johns Hopkins University, Bytedance
 
 
-Load from HuggingFace:
+Load from HuggingFace with transformers.AutoModel:
 ```python
 import torch
 import open_clip
@@ -31,8 +39,7 @@ with torch.no_grad(), torch.cuda.amp.autocast():
     image_features, text_features, logit_scale = model(pixel_values, text)
     text_probs = (100.0 * image_features @ text_features.to(torch.float).T).softmax(dim=-1)
 
-print("Label probs:", text_probs) 
-
+print("Label probs:", text_probs)
 ```
 
 ## Main Results with CLIP Pre-training on DataComp-1B