Update app.py

app.py

@@ -1,4 +1,4 @@
-from transformers import AutoTokenizer, CLIPImageProcessorProcessor, AutoProcessor, pipeline, CLIPTokenizer
+from transformers import CLIPImageProcessor, pipeline, CLIPTokenizer
 import torchvision.transforms as T
 import torch.nn.functional as F
 from PIL import Image, ImageFile
@@ -60,6 +60,9 @@ def infer(image, labels):
 
 with gr.Blocks() as demo:
   gr.Markdown("# EVACLIP vs CLIP 💥 ")
+  gr.Markdown("[EVACLIP](https://huggingface.co/BAAI/EVA-CLIP-8B) is CLIP scaled to the moon! 🔥")
+  gr.Markdown("It's a state-of-the-art zero-shot image classification model, which is also outperforming predecessors on text-image retrieval and linear probing.")
+  gr.Markdown("In this demo, compare EVACLIP outputs to CLIP outputs ✨")
   with gr.Row():
     with gr.Column():
         image_input = gr.Image(type="pil")