Update app.py

app.py

@@ -1,22 +1,25 @@
 import streamlit as st
-from transformers import CLIPProcessor, CLIPModel, DiffusionModel
+from transformers import CLIPProcessor, CLIPModel
 import torch
 from PIL import Image
 
-st.title("Text to Image Generation")
+st.title("Text to Image Generation with CLIP")
 
 # Load pretrained models
 clip_processor = CLIPProcessor.from_pretrained("openai/clip-vit-base-patch16")
 clip_model = CLIPModel.from_pretrained("openai/clip-vit-base-patch16")
-diffusion_model = DiffusionModel.from_pretrained("openai/guided-diffusion-clipped-coco")
 
 text = st.text_area("Enter a description:")
 if st.button("Generate Image") and text:
     # Process text and get CLIP features
     text_features = clip_processor(text, return_tensors="pt", padding=True)
 
-    # Generate image from text using Guided Diffusion
-    image = diffusion_model.generate_text_to_image(text_features["pixel_values"])
+    # Use CLIP's image projection to generate an image representation
+    image_representation = clip_model.get_image_features(pixel_values=text_features.pixel_values)
+    
+    # For visualization, you can convert the image representation back to an image
+    image_array = image_representation.squeeze().permute(1, 2, 0).cpu().numpy()
+    image = Image.fromarray((image_array * 255).astype('uint8'))
 
     # Display the generated image
     st.image(image, caption="Generated Image", use_column_width=True)