Create app.py

app.py

@@ -0,0 +1,72 @@
+import gradio as gr
+from transformers import AutoProcessor, AutoModelForVision2Seq
+import torch
+from PIL import Image
+
+# Load NanoLLaVA model and processor
+model_name = "facebook/nano-llava"
+processor = AutoProcessor.from_pretrained(model_name)
+model = AutoModelForVision2Seq.from_pretrained(model_name)
+
+def generate_caption(image):
+    # Process the image
+    inputs = processor(images=image, text="Describe this image in detail", return_tensors="pt")
+    
+    # Generate caption
+    outputs = model.generate(
+        **inputs,
+        max_length=100,
+        num_beams=4,
+        temperature=0.8
+    )
+    
+    # Decode the caption
+    caption = processor.decode(outputs[0], skip_special_tokens=True)
+    return caption
+
+def create_persona(caption):
+    # Template for transforming caption into a persona
+    persona_prompt = f"""You are a character based on this description: {caption}
+
+Role: An entity exactly as described in the image
+Background: Your appearance and characteristics match the image description
+Personality: Reflect the mood, style, and elements captured in the image
+Goal: Interact authentically based on your visual characteristics
+
+Please stay in character and respond as this entity would, incorporating visual elements from your description into your responses."""
+    
+    return persona_prompt
+
+def process_image_to_persona(image):
+    # Generate caption from image
+    caption = generate_caption(image)
+    
+    # Transform caption into persona
+    persona = create_persona(caption)
+    
+    return caption, persona
+
+# Create Gradio interface
+with gr.Blocks() as app:
+    gr.Markdown("# Image to Chatbot Persona Generator")
+    gr.Markdown("Upload an image of a character to generate a persona for a chatbot based on the image.")
+    
+    with gr.Row():
+        image_input = gr.Image(type="pil", label="Upload Character Image")
+    
+    with gr.Row():
+        generate_button = gr.Button("Generate Persona")
+    
+    with gr.Row():
+        caption_output = gr.Textbox(label="Generated Caption", lines=3)
+        persona_output = gr.Textbox(label="Chatbot Persona", lines=10)
+    
+    generate_button.click(
+        fn=process_image_to_persona,
+        inputs=[image_input],
+        outputs=[caption_output, persona_output]
+    )
+
+# Launch the app
+if __name__ == "__main__":
+    app.launch(share=True)