Update app.py

app.py

@@ -2,7 +2,6 @@ import gradio as gr
 from PIL import Image
 import io
 import base64
-import requests  # For making API requests
 
 # Function to encode image as base64
 def image_to_base64(image):
@@ -12,63 +11,51 @@ def image_to_base64(image):
     return img_str
 
 # Function to interact with LLAVA model
-def chat_with_llava(image, question):
-    try:
+def respond(
+    message,
+    history: list[tuple[str, str]],
+    system_message,
+    max_tokens,
+    temperature,
+    top_p,
+    image=None
+):
+    messages = [{"role": "system", "content": system_message}]
+
+    for val in history:
+        if val[0]:
+            messages.append({"role": "user", "content": val[0]})
+        if val[1]:
+            messages.append({"role": "assistant", "content": val[1]})
+
+    if image:
         # Convert image to base64
         image_b64 = image_to_base64(image)
+        messages.append({"role": "user", "content": "Image uploaded", "image": image_b64})
 
-        # Crafting a prompt to instruct the model to respond as a doctor
-        doctor_prompt = (
-            "You are a highly experienced and knowledgeable medical doctor. "
-            "Please analyze the provided medical image and give a detailed medical explanation in response to the following question.\n\n"
-            f"Question: {question}\n"
-            "Please include relevant medical terminology and explanations in your response."
-        )
+    messages.append({"role": "user", "content": message})
 
-        # URL for Ollama chat API endpoint
-        api_url = "https://api.ollama.com/chat"
-        
-        # Prepare headers with API key (replace 'YOUR_API_KEY' with your actual API key)
-        headers = {
-            "Authorization": "Bearer YOUR_API_KEY",
-            "Content-Type": "application/json"
-        }
+    response = ""
 
-        # Prepare payload for the API request
-        payload = {
-            "model": "rohithbojja/llava-med-v1.6",
-            "messages": [
-                {
-                    "role": "user",
-                    "content": doctor_prompt,
-                    "image": image_b64
-                }
-            ]
-        }
+    # Simulate a response (replace with your logic)
+    for i in range(3):  # Simulate 3 responses
+        response += f"Response {i+1}\n"
 
-        # Make POST request to Ollama chat API
-        response = requests.post(api_url, json=payload, headers=headers)
-        response_data = response.json()
-
-        # Extract and return model response
-        return response_data['message']['content']
-
-    except Exception as e:
-        return f"Error occurred: {str(e)}"
+    yield response
 
 # Create a Gradio interface
-iface = gr.Interface(
-    fn=chat_with_llava,
+demo = gr.Interface(
+    fn=respond,
     inputs=[
-        gr.inputs.Image(type="pil", label="Upload Medical Image"),
-        gr.inputs.Textbox(lines=2, label="Ask a medical question about the image")
+        gr.inputs.Textbox(value="You are a friendly Chatbot.", label="System message"),
+        gr.inputs.Image(type="pil", label="Upload Image"),
+        gr.inputs.Textbox(lines=2, label="Message")
     ],
     outputs=gr.outputs.Textbox(label="Response", placeholder="Model response will appear here..."),
-    title="LLAVA Model - Medical Image and Question",
-    description="Upload a medical image and ask a specific question about the image for a medical description."
+    title="Chat Interface Demo",
+    description="Simulate a chat interface with customizable parameters."
 )
 
 # Launch the Gradio interface
 if __name__ == "__main__":
-    iface.launch()
-
+    demo.launch()