Update app.py

app.py

@@ -1,19 +1,19 @@
+import os
 import gradio as gr
 from langchain_google_genai.chat_models import ChatGoogleGenerativeAI
 from PIL import Image
 import torch
 from torchvision import models, transforms
-import json
-import requests
 
-# Initialize the chat model with Hugging Face-specific environment variables
-llm = ChatGoogleGenerativeAI(model='gemini-1.5-pro')
+# Set up the environment for Google Generative AI
+os.environ["GOOGLE_APPLICATION_CREDENTIALS"] = "./firm-catalyst-437006-s4-407500537db5.json"
+llm = ChatGoogleGenerativeAI(model='gemini-1.5-pro')  
 
-# Load a pre-trained ResNet50 model for image classification
+# Load a pre-trained ResNet50 model for image analysis
 model = models.resnet50(pretrained=True)
-model.eval()
+model.eval()  # Set the model to evaluation mode
 
-# Transformation pipeline for image preprocessing
+# Define the transformation for the image
 transform = transforms.Compose([
     transforms.Resize(256),
     transforms.CenterCrop(224),
@@ -21,62 +21,68 @@ transform = transforms.Compose([
     transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
 ])
 
-# Load ImageNet labels
+# Load the ImageNet labels
 LABELS_URL = "https://raw.githubusercontent.com/anishathalye/imagenet-simple-labels/master/imagenet-simple-labels.json"
-labels = json.loads(requests.get(LABELS_URL).text)
+labels = None
+
+if not os.path.exists("imagenet_labels.json"):
+    import requests
+    response = requests.get(LABELS_URL)
+    with open("imagenet_labels.json", "wb") as f:
+        f.write(response.content)
 
-# Global chat history variable
-chat_history = []
+import json
+with open("imagenet_labels.json") as f:
+    labels = json.load(f)
 
-def chat_with_gemini(message):
-    global chat_history
-    # Get a response from the language model
+def chat_with_gemini(message, chat_history):
+    # Generate a response from the language model
     bot_response = llm.predict(message)
     chat_history.append((message, bot_response))
-    return chat_history
+    
+    return chat_history, chat_history
 
-def analyze_image(image_path):
-    global chat_history
-    # Open, preprocess, and classify the image
+def analyze_image(image_path, chat_history):
+    # Load and preprocess the image
     image = Image.open(image_path).convert("RGB")
     image_tensor = transform(image).unsqueeze(0)
-
+    
+    # Predict the image class
     with torch.no_grad():
         outputs = model(image_tensor)
         _, predicted_idx = outputs.max(1)
     
+    # Retrieve the label
     label = labels[predicted_idx.item()]
+    
+    # Respond with the classification result
     bot_response = f"The image seems to be: {label}."
     chat_history.append(("Uploaded an image for analysis", bot_response))
-    return chat_history
+    
+    return chat_history, chat_history
 
-# Build the Gradio interface
-with gr.Blocks() as demo:
+# Create Gradio interface
+with gr.Blocks() as iface:
     gr.Markdown("# Ken Chatbot")
     gr.Markdown("Ask me anything or upload an image for analysis!")
 
-    # Chatbot display without "User" or "Bot" labels
+    # Chatbot component without usernames
     chatbot = gr.Chatbot(elem_id="chatbot")
-
+    
     # User input components
-    msg = gr.Textbox(label="Type your message here...", placeholder="Enter your message...", show_label=False)
+    msg = gr.Textbox(label="Type your message here...", placeholder="Enter your message...")
     send_btn = gr.Button("Send")
     img_upload = gr.Image(type="filepath", label="Upload an image for analysis")
 
-    # Define interactions
-    def handle_text_message(message):
-        return chat_with_gemini(message)
-
-    def handle_image_upload(image_path):
-        return analyze_image(image_path)
+    # State for chat history
+    state = gr.State([])
 
-    # Set up Gradio components with Enter key for sending
-    msg.submit(handle_text_message, msg, chatbot)
-    send_btn.click(handle_text_message, msg, chatbot)
-    send_btn.click(lambda: "", None, msg)  # Clear input field
-    img_upload.change(handle_image_upload, img_upload, chatbot)
+    # Define interactions
+    send_btn.click(chat_with_gemini, [msg, state], [chatbot, state])  # Handle text input
+    send_btn.click(lambda: "", None, msg)  # Clear textbox
+    img_upload.change(analyze_image, [img_upload, state], [chatbot, state])  # Handle image uploads
 
-    # Custom CSS for styling without usernames
+    # Custom CSS for styling chat bubbles without usernames
     gr.HTML("""
     <style>
     #chatbot .message-container {
@@ -104,5 +110,5 @@ with gr.Blocks() as demo:
     </style>
     """)
 
-# Launch for Hugging Face Spaces
-demo.launch()
+# Launch the Gradio interface
+iface.launch(debug=True)