Create app.py

app.py

@@ -0,0 +1,85 @@
+import gradio as gr
+import chromadb
+from sentence_transformers import SentenceTransformer
+import pandas as pd
+
+# --- 1. Load Model (No changes here) ---
+print("Loading sentence-transformer model...")
+model = SentenceTransformer('all-MiniLM-L6-v2')
+print("Model loaded.")
+
+# --- 2. Setup ChromaDB (No changes here) ---
+client = chromadb.Client()
+
+try:
+    collection = client.create_collection("my_documents")
+    print("ChromaDB collection created.")
+    
+    documents = [
+        "The Eiffel Tower is a wrought-iron lattice tower on the Champ de Mars in Paris, France.",
+        "The Great Wall of China is a series of fortifications made of stone, brick, and other materials.",
+        "ChromaDB is an open-source embedding database designed to store and retrieve vector embeddings.",
+        "Hugging Face Spaces is a platform for building, deploying, and sharing ML apps.",
+        "A chatbot is a software application used to conduct an on-line chat conversation."
+    ]
+    
+    embeddings = model.encode(documents)
+    
+    collection.add(
+        embeddings=embeddings.tolist(),
+        documents=documents,
+        ids=[f"id_{i}" for i in range(len(documents))]
+    )
+    print("Documents added to ChromaDB.")
+
+except ValueError:
+    collection = client.get_collection("my_documents")
+    print("ChromaDB collection loaded.")
+
+
+# --- 3. Define the Chatbot Logic (Minor change to fit ChatInterface) ---
+def chatbot_response(message, history):
+    """
+    This function processes the user message and returns a response.
+    'message' is the user's input, 'history' is the conversation history.
+    """
+    # 1. Create an embedding for the user's message
+    query_embedding = model.encode([message]).tolist()
+    
+    # 2. Query ChromaDB to find the most relevant documents
+    results = collection.query(
+        query_embeddings=query_embedding,
+        n_results=2 
+    )
+    
+    retrieved_documents = results['documents'][0]
+    
+    if not retrieved_documents:
+        return "I'm sorry, I couldn't find any information on that topic. 🤔"
+        
+    # 4. Formulate the response
+    context = "\n- ".join(retrieved_documents)
+    response = f"Here's what I found related to your question:\n- {context}"
+    
+    return response
+
+# --- 4. Create the NEW Gradio Chat Interface ---
+iface = gr.ChatInterface(
+    fn=chatbot_response,
+    title="ChromaDB Knowledge Bot 🤖",
+    description="Ask me anything! I'll search my knowledge base to find the best answer for you.",
+    theme="soft", # Try "soft", "glass", "monochrome", or "base"
+    examples=[
+        "What is ChromaDB?", 
+        "Tell me about Hugging Face Spaces", 
+        "What is the Eiffel Tower made of?"
+    ],
+    chatbot=gr.Chatbot(avatar_images=("user.png", "bot.png")), # Optional: Add avatar images
+    cache_examples=False,
+    retry_btn=None,
+    undo_btn=None,
+    clear_btn="Clear Conversation",
+)
+
+# Launch the app
+iface.launch()