Update app.py

app.py

@@ -3,12 +3,23 @@ import gradio as gr
 from langchain.vectorstores import FAISS
 from langchain.embeddings import HuggingFaceEmbeddings
 from langchain.schema import Document
+import os
+import pickle
 
-# Initialize once when the app starts
+EMBEDDING_MODEL = "sentence-transformers/multi-qa-mpnet-base-dot-v1"
+DATASET_PATH = "qa_dataset.csv"
+FAISS_INDEX_PATH = "faiss_index"
+
+# Initialize system (Load dataset and FAISS index)
 def initialize_system():
-    # Load dataset
-    data = pd.read_csv("qa_dataset.csv")
-    
+    if os.path.exists(FAISS_INDEX_PATH):
+        print("Loading FAISS index from cache...")
+        with open(FAISS_INDEX_PATH, "rb") as f:
+            return pickle.load(f)
+
+    print("Initializing FAISS from scratch...")
+    data = pd.read_csv(DATASET_PATH).dropna().head(500)  # Limit rows for speed
+
     # Create documents
     documents = [
         Document(
@@ -16,30 +27,35 @@ def initialize_system():
             metadata={"question": row['Question'], "answer": row['Answer']}
         ) for _, row in data.iterrows()
     ]
-    
-    # Create vector store
-    embeddings = HuggingFaceEmbeddings(
-        model_name="sentence-transformers/multi-qa-mpnet-base-dot-v1"
-    )
-    
-    return FAISS.from_documents(documents, embeddings)
+
+    # Load embedding model once
+    embeddings = HuggingFaceEmbeddings(model_name=EMBEDDING_MODEL)
+
+    # Create FAISS vector store
+    vector_store = FAISS.from_documents(documents, embeddings)
+
+    # Cache the FAISS index
+    with open(FAISS_INDEX_PATH, "wb") as f:
+        pickle.dump(vector_store, f)
+
+    return vector_store
 
 vector_store = initialize_system()
 
 def classify_question(query: str, k: int = 3):
-    # Retrieve similar Q&A pairs
     results = vector_store.similarity_search(query, k=k)
-    
-    # Generate category from answers
+
+    if not results:
+        return {"Category": "Unknown", "Top Matches": "No matches found", "Confidence": "0%"}
+
     answers = " ".join([doc.metadata['answer'] for doc in results])
-    keywords = list(dict.fromkeys(answers.split()))[:5]
+    keywords = list(dict.fromkeys(answers.split()))[:5]  # Extract first 5 unique words
     category = " ".join(keywords)
-    
-    # Format output
+
     return {
         "Category": category,
         "Top Matches": "\n\n".join([f"Q: {doc.metadata['question']}\nA: {doc.metadata['answer']}" 
-                                  for doc in results]),
+                                    for doc in results]),
         "Confidence": f"{len(results)/k:.0%}"
     }
 
@@ -53,8 +69,8 @@ interface = gr.Interface(
         gr.Textbox(label="Confidence")
     ],
     title="Question Classification System",
-    description="Classify questions based on existing Q&A pairs using RAG"
+    description="Classify questions based on existing Q&A pairs using FAISS"
 )
 
 if __name__ == "__main__":
-    interface.launch()
+    interface.launch(share=True)