Update app.py

app.py

@@ -9,8 +9,6 @@ from langchain_text_splitters import RecursiveCharacterTextSplitter
 from langchain_google_genai import GoogleGenerativeAIEmbeddings
 from langchain_community.vectorstores import FAISS
 from langchain_google_genai import ChatGoogleGenerativeAI
-from langchain.chains.question_answering import load_qa_chain
-from langchain.prompts import PromptTemplate
 
 # --- Get API key from Hugging Face Secrets ---
 GOOGLE_API_KEY = os.environ.get("GOOGLE_API_KEY", "")
@@ -48,8 +46,22 @@ def get_vector_store(text_chunks, api_key):
     vector_store.save_local(FAISS_INDEX_PATH)
 
 
-def get_conversational_chain(api_key):
-    prompt_template = """
+def get_response(user_question, api_key, chat_history):
+    """Get response from the AI model with chat history context"""
+    embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001", google_api_key=api_key)
+    new_db = FAISS.load_local(FAISS_INDEX_PATH, embeddings, allow_dangerous_deserialization=True)
+    docs = new_db.similarity_search(user_question, k=4)
+    
+    # Format chat history for context
+    history_text = ""
+    for msg in chat_history[-10:]:  # Keep last 10 messages for context
+        role = "User" if msg["role"] == "user" else "Assistant"
+        history_text += f"{role}: {msg['content']}\n"
+    
+    # Combine document contents
+    context = "\n\n".join([doc.page_content for doc in docs])
+    
+    prompt_template = f"""
     You are a helpful assistant for Antimicrobial Pharmacology. You answer questions based ONLY on the context provided from the PDF documents.
     
     IMPORTANT RULES:
@@ -61,13 +73,13 @@ def get_conversational_chain(api_key):
     6. IMPORTANT: When referencing information from the course materials, always say "your professor says" or "according to your professor" instead of "the text states", "the document states", "the text says", or similar phrases. This makes the learning experience more personal and connected to the course.
     
     Chat History:
-    {chat_history}
+    {history_text}
     
     Context from PDF:
     {context}
     
     Current Question:
-    {question}
+    {user_question}
     
     Instructions:
     - If the user asks for a multiple choice question (MCQ), quiz, or test question:
@@ -87,34 +99,10 @@ def get_conversational_chain(api_key):
     
     Answer:
     """
-    model = ChatGoogleGenerativeAI(model="gemini-2.0-flash", temperature=0.3, google_api_key=api_key)
-    prompt = PromptTemplate(template=prompt_template, input_variables=["context", "question", "chat_history"])
-    chain = load_qa_chain(model, chain_type="stuff", prompt=prompt)
-    return chain
-
-
-def get_response(user_question, api_key, chat_history):
-    """Get response from the AI model with chat history context"""
-    embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001", google_api_key=api_key)
-    new_db = FAISS.load_local(FAISS_INDEX_PATH, embeddings, allow_dangerous_deserialization=True)
-    docs = new_db.similarity_search(user_question, k=4)
-    
-    # Format chat history for context
-    history_text = ""
-    for msg in chat_history[-10:]:  # Keep last 10 messages for context
-        role = "User" if msg["role"] == "user" else "Assistant"
-        history_text += f"{role}: {msg['content']}\n"
     
-    chain = get_conversational_chain(api_key)
-    response = chain(
-        {
-            "input_documents": docs, 
-            "question": user_question,
-            "chat_history": history_text
-        }, 
-        return_only_outputs=True
-    )
-    return response["output_text"]
+    model = ChatGoogleGenerativeAI(model="gemini-2.0-flash", temperature=0.3, google_api_key=api_key)
+    response = model.invoke(prompt_template)
+    return response.content
 
 
 def main():
@@ -291,4 +279,4 @@ def main():
 
 
 if __name__ == "__main__":
-    main() 
+    main()