Fix: Update deprecated imports, add tiktoken, migrate to langchain_community

app.py

@@ -1,17 +1,17 @@
 import os
 import gradio as gr
 import torch
-from langchain.document_loaders import PyPDFLoader
+from langchain_community.document_loaders import PyPDFLoader
 from langchain.text_splitter import RecursiveCharacterTextSplitter
-from langchain.embeddings import OpenAIEmbeddings
-from langchain.vectorstores import FAISS
+from langchain_community.embeddings import OpenAIEmbeddings
+from langchain_community.vectorstores import FAISS
 from langchain.chains import RetrievalQA
-from langchain.chat_models import ChatOpenAI
+from langchain_openai import ChatOpenAI
 from transformers.pipelines import pipeline
 
-# ✅ Load API key from environment variable (set in Hugging Face Secrets)
+# ✅ Load API key from Hugging Face secret
 openai_key = os.environ.get("OPENAI_API_KEY")
-llm = ChatOpenAI(openai_api_key=openai_key, model_name="gpt-3.5-turbo", temperature=0)
+llm = ChatOpenAI(api_key=openai_key, model="gpt-3.5-turbo", temperature=0)
 
 # ✅ Build RAG agent
 def build_rag_agent(pdf_path):
@@ -19,7 +19,8 @@ def build_rag_agent(pdf_path):
     docs = loader.load()
     splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
     chunks = splitter.split_documents(docs)
-    vectorstore = FAISS.from_documents(chunks, OpenAIEmbeddings(openai_api_key=openai_key))
+    embeddings = OpenAIEmbeddings(api_key=openai_key)
+    vectorstore = FAISS.from_documents(chunks, embeddings)
     retriever = vectorstore.as_retriever()
     return RetrievalQA.from_chain_type(llm=llm, retriever=retriever, chain_type="stuff")
 
@@ -27,10 +28,10 @@ def build_rag_agent(pdf_path):
 stat6371_agent = build_rag_agent("PDFs/DS 6371 Syllabus Ver 6.pdf")
 ds7333_agent = build_rag_agent("PDFs/ds-7333_syllabus.pdf")
 
-# ✅ Load HF fine-tuned model for general stats
+# ✅ Load Hugging Face fine-tuned model
 general_stat_agent = pipeline("text2text-generation", model="BivinSadler/llama3-finetuned-Statistics")
 
-# ✅ Routing agent
+# ✅ Routing logic
 def route_question_llm(question):
     prompt = f"""
 You are a classification agent that helps route questions to the appropriate expert.
@@ -44,10 +45,10 @@ Classify the following question into one of those three categories by answering
 
 Question: "{question}"
 Answer:"""
-    route_response = llm.invoke(prompt).content.strip().upper()
-    if route_response.startswith("A"):
+    response = llm.invoke(prompt).content.strip().upper()
+    if response.startswith("A"):
         return "stat6371"
-    elif route_response.startswith("B"):
+    elif response.startswith("B"):
         return "ds7333"
     else:
         return "general"
@@ -64,7 +65,7 @@ Write your response in 2–3 sentences. Avoid technical jargon.
 """
     return llm.invoke(prompt).content
 
-# ✅ Multi-agent logic
+# ✅ Main app logic
 def multiagent_system(question):
     print(f"🧭 Routing: {question}")
     route = route_question_llm(question)