Update app.py

app.py

@@ -1,42 +1,44 @@
 import os
 import chainlit as cl
+from langchain_community.vectorstores import FAISS
+from langchain_community.embeddings import HuggingFaceEmbeddings
+from langchain_groq import ChatGroq
 from langchain_community.document_loaders import PyPDFLoader
-from langchain.text_splitter import CharacterTextSplitter
-from langchain.vectorstores.faiss import FAISS
-from langchain.embeddings import HuggingFaceEmbeddings
-from langchain.chains import RetrievalQA
-from langchain_community.llms import Groq
+from langchain_text_splitters import RecursiveCharacterTextSplitter
+from langchain.chains.question_answering import load_qa_chain
+from langchain_community.llms import HuggingFaceHub
+from dotenv import load_dotenv
 
-# Load your API key
-os.environ["GROQ_API_KEY"] = os.getenv("GROQ_API_KEY") or "your-groq-api-key"  # Replace with actual key if needed
+load_dotenv()
 
-# Initialize the LLM
-llm = Groq(model="llama3-8b-8192")
-
-# Load and split PDF (do this only once — in real apps, optimize this with caching)
-def load_pdf_and_create_retriever():
-    loader = PyPDFLoader("document.pdf")  # Make sure document.pdf is in the same folder
-    documents = loader.load()
+@cl.on_message
+async def main(message: cl.Message):
+    # Load PDF
+    loader = PyPDFLoader("document.pdf")
+    pages = loader.load_and_split()
 
-    text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
-    docs = text_splitter.split_documents(documents)
+    # Split text
+    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
+    documents = text_splitter.split_documents(pages)
 
+    # Embeddings & Vector DB
     embeddings = HuggingFaceEmbeddings()
-    db = FAISS.from_documents(docs, embeddings)
-    retriever = db.as_retriever()
-    return retriever
+    vectorstore = FAISS.from_documents(documents, embeddings)
+
+    # Get relevant docs
+    query = message.content
+    matching_docs = vectorstore.similarity_search(query)
 
-# Create the retriever once
-retriever = load_pdf_and_create_retriever()
+    # LLM setup
+    llm = ChatGroq(
+        api_key=os.getenv("GROQ_API_KEY"),
+        model_name="LLaMA3-8b-8192"
+    )
 
-# Create Retrieval-QA chain
-qa_chain = RetrievalQA.from_chain_type(llm=llm, retriever=retriever)
+    # QA chain
+    chain = load_qa_chain(llm, chain_type="stuff")
+    answer = chain.run(input_documents=matching_docs, question=query)
 
-# Chainlit callback
-@cl.on_message
-async def handle_message(message: cl.Message):
-    query = message.content
-    result = qa_chain.run(query)
-    await cl.Message(content=result).send()
+    await cl.Message(content=answer).send()