Update app.py

app.py

@@ -1,9 +1,9 @@
 import pandas as pd
 import fitz  # PyMuPDF for PDF extraction
 import spacy
-from langchain_community.chains import ConversationalRetrievalChain
-from langchain_community.llms import OpenAI  # Updated import
-from langchain_community.vectorstores import FAISS  # Updated import
+from langchain.chains import ConversationalRetrievalChain  # Ensure this class is available or use an alternative
+from langchain.llms import OpenAI
+from langchain.vectorstores import FAISS
 import torch
 from transformers import AutoTokenizer, AutoModel
 import gradio as gr
@@ -38,7 +38,7 @@ class CustomEmbeddingModel:
 embedding_model = CustomEmbeddingModel('distilbert-base-uncased')  # Replace with your model name
 
 # Load Spacy model for preprocessing
-nlp = spacy.load("en_core_web_sm")  # Ensure the model is installed
+nlp = spacy.load("en_core_web_sm")
 
 def preprocess_text(text):
     doc = nlp(text)
@@ -57,6 +57,8 @@ vector_store = FAISS.from_documents(documents, embeddings)
 # Create LangChain model and chain
 llm_model = OpenAI('gpt-3.5-turbo')  # You can replace this with a different LLM if desired
 retriever = vector_store.as_retriever()
+
+# Create a conversational chain
 chain = ConversationalRetrievalChain.from_llm(llm_model, retriever=retriever)
 
 # Function to generate a response
@@ -77,3 +79,4 @@ iface = gr.Interface(
 if __name__ == "__main__":
     iface.launch()
 
+