Update processing.py

processing.py

@@ -4,7 +4,7 @@ from langchain_community.embeddings import OpenAIEmbeddings
 from langchain_community.vectorstores import FAISS
 from llm_loader import load_model  # Import the function to load the model
 from config import openai_api_key  # Import the API key from config.py
-from langchain_community.chains import RetrievalQAChain  # Assuming this is the updated class
+from langchain.chains import RetrievalQA
 import os
 
 # Initialize embeddings and FAISS index
@@ -37,7 +37,8 @@ faiss_index.save_local("faiss_index")
 llm = load_model(openai_api_key)  # Load the model using your custom loader
 
 # Initialize the retrieval chain
-qa_chain = RetrievalQAChain(llm=llm, retriever=faiss_index.as_retriever())  # Replace with the correct chain class
+retriever = faiss_index.as_retriever()
+qa_chain = RetrievalQA.from_llm(llm=llm, retriever=retriever)
 
 def load_text(file_path: str) -> str:
     with open(file_path, 'r', encoding='utf-8') as file:
@@ -56,8 +57,8 @@ def process_task(llm, input_text: str, general_task: str, specific_task: str, ou
     truncated_input = truncate_text(input_text)
 
     # Perform retrieval to get the most relevant context
-    relevant_docs = qa_chain({"query": truncated_input})
-    retrieved_knowledge = "\n".join([doc.page_content for doc in relevant_docs['documents']])
+    query = f"{general_task}\n\n{specific_task}\n\n{truncated_input}"
+    retrieved_knowledge = qa_chain.run(query)
 
     # Combine the retrieved knowledge with the original prompt
     prompt = f"""{general_task}