Update QnA.py

QnA.py

@@ -64,7 +64,7 @@ def get_hugging_face_model(model_id='mistralai/Mistral-7B-Instruct-v0.2',tempera
         )
     return llm
 
-def Q_A(vectorstore,question,API_KEY):
+def Q_A(vectorstore,question,API_KEY,compressor=False):
     
     if API_KEY.startswith('gsk'):
         os.environ["GROQ_API_KEY"] = API_KEY
@@ -74,9 +74,13 @@ def Q_A(vectorstore,question,API_KEY):
     
     # Create a retriever
     retriever = vectorstore.as_retriever(search_type = 'similarity',search_kwargs = {'k':2},)
-    #Create a contextual compressor
-    compressor = LLMChainExtractor.from_llm(chat_llm)
-    compression_retriever = ContextualCompressionRetriever(base_compressor=compressor,base_retriever=retriever)
+    
+    if compressor:
+        #Create a contextual compressor
+        compressor = LLMChainExtractor.from_llm(chat_llm)
+        compression_retriever = ContextualCompressionRetriever(base_compressor=compressor,base_retriever=retriever)
+        retriever = compression_retriever
+        
     
     if  'reliable' in question.lower() or 'relaibility' in question.lower():
         question_answer_chain = create_stuff_documents_chain(chat_llm, prompt_template_for_relaibility())
@@ -84,7 +88,7 @@ def Q_A(vectorstore,question,API_KEY):
     else:
         question_answer_chain = create_stuff_documents_chain(chat_llm, prompt_template_to_analyze_resume())
     
-    chain = create_retrieval_chain(compression_retriever, question_answer_chain)
+    chain = create_retrieval_chain(retriever, question_answer_chain)
     result = chain.invoke({'input':question})
     return result['answer']