Update app.py

app.py

@@ -86,9 +86,14 @@ retriever = PineconeHybridSearchRetriever(
 llm = ChatPerplexity(temperature=0, pplx_api_key=GROQ_API_KEY, model="llama-3.1-sonar-large-128k-chat", max_tokens=512, max_retries=2)
 
 # Initialize Reranker
-from langchain.retrievers.document_compressors import FlashrankRerank
+# from langchain.retrievers.document_compressors import FlashrankRerank
+
+# compressor = FlashrankRerank(model="ms-marco-MultiBERT-L-12", score_threshold=0.6)
+
+from langchain_community.document_compressors.rankllm_rerank import RankLLMRerank
+
+compressor = RankLLMRerank(top_n=6, model="zephyr")
 
-compressor = FlashrankRerank(model="ms-marco-MultiBERT-L-12", score_threshold=0.6)
 compression_retriever = ContextualCompressionRetriever(
     base_compressor=compressor, base_retriever=retriever
 )
@@ -98,7 +103,7 @@ compression_retriever = ContextualCompressionRetriever(
 )
 
 # Contextualization prompt and retriever
-contextualize_q_system_prompt = """Given a chat history and the latest user question \
+contextualize_q_system_prompt = """ Given a chat history and the latest user question \
 which might reference context in the chat history, formulate a standalone question \
 which can be understood without the chat history. Do NOT answer the question, \
 just reformulate it if needed and otherwise return it as is.