Update app.py

app.py

@@ -86,12 +86,12 @@ 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
-model = HuggingFaceCrossEncoder(model_name="BAAI/bge-reranker-base")
-compressor = CrossEncoderReranker(model=model, top_n=10)
+# model = HuggingFaceCrossEncoder(model_name="BAAI/bge-reranker-base")
+# compressor = CrossEncoderReranker(model=model, top_n=10)
 
-compression_retriever = ContextualCompressionRetriever(
-    base_compressor=compressor, base_retriever=retriever
-)
+# compression_retriever = ContextualCompressionRetriever(
+#     base_compressor=compressor, base_retriever=retriever
+# )
 # from langchain.retrievers.document_compressors import LLMChainExtractor
 
 # compressor = LLMChainExtractor.from_llm(llm)
@@ -141,7 +141,7 @@ When responding to queries, follow these guidelines:
 3. Proper References:
    - Always use inline references with embedded source URLs. 
    - The inline references should be in the format [1], [2], etc.
-   - INCLUDE THE 'References' SECTION AT THE END OF RESPONSE.
+   - INCLUDE THE 'References' SECTION WITH REFERENCES URLs AT THE END OF RESPONSE.
    
 FOLLOW ALL THE GIVEN INSTRUCTIONS, FAILURE TO DO SO WILL RESULT IN THE TERMINATION OF THE CHAT.