Changed title

app.py

@@ -1,18 +1,21 @@
+                      
+                      
+                          # HyDE + ReRank RAG for Freights Rates
+
 # Import modules and classes
 from llama_index.core import VectorStoreIndex, StorageContext, load_index_from_storage
-from langchain_nvidia_ai_endpoints import NVIDIARerank
 from llama_index.core.indices.query.query_transform import HyDEQueryTransform
+from llama_index.core.query_engine import TransformQueryEngine
+from langchain_nvidia_ai_endpoints import NVIDIARerank
+from langchain_core.documents import Document as LangDocument
 from llama_index.core.llms import ChatMessage, MessageRole
 from llama_index.llms.nvidia import NVIDIA
 from llama_index.embeddings.nvidia import NVIDIAEmbedding
-from llama_index.core.query_engine import TransformQueryEngine
-from langchain_core.documents import Document as LangDocument
 from llama_index.core import Document as LlamaDocument
 from llama_index.core import Settings
 from llama_parse import LlamaParse
 import streamlit as st
 import os
-
 # Set environmental variables
 nvidia_api_key = os.getenv("NVIDIA_KEY")
 llamaparse_api_key = os.getenv("PARSE_KEY")
@@ -33,10 +36,12 @@ embed_model = NVIDIAEmbedding(
 )
 
 reranker = NVIDIARerank(
-  model="nvidia/nv-rerankqa-mistral-4b-v3", 
-  api_key=nvidia_api_key,
+    model="nvidia/nv_embedqa_e5-v5",
+    api_key=nvidia_api_key,
+    truncate="NONE"
 )
 
+
 # Set the NVIDIA models globally
 Settings.embed_model = embed_model
 Settings.llm = client
@@ -121,10 +126,7 @@ def query_model_with_context(question):
     retriever = index.as_retriever(similarity_top_k=3)
     nodes = retriever.retrieve(hyde_query)
 
-    for node in nodes:
-        print(node)
-
-    # Rerank the retrieved documents
+    # Rerank the retrieved nodes
     ranked_documents = reranker.compress_documents(
         query=question,
         documents=[LangDocument(page_content=node.text) for node in nodes]
@@ -146,8 +148,6 @@ def query_model_with_context(question):
     # Call the chat method to get the response
     completion = client.chat(messages)
 
-    print(completion)
-
     # Process response - assuming completion is a single string or a tuple containing a string
     response_text = ""
 
@@ -167,7 +167,7 @@ def query_model_with_context(question):
 
 
 # Streamlit UI
-st.title("Chat with HyDE + Rerank Freights RAG App")
+st.title("Chat with HyDE + Rerank Freights App")
 question = st.text_input("Enter a relevant question to chat with the attached FreightsDataset file:")
 
 if st.button("Submit"):
@@ -176,4 +176,5 @@ if st.button("Submit"):
         response = query_model_with_context(question)
         st.write(response)
     else:
-        st.warning("Please enter a question.")
+        st.warning("Please enter a question.")
+