fixing indention in app.py

app.py

@@ -2,14 +2,13 @@
 import streamlit as st
 import os
 import yaml
-from langchain_nvidia_ai_endpoints import ChatNVIDIA
 from dotenv import load_dotenv
 import torch
 from src.generator import answer_with_rag
 from ragatouille import RAGPretrainedModel
 from src.data_preparation import split_documents
 from src.embeddings import init_embedding_model
-from langchain_nvidia_ai_endpoints.embeddings import NVIDIAEmbeddings
+from langchain_nvidia_ai_endpoints import NVIDIAEmbeddings, ChatNVIDIA
 
 from transformers import pipeline
 from langchain_community.document_loaders import PyPDFLoader
@@ -70,58 +69,21 @@ def main():
         st.session_state.embedding_model=NVIDIAEmbeddings()
         st.session_state.KNOWLEDGE_VECTOR_DATABASE= init_vectorDB_from_doc(st.session_state.docs_processed, 
                                                                            st.session_state.embedding_model)
-    
-    #if os.path.exists(VECTORDB_PATH):
-      #  KNOWLEDGE_VECTOR_DATABASE = FAISS.load_local(
-      #      VECTORDB_PATH, embedding_model, 
-       #     allow_dangerous_deserialization=True)
-    #else:
-        #KNOWLEDGE_VECTOR_DATABASE=init_vectorDB_from_doc(docs_processed, embedding_model)
-       # KNOWLEDGE_VECTOR_DATABASE.save_local(VECTORDB_PATH)
-
-
-    if st.button("Get Answer"):
-    # Get the answer and relevant documents
-        #bnb_config = BitsAndBytesConfig(
-            #load_in_8bit=True
-         #   load_in_4bit=True,
-         #   bnb_4bit_use_double_quant=True,
-         #   bnb_4bit_quant_type="nf4",
-         #   bnb_4bit_compute_dtype=torch.bfloat16,
-        #)   
+    if (user_query) and (st.button("Get Answer")):
+        num_doc_before_rerank=15 
+        st.session_state.retriever= st.session_state.KNOWLEDGE_VECTOR_DATABASE.as_retriever(search_type="similarity", 
         
-
-    llm = ChatNVIDIA(
-        model=READER_MODEL_NAME,
-        api_key= os.get("NVIDIA_API_KEY"), 
-        temperature=0.2,
-        top_p=0.7,
-        max_tokens=1024,
-        )
-        #tokenizer = AutoTokenizer.from_pretrained(READER_MODEL_NAME)
-
-        #READER_LLM = pipeline(
-        #    model=model,
-        #    tokenizer=tokenizer,
-        #    task="text-generation",
-        #    do_sample=True,
-       #     temperature=0.2,
-        #    repetition_penalty=1.1,
-        #    return_full_text=False,
-        #    max_new_tokens=500,
-        #    token = os.getenv("HF_TOKEN")
-       # )
-       # RERANKER = RAGPretrainedModel.from_pretrained(RERANKER_MODEL_NAME)
-       # num_doc_before_rerank=15
-       # num_final_releveant_docs=5
-       # answer, relevant_docs = answer_with_rag(query=user_query, READER_MODEL_NAME=READER_MODEL_NAME,embedding_model=embedding_model,vectorDB=KNOWLEDGE_VECTOR_DATABASE,reranker=RERANKER, llm=READER_LLM,num_doc_before_rerank=num_doc_before_rerank,num_final_relevant_docs=num_final_releveant_docs,rerank=True)
-        #print(answer) 
-
-    
-        # Display the answer
+        st.write("### Please wait while we are getting the answer.....")                                                                                  search_kwargs={"k": num_doc_before_rerank})
+        llm = ChatNVIDIA(
+            model=READER_MODEL_NAME,
+            api_key= os.get("NVIDIA_API_KEY"), 
+            temperature=0.2,
+            top_p=0.7,
+            max_tokens=1024,
+            )
+        answer, relevant_docs = answer_with_rag(query=user_query, llm=llm, retriever=st.session_state.retriever)
         st.write("### Answer:")
         st.write(answer)
-    
         # Display the relevant documents
         st.write("### Relevant Documents:")
         for i, doc in enumerate(relevant_docs):