Update app.py

app.py

@@ -544,8 +544,47 @@ memory.save_context(
 
 
 
+questions = ["what does DTC stands for?",
+             "plaese give more details about it, including its use cases and implementation.",
+             "does it outperform other diffusion-based models? explain in details.",
+             "what is Langchain?"]
+
+# Instantiate the retriever and the ConversationalRetrievalChain :
+
+retriever_Google = retrieval_blocks(
+    create_vectorstore=False,
+    LLM_service="Google",
+    vectorstore_name="Vit_All_Google_Embeddings",
+    retriever_type="Cohere_reranker",
+    base_retriever_search_type="similarity", base_retriever_k=12,
+    compression_retriever_k=16,
+    cohere_api_key=cohere_api_key,cohere_top_n=10,
+)
+
+chain_gemini,memory_gemini = custom_ConversationalRetrievalChain(
+    llm = instantiate_LLM(
+        LLM_provider="Google",api_key=google_api_key,temperature=0.5,model_name="gemini-pro"
+    ),
+    condense_question_llm = instantiate_LLM(
+        LLM_provider="Google",api_key=google_api_key,temperature=0.1,model_name="gemini-pro"),
+    retriever=retriever_Google,
+    language="english",
+    llm_provider="Google",
+    model_name="gemini-pro"
+)
 
 
+memory_gemini.clear()
+
+for i,question in enumerate(questions):
+    response = chain_gemini.invoke({"question":question})
+    answer = response['answer'].content
+    print(f"Question[{i}]:",question)
+    print("Standalone_question:",response['standalone_question'])
+    print("Answer:\n",answer,f"\n\n{'-' * 100}\n")
+    
+    memory_gemini.save_context( {"question": question}, {"answer": answer} ) # update memory
+