Update app.py

app.py

@@ -19,41 +19,19 @@ st.set_page_config(page_title="SEARCH IATI",layout='wide')
 st.title("SEARCH IATI Database")
 var=st.text_input("enter keyword")
   
-
-def get_context(vectorstore,query):
-    # create metadata filter
-
-
-    # getting context
-    retriever = vectorstore.as_retriever(search_type="similarity_score_threshold", 
-                                         search_kwargs={"score_threshold": 0.5, 
-                                                        "k": 10,})
-#    # re-ranking the retrieved results
-#    model = HuggingFaceCrossEncoder(model_name=model_config.get('ranker','MODEL'))
-#   compressor = CrossEncoderReranker(model=model, top_n=int(model_config.get('ranker','TOP_K')))
-#    compression_retriever = ContextualCompressionRetriever(
-#           base_compressor=compressor, base_retriever=retriever
-#       )
-    context_retrieved = retriever.invoke(query)
-    print(f"retrieved paragraphs:{len(context_retrieved)}")
-
-    return context_retrieved
-
-# first we process and create the chunks for relvant data source
+####################  Create the embeddings collection and save ######################
+# the steps below need to be performed only once and then commented out any unnecssary compute over-run
+##### First we process and create the chunks for relvant data source
 #chunks = process_giz_worldwide()
-# we convert to langchain documents
+##### Convert to langchain documents
 #temp_doc = create_documents(chunks,'chunks')
-# embed and store docs
+##### Embed and store docs, check if collection exist then you need to update the collection
 #hybrid_embed_chunks(docs= temp_doc, collection_name = "giz_worldwide")
 
+################### Hybrid Search ######################################################
 
-print("embedding done")
 
-# once the chunks are done, we perform hybrid emebddings
-#embed_chunks(chunks)
 
-#vectorstores = get_local_qdrant('giz_worldwide') 
-#vectorstore = vectorstores['giz_worldwide']
 button=st.button("search")
 #found_docs = vectorstore.similarity_search(var)
 #print(found_docs)