switch owid vectorstore to pinecone

app.py

@@ -92,11 +92,12 @@ share_client = service.get_share_client(file_share_name)
 
 user_id = create_user_id()
 
-vectorstore_graphs = Chroma(persist_directory="/home/tim/ai4s/climate_qa/climate-question-answering/data/vectorstore_owid", embedding_function=embeddings_function) # TODO make it api call
 
 
 # Create vectorstore and retriever
-vectorstore = get_pinecone_vectorstore(embeddings_function)
+vectorstore = get_pinecone_vectorstore(embeddings_function, index_name = os.getenv("PINECONE_API_INDEX"))
+vectorstore_graphs = get_pinecone_vectorstore(embeddings_function, index_name = os.getenv("PINECONE_API_INDEX_OWID"), text_key="title")
+
 llm = get_llm(provider="openai",max_tokens = 1024,temperature = 0.0)
 reranker = get_reranker("nano")
 # agent = make_graph_agent(llm,vectorstore,reranker)

climateqa/engine/chains/retriever.py

@@ -87,9 +87,10 @@ def make_retriever_node(vectorstore,reranker,rerank_by_question=True, k_final=15
                 vectorstore=vectorstore,
                 sources = sources,
                 # reports = ias_reports,
-            min_size = 200,
-            k_summary = k_summary,k_total = k_before_reranking,
-            threshold = 0.5,
+                min_size = 200,
+                k_summary = k_summary,
+                k_total = k_before_reranking,
+                threshold = 0.5,
             )
             docs_question = retriever.get_relevant_documents(question)
             

climateqa/engine/vectorstore.py

@@ -19,7 +19,7 @@ def get_chroma_vectorstore(embedding_function, persist_directory="/home/dora/cli
     return vectorstore
 
 
-def get_pinecone_vectorstore(embeddings,text_key = "content"):
+def get_pinecone_vectorstore(embeddings,text_key = "content", index_name = os.getenv("PINECONE_API_INDEX")):
 
     # # initialize pinecone
     # pinecone.init(
@@ -33,7 +33,7 @@ def get_pinecone_vectorstore(embeddings,text_key = "content"):
     # return vectorstore
 
     pc = Pinecone(api_key=os.getenv("PINECONE_API_KEY"))
-    index = pc.Index(os.getenv("PINECONE_API_INDEX"))
+    index = pc.Index(index_name)
 
     vectorstore = PineconeVectorstore(
         index, embeddings, text_key,