add pinecone

kb.py

@@ -0,0 +1,18 @@
+
+from langchain_pinecone import PineconeVectorStore
+from langchain_core.vectorstores import VectorStore
+from langchain_core.vectorstores import VectorStoreRetriever
+from langchain_aws import BedrockEmbeddings
+
+INDEX_NAME = "zeteo-health"
+NAMESPACE = "knowedgebase"
+EMBEDDING_DIMENSION = 1536
+EMBEDDING_MODEL_NAME = "amazon.titan-embed-text-v1"
+
+vectorstore = PineconeVectorStore(INDEX_NAME, 
+                    embedding=BedrockEmbeddings(model_id=EMBEDDING_MODEL_NAME, 
+                    namespace=NAMESPACE))
+
+def search_with_score(self, query):
+    documents = [doc[0] for doc in self.vectorstore.similarity_search_with_score(query, k=6)  if doc[1] > self.similarity_threshold]
+    return documents

main.py

@@ -2,6 +2,7 @@ from fastapi import FastAPI, Request, Query
 import logging
 from whatsapp_client import WhatsAppWrapper
 import json
+from kb import search_with_score
 
 logger = logging.getLogger()
 
@@ -42,8 +43,10 @@ async def callback(request: Request):
     if 'from' in message:
         phone_num = message['from']
         mesg_body = value['messages'][0]['text']['body']
+        documents = search_with_score(mesg_body)
+        print(f"DOCUMENT: {len(documents)}")
         client = WhatsAppWrapper()
         client.send_text_message("hello, right back at ya ..." + mesg_body,  phone_num)
 
     return {"status": "success"}
-
+    

requirements.txt

@@ -1,4 +1,13 @@
 fastapi
 uvicorn[standard]
 python-dotenv
-requests
+requests
+pinecone-client
+langchain
+langchain-aws
+langchain-community
+langchain-pinecone
+pinecone-client
+python-dotenv
+tiktoken
+boto3