Update appChatbot.py

appChatbot.py

@@ -31,6 +31,8 @@ embeddings = HuggingFaceEmbeddings(model_name="all-MiniLM-L6-v2")
 ABS_PATH = os.path.dirname(os.path.abspath(__file__))
 DB_DIR = os.path.join(ABS_PATH, "db")
 
+vectorstore = None
+
 def replace_newlines_and_spaces(text):
     # Replace all newline characters with spaces
     text = text.replace("\n", " ")
@@ -63,8 +65,7 @@ def init_chromadb():
     # Create the vectorestore to use as the index
     vectorstore = Chroma.from_documents(texts, embeddings, persist_directory=DB_DIR)
     vectorstore.persist()
-    print(vectorstore)
-    vectorstore = None
+    print("vectorstore::", vectorstore)
 
 def query_chromadb(ASK):
     if not os.path.exists(DB_DIR):
@@ -75,8 +76,8 @@ def query_chromadb(ASK):
 
     result = vectorstore.similarity_search_with_score(query=ASK, k=4)
     jsonable_result = jsonable_encoder(result)
-    print(json.dumps(jsonable_result, indent=2))
-    return json.dumps(jsonable_result, indent=2)
+    print("Json pdf response ::", json.dumps(jsonable_result, indent=2))
+    #return json.dumps(jsonable_result, indent=2)
 
 
 def respond(
@@ -106,11 +107,18 @@ def respond(
         temperature=temperature,
         top_p=top_p,
     ):
+               
         token = message.choices[0].delta.content
 
         response += token
-        #yield response
-        yield query_chromadb(message)
+
+        print ("message::",message)
+        print ("token::",token)
+        print ("response::",response)
+        print ("query_chromadb::",query_chromadb("how could an AI be used in smart water management systems?"))
+        
+        yield response
+        #yield query_chromadb(message)
 
 
 """