Upload app.py

app.py

@@ -20,22 +20,22 @@ collection = client[db_name][collection_name]
 
 ## Create a vector search index
 print ('Creating vector search index')
-collection.create_search_index(model={"definition": {"mappings":{
-  "dynamic":True,
-  "fields": {
-    "plot_embedding": {
-      "type": "knnVector",
-      "dimensions": 1536,
-      "similarity": "euclidean"
-    }
-  }
-}}, "name":'default'})
+# collection.create_search_index(model={"definition": {"mappings":{
+#   "dynamic":True,
+#   "fields": {
+#     "plot_embedding": {
+#       "type": "knnVector",
+#       "dimensions": 1536,
+#       "similarity": "euclidean"
+#     }
+#   }
+# }}, "name":'default'})
 
 # sleep for minute
-print ('Waiting for vector index on field "embedding" to be created')
-time.sleep(60)
+# print ('Waiting for vector index on field "embedding" to be created')
+# time.sleep(60)
 
-vector_store = MongoDBAtlasVectorSearch(embedding=OpenAIEmbeddings(), collection=collection, index_name='default', text_key='plot', embedding_key='plot_embedding')
+vector_store = MongoDBAtlasVectorSearch(embedding=OpenAIEmbeddings(), collection=collection, index_name='vector_index', text_key='plot', embedding_key='plot_embedding')
 
 def get_movies(message, history):
     movies = vector_store.similarity_search(message, 3)