fixing issue with embedding model initialization.

src/infra/cloudformation/eventbridge.yaml

@@ -28,7 +28,7 @@ Resources:
     Properties:
       Name: rss-feed-processor-schedule
       Description: Runs the RSS Feed Processor Lambda function every hour
-      ScheduleExpression: rate(5 minutes)
+      ScheduleExpression: rate(30 minutes)
       FlexibleTimeWindow: 
         Mode: FLEXIBLE
         MaximumWindowInMinutes: 1

src/infra/lambdas/RSSFeedProcessorLambda/src/analytics/embeddings/vector_db.py

@@ -11,8 +11,7 @@ embedding_dim = os.getenv("VECTOR_EMBEDDING_DIM")
 vector_search_metric = os.getenv("VECTOR_SEARCH_METRIC")
 index_name = os.getenv("PINECONE_DB_NAME")
 
-client = OpenAI()
-
+client = OpenAI() # For Embedding Models, Not LLMs
 pc = Pinecone(api_key=api_key)
 
 def get_index():

src/infra/lambdas/RSSFeedProcessorLambda/src/lambda_function.py

@@ -36,7 +36,6 @@ def lambda_handler(event, context):
 
         # Process the feed
         result = extract_feed(feed)
-        print(type(result))
         logger.info("Process Feed Result Dictionary: ", result)
         last_pub_dt = result['max_date']
 

src/infra/lambdas/lambda_utils/update_lambda_env_vars.py

@@ -20,7 +20,8 @@ def update_env_vars(function_name):
         'VECTOR_EMBEDDING_MODEL': os.environ.get('VECTOR_EMBEDDING_MODEL'),
         'VECTOR_EMBEDDING_DIM': os.environ.get('VECTOR_EMBEDDING_DIM'),
         'VECTOR_SEARCH_METRIC': os.environ.get('VECTOR_SEARCH_METRIC'),
-        'PINECONE_DB_NAME': os.environ.get('PINECONE_DB_NAME')
+        'PINECONE_DB_NAME': os.environ.get('PINECONE_DB_NAME'),
+        'OPENAI_API_KEY': os.environ.get('OPENAI_API_KEY')
     }
     
     return lambda_client.update_function_configuration(

todo.md

@@ -1,5 +1,5 @@
 # Testing🧪
-* Testing from 3rd party aws account. [ Today ]
+* Testing from 3rd party aws account. 
 * Test Large Amounts of Feeds ( Decrease the cadence of ingesting. ) [ Today ]
 * Test out Vector Databases at Small Scale
 * Test out Vector Databases at Scale.