Dataset to Model to Endpoint
The process I completed links these resources across the two projects:
-
Dataset Creation: I have created the raw data container, Rag_AutoML, in the project (PII Removed by Staff). (Caller Project)
-
Model Training: I ran the AutoML pipeline, which consumed the dataset from pristine-valve-477208-i1 and used it to train a machine learning model named solar_forecast_automl_model.
-
Model Deployment (The Critical Step): Google Cloud automatically deployed the resulting, trained model to an Endpoint in the Core Project (PII Removed by Staff).
The Final Resource
The final resource my Python script needs to call is the Deployed Endpoint, which lives in the Core Project:
-
Resource to Call: Endpoint ID 7273285910412656640
-
Hosting Project: (PII Removed by Staff)
Even though the data originated in the auxiliary project, the final model asset and the serving endpoint (the active prediction service) are bound to the core project. This is why I must set my configuration variables to target (PII Removed by Staff) to solve the 403 Permission Denied error , and I don’t know how that core project was created , why I have been given a sub-Project Id from (PII Removed by Staff). with “i” . If anyone know how to sort that out , I would highly appreciate it .