将数据从Google数据存储加载到Google Cloud ML Engine

Question

I was trying to run a ML Engine training job reading data from Cloud Datastore, but a got a permission error:

line 434, in _end_unary_response_blocking raise _Rendezvous(state, None, None, deadline) grpc._channel._Rendezvous: <_Rendezvous of RPC that terminated with (StatusCode.PERMISSION_DENIED, Missing or insufficient permissions.)> The above exception was the direct cause of the following exception: Traceback (most recent call last): File "/usr/lib/python3.5/runpy.py", 

I ran the same script locally and it run well. I searched how to add Datastore permission to Cloud ML but didn't find anything...

Answer 1

It's unclear if you are using Cloud ML Engine & Cloud Datastore in the same project, or if you have the Cloud Datastore API enabled. You can probably follow the instructions at https://cloud.google.com/ml-engine/docs/tensorflow/working-with-cloud-storage#setup-different-project to ensure the service account your Cloud ML Engine workload runs as is able to access Cloud Datastore.