[
https://issues.apache.org/jira/browse/AIRFLOW-1271?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16051237#comment-16051237
]
Yu Ishikawa commented on AIRFLOW-1271:
--------------------------------------
[~pdwryst] Thank you for suggesting it. Do you have any update this issue? I am
really interested in this issue. If you have much time to develop this feature,
I can take over it.
> Google Cloud ML Training Operator
> ---------------------------------
>
> Key: AIRFLOW-1271
> URL: https://issues.apache.org/jira/browse/AIRFLOW-1271
> Project: Apache Airflow
> Issue Type: New Feature
> Components: operators
> Reporter: Peter Dolan
> Assignee: Peter Dolan
>
> There should be an Operator for the Google Cloud ML Training Job, which
> supports executing Tensorflow neural network training as a service.
> Documentation is here:
> https://cloud.google.com/ml-engine/docs/how-tos/training-jobs
> This should support launching the training job on the service via the API
> call in operation, or running the job locally on your machine via the 'gcloud
> ml-engine local' command for end-to-end testing. Documentation is here:
> https://cloud.google.com/sdk/gcloud/reference/ml-engine/local/
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
