GitHub Issue Summarization for 1.0

[jlewi]

The GitHub issue summarization example needs to be updated for 1.0

A couple suggestions

• We should create a notebook that can run in KF for running the example E2E

  • This eliminates the need for users to type and copy and paste a lot of things
  • This makes it easier to test
  • We should create Cloud/platform specific variants of the notebooks

• We should get rid of distributed training

  • I don't think this every worked; see [gh_issue_summarization] distributed training using Keras #196 and [GH Issue Summarization] Keras model doesn't work with keras in TensorFlow #280

• We should try to use object storage when possible

  • This is easier then trying to move PVs around or creating read write many PVCs like NFS
  • If Keras can't write directly to object store we can just add some code to upload the model
    after training ends

[issue-label-bot]

Issue-Label Bot is automatically applying the labels:

Label	Probability
kind/feature	0.70

Please mark this comment with 👍 or 👎 to give our bot feedback!
Links: app homepage, dashboard and code for this bot.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.