[Request for Comments] KerasTuner new feature: tuning end-to-end ML workflows

Hi Everyone,

we propose some exciting improvements in KerasTuner to be implemented in the near future.

It allows you to:

  • Tune the hyperparameters in any step in a ML workflow, including data preprocessing, model building, and training.
  • Tune your existing Keras code with little modification.

You feedback is valuable to us!

Please let us know what you think of these feature by commenting on the pull request. [Link]

To view the proposal: [Link]