This repository was archived by the owner on Dec 11, 2023. It is now read-only.

Description
I need to try out different hyperparameters and compare their performance. I would be interested to know if an automated hyperparameter tuning option is available to do a Bayesian optimization as an example. However, it looks like the only option I have is to do a manual search.
To summarize what I want to know whether there are built-in methods in Tensorflow NMT to optimize hyperparameters. Unfortunately, I didn't find any. Ex: Grid search, Random Search.