optuna-worker

optuna-worker is a python package to help using optuna without code modification to existing ML training project, with two assumptions.

1. You could pass hyperparameters to your trainer via cli arguments.
2. Metrics to be optimized are printed out as formatted logs during training.

A CLI named optuna-worker is included in this package, which runs optuna worker with a configuration yaml file.

optuna-worker run $CONFIG_FILE

All it does is simply creating a proxy objective that

1. Creates a training subprocess, passing suggested parameters via CLI arguments.
2. Parses metric values from stdout of the training subprocess, so that they could be reported and returned.

Details of the configuration file could be found from the documentation. You can also find working example configuration files from examples.

Installation

pip install optuna-worker

License

optuna-worker is distributed under the terms of the MIT license.