Monitor Scikit Optimize hyperparameter optimization in Neptune¶

Prerequisites¶

Integration with Scikit Optimize framework is introduced as a part of logging module so just need to have neptune-client and neptune-contrib installed.

pip install neptune-client neptune-contrib['monitoring']

Initialize Neptune and create an experiment¶

import neptune
neptune.init('jakub-czakon/blog-hpo')
neptune.create_experiment(name='skopt sweep')

Create NeptuneMonitor callback¶

Pass the experiment object as first argument.

Note

To be able to log information after the .fit() method finishes remember to pass close_after_train=False

import neptunecontrib.monitoring.skopt as sk_utils
neptune_monitor = sk_utils.NeptuneMonitor()

Pass neptune_monitor to skopt.forest_minimize or others¶

It will monitor the metrics and parameters checked at each run.

results = skopt.forest_minimize(objective, space, callback=[neptune_monitor],
                                base_estimator='ET', n_calls=100, n_random_starts=10)
sk_utils.log_results(results)

Log all results¶

It will log the following things to Neptune: * best score * best parameters * plot_convergence figure * plot_evaluations figure * plot_objective figure

sk_utils.log_results(results)

Monitor your Scikit Optimize training in Neptune¶

Now you can watch your Scikit Optimize hyperparameter optimization in Neptune!

Check out this example experiment.