automatically set refit_with_val when early stopping is enabled

Nixtla / neuralforecast

Scalable and user friendly neural :brain: forecasting algorithms.

https://nixtlaverse.nixtla.io/neuralforecast

Apache License 2.0

2.69k stars 312 forks source link

automatically set refit_with_val when early stopping is enabled #1031

Closed jmoralez closed 3 days ago

jmoralez commented 3 weeks ago

Automatically sets refit_with_val=True if the user provided early_stop_patience_steps > 0 in the auto models to avoid errors like the one in #1009. I decided to automatically set it here instead of raising an error to avoid cases where the hyperparameter optimization concluded and there's an error when fitting the best model which also makes the user loose the results of the optimization.

review-notebook-app[bot] commented 3 weeks ago

Check out this pull request on

See visual diffs & provide feedback on Jupyter Notebooks.