optuna.integration.PyTorchIgnitePruningHandler¶
- class optuna.integration.PyTorchIgnitePruningHandler(trial: Trial, metric: str, trainer: Engine)[source]¶
PyTorch Ignite handler to prune unpromising trials.
See the example if you want to add a pruning handler which observes validation accuracy.
- Parameters
trial – A
Trial
corresponding to the current evaluation of the objective function.metric – A name of metric for pruning, e.g.,
accuracy
andloss
.trainer – A trainer engine of PyTorch Ignite. Please refer to ignite.engine.Engine reference for further details.
Methods
__init__
(trial, metric, trainer)