Source code for optuna.integration.pytorch_ignite

import optuna
from optuna.trial import Trial


with optuna._imports.try_import() as _imports:
    from ignite.engine import Engine


[docs]class PyTorchIgnitePruningHandler: """PyTorch Ignite handler to prune unpromising trials. See `the example <https://github.com/optuna/optuna-examples/blob/main/ pytorch/pytorch_ignite_simple.py>`__ if you want to add a pruning handler which observes validation accuracy. Args: trial: A :class:`~optuna.trial.Trial` corresponding to the current evaluation of the objective function. metric: A name of metric for pruning, e.g., ``accuracy`` and ``loss``. trainer: A trainer engine of PyTorch Ignite. Please refer to `ignite.engine.Engine reference <https://pytorch.org/ignite/engine.html#ignite.engine.Engine>`_ for further details. """ def __init__(self, trial: Trial, metric: str, trainer: "Engine") -> None: _imports.check() self._trial = trial self._metric = metric self._trainer = trainer def __call__(self, engine: "Engine") -> None: score = engine.state.metrics[self._metric] self._trial.report(score, self._trainer.state.epoch) if self._trial.should_prune(): message = "Trial was pruned at {} epoch.".format(self._trainer.state.epoch) raise optuna.TrialPruned(message)