Source code for optuna.integration.chainer

import optuna
from optuna import type_checking

try:
    import chainer
    from chainer.training.extension import Extension
    from chainer.training import triggers

    _available = True
except ImportError as e:
    _import_error = e
    # ChainerPruningExtension is disabled because Chainer is not available.
    _available = False
    # This alias is required to avoid ImportError at ChainerPruningExtension definition.
    Extension = object

if type_checking.TYPE_CHECKING:
    from typing import Tuple
    from typing import Union

    TriggerType = Union[
        Tuple[(int, str)], triggers.IntervalTrigger, triggers.ManualScheduleTrigger
    ]


[docs]class ChainerPruningExtension(Extension): """Chainer extension to prune unpromising trials. See `the example <https://github.com/optuna/optuna/blob/master/ examples/pruning/chainer_integration.py>`__ if you want to add a pruning extension which observes validation accuracy of a `Chainer Trainer <https://docs.chainer.org/en/stable/ reference/generated/chainer.training.Trainer.html>`_. Args: trial: A :class:`~optuna.trial.Trial` corresponding to the current evaluation of the objective function. observation_key: An evaluation metric for pruning, e.g., ``main/loss`` and ``validation/main/accuracy``. Please refer to `chainer.Reporter reference <https://docs.chainer.org/en/stable/reference/ util/generated/chainer.Reporter.html>`_ for further details. pruner_trigger: A trigger to execute pruning. ``pruner_trigger`` is an instance of `IntervalTrigger <https://docs.chainer.org/en/stable/reference/generated/ chainer.training.triggers.IntervalTrigger.html>`_ or `ManualScheduleTrigger <https://docs.chainer.org/en/stable/reference/generated/ chainer.training.triggers.ManualScheduleTrigger.html>`_. `IntervalTrigger <https:// docs.chainer.org/en/stable/reference/generated/chainer.training.triggers. IntervalTrigger.html>`_ can be specified by a tuple of the interval length and its unit like ``(1, 'epoch')``. """ def __init__(self, trial, observation_key, pruner_trigger): # type: (optuna.trial.Trial, str, TriggerType) -> None _check_chainer_availability() self._trial = trial self._observation_key = observation_key self._pruner_trigger = chainer.training.get_trigger(pruner_trigger) if not ( isinstance(self._pruner_trigger, triggers.IntervalTrigger) or isinstance(self._pruner_trigger, triggers.ManualScheduleTrigger) ): pruner_type = type(self._pruner_trigger) raise TypeError( "Invalid trigger class: " + str(pruner_type) + "\n" "Pruner trigger is supposed to be an instance of " "IntervalTrigger or ManualScheduleTrigger." ) @staticmethod def _get_float_value(observation_value): # type: (Union[float, chainer.Variable]) -> float _check_chainer_availability() if isinstance(observation_value, chainer.Variable): observation_value = observation_value.data try: observation_value = float(observation_value) except TypeError: raise TypeError( "Type of observation value is not supported by ChainerPruningExtension.\n" "{} cannot be casted to float.".format(type(observation_value)) ) return observation_value def _observation_exists(self, trainer): # type: (chainer.training.Trainer) -> bool return self._pruner_trigger(trainer) and self._observation_key in trainer.observation def __call__(self, trainer): # type: (chainer.training.Trainer) -> None if not self._observation_exists(trainer): return current_score = self._get_float_value(trainer.observation[self._observation_key]) current_step = getattr(trainer.updater, self._pruner_trigger.unit) self._trial.report(current_score, step=current_step) if self._trial.should_prune(): message = "Trial was pruned at {} {}.".format(self._pruner_trigger.unit, current_step) raise optuna.TrialPruned(message)
def _check_chainer_availability(): # type: () -> None if not _available: raise ImportError( "Chainer is not available. Please install Chainer to use this feature. " "Chainer can be installed by executing `$ pip install chainer`. " "For further information, please refer to the installation guide of Chainer. " "(The actual import error is as follows: " + str(_import_error) + ")" )