def _find_similar_experiments(self):
"""After locating similar experiments by way of the parent's :meth:`_find_similar_experiments`, fit :attr:`optimizer`
with the hyperparameters and results of each located experiment"""
super()._find_similar_experiments()
for _i, _experiment in enumerate(self.similar_experiments[::-1]):
_hyperparameters = dimension_subset(
_experiment[0], self.hyperparameter_space.get_names())
_evaluation = _experiment[1]
_experiment_id = _experiment[2] if len(_experiment) > 2 else None
self.logger.print_result(_hyperparameters,
_evaluation,
experiment_id=_experiment_id)
# FLAG: Resolve switching between below options depending on `target_metric`
# self.optimizer_result = self.optimizer.tell(_hyperparameters, _evaluation)
self.optimizer_result = self.optimizer.tell(
_hyperparameters, -_evaluation)
# FLAG: Resolve switching between above options depending on `target_metric`
# self.optimizer_result = self.optimizer.tell(
# _hyperparameters, _evaluation, fit=(_i == len(self.similar_experiments) - 1))
if eval_callbacks(self.callbacks, self.optimizer_result):
return self.optimizer_result
def filter_by_space(hyperparameters_and_scores, hyperparameter_space):
"""Reject any `hyperparameters_and_scores` tuples whose hyperparameters do not fit within
`hyperparameter_space`
Parameters
----------
hyperparameters_and_scores: List of tuples
Each tuple in list should be a pair of form (hyperparameters <dict>, evaluation <float>),
where the hyperparameter dict should contain at least the following keys:
['model_init_params', 'model_extra_params', 'preprocessing_pipeline',
'preprocessing_params', 'feature_selector']
hyperparameter_space: instance of :class:`space.Space`
The boundaries of the hyperparameters to be searched
Returns
-------
hyperparameters_and_scores: List of tuples
Filtered to include only those whose hyperparameters fit within `hyperparameter_space`"""
dimension_names = hyperparameter_space.names()
hyperparameters_and_scores = list(
filter(
lambda _: dimension_subset(_[0], dimension_names) in
hyperparameter_space,
hyperparameters_and_scores,
))
return hyperparameters_and_scores
def _find_similar_experiments(self):
"""After locating similar experiments by way of the parent's
:meth:`_find_similar_experiments`, fit :attr:`optimizer` with the hyperparameters and
results of each located experiment"""
super()._find_similar_experiments()
# TODO: Remove below reversal of `similar_experiments` when `result_reader.ResultFinder.sort` finished
for _i, _experiment in enumerate(self.similar_experiments[::-1]):
_hyperparameters = dimension_subset(_experiment[0], self.space.names())
_evaluation = _experiment[1]
_experiment_id = _experiment[2] if len(_experiment) > 2 else None
self.logger.print_result(_hyperparameters, _evaluation, experiment_id=_experiment_id)
self._update_optimizer(_hyperparameters, _evaluation)
# self.optimizer_result = self.optimizer.tell(
# _hyperparameters, _evaluation, fit=(_i == len(self.similar_experiments) - 1))
if eval_callbacks(self.callbacks, self.optimizer_result):
return self.optimizer_result
def does_fit_in_space(root, space):
"""Determine if the subset of `root` identified by `space` fits within dimensions of `space`
Parameters
----------
root: Object
Iterable, whose values at the locations specified in `space` will be checked. For each
dimension in `space`, the dimension's `location`/`name` is looked up in `root`, and the
value is tested to see if it falls within the dimension's range of allowed values
space: `space.Space`
Instance of :class:`space.Space` that defines dimension choices for select hyperparameters.
Each dimension in `space` should have an appropriate `name` (or `location`, if necessary)
attribute to match `root`
Returns
-------
Boolean
True if `root` subset (at `space` locations) fits in `space` dimensions. Else, False"""
return dimension_subset(root, space.names()) in space