gemseo_mlearning / adaptive / criteria / optimum

Show inherited members

criterion module

Common expected improvement of the regression model.

This is the same as the expected improvement of the regression model for the minimum.

Statistics:

\[EI[x] = E[\max(y_{min}-Y(x),0)]\]

where \(y_{min}=\min_{1\leq i \leq n}~y^{(i)}\).

Bootstrap estimator:

\[\widehat{EI}[x] = \frac{1}{B}\sum_{b=1}^B \max(f_{min}-Y_b(x),0)\]
class gemseo_mlearning.adaptive.criteria.optimum.criterion.ExpectedImprovement(algo_distribution, **options)[source]

Bases: MinExpectedImprovement

The expected improvement.

This criterion is scaled by the output range.

# noqa: D205 D212 D415

Parameters:

  • algo_distribution (MLRegressorDistribution) – The distribution of a machine learning algorithm.

  • **options (MLDataAcquisitionCriterionOptionType) – The acquisition criterion options.

algo_distribution: MLRegressorDistribution

The distribution of a machine learning algorithm assessor.

force_real: bool

Whether to cast the results to real value.

has_default_name: bool

Whether the name has been set with a default value.

last_eval: OutputType | None

The value of the function output at the last evaluation.

None if it has not yet been evaluated.

output_range: float

The output range.

special_repr: str

The string representation of the function overloading its default string ones.