deephyper.problem#

This sub-package provides tools to define hyperparameter and neural architecture search problems. Some features of this module are based on the ConfigSpace project. The main classes provided by this module are: * deephyper.problem.HpProblem: A class to define a hyperparameter search problem. * deephyper.problem.NaProblem: A class to define a neural architecture search problem.

Functions

`Categorical`	Creates a Categorical Hyperparameter.
`Float`	Create a FloatHyperparameter.
`Integer`	Create an IntegerHyperparameter.

Classes

`AndConjunction`
`Beta`	Represents a beta distribution.
`BetaFloatHyperparameter`
`BetaIntegerHyperparameter`
`CategoricalHyperparameter`
`Configuration`
`ConfigurationSpace`
`Constant`
`Distribution`	Base distribution type
`EqualsCondition`
`ForbiddenAndConjunction`	A ForbiddenAndConjunction.
`ForbiddenEqualsClause`	A ForbiddenEqualsClause
`ForbiddenEqualsRelation`	A ForbiddenEquals relation between two hyperparameters.
`ForbiddenGreaterThanRelation`	A ForbiddenGreaterThan relation between two hyperparameters.
`ForbiddenInClause`
`ForbiddenLessThanRelation`	A ForbiddenLessThan relation between two hyperparameters.
`GreaterThanCondition`
`HpProblem`	Class to define an hyperparameter problem.
`InCondition`
`LessThanCondition`
`NaProblem`	A Neural Architecture Problem specification for Neural Architecture Search.
`Normal`	Represents a normal distribution.
`NormalFloatHyperparameter`
`NormalIntegerHyperparameter`
`NotEqualsCondition`
`OrConjunction`
`OrdinalHyperparameter`
`UnParametrizedHyperparameter`
`Uniform`	A uniform distribution
`UniformFloatHyperparameter`
`UniformIntegerHyperparameter`