deephyper.skopt.optimizer.optimizer.Categorical#

class deephyper.skopt.optimizer.optimizer.Categorical(categories, prior=None, transform=None, name=None)[source]#

Bases: Dimension

Search space dimension that can take on categorical values.

Parameters:
  • categories (list, shape=(n_categories,)) – Sequence of possible categories.

  • prior (list, shape=(categories,), default=None) – Prior probabilities for each category. By default all categories are equally likely.

  • transform ("onehot", "string", "identity", "label", default="onehot") –

    • “identity”, the transformed space is the same as the original space.

    • ”string”, the transformed space is a string encoded representation of the original space.

    • ”label”, the transformed space is a label encoded representation (integer) of the original space.

    • ”onehot”, the transformed space is a one-hot encoded representation of the original space.

  • name (str or None) – Name associated with dimension, e.g., “colors”.

Methods

distance

Compute distance between category a and b.

inverse_transform

Inverse transform samples from the warped space back into the original space.

rvs

Draw random samples.

set_transformer

Define _rvs and transformer spaces.

transform

Transform samples form the original space to a warped space.

Attributes

bounds

is_constant

name

prior

size

transformed_bounds

transformed_size

distance(a, b)[source]#

Compute distance between category a and b.

As categories have no order the distance between two points is one if a != b and zero otherwise.

Parameters:
  • a (category) – First category.

  • b (category) – Second category.

inverse_transform(Xt)[source]#

Inverse transform samples from the warped space back into the original space.

rvs(n_samples=None, random_state=None)[source]#

Draw random samples.

Parameters:
  • n_samples (int or None) – The number of samples to be drawn.

  • random_state (int, RandomState instance, or None (default)) – Set random state to something other than None for reproducible results.

set_transformer(transform='onehot')[source]#

Define _rvs and transformer spaces.

Parameters:

transform (str) – Can be ‘normalize’, ‘onehot’, ‘string’, ‘label’, or ‘identity’

transform(X)#

Transform samples form the original space to a warped space.