Developer API for design spaces

class smt.utils.design_space.BaseDesignSpace(design_variables: List[DesignVariable] = None)

Interface for specifying (hierarchical) design spaces.

This class itself only specifies the functionality that any design space definition should implement: - a way to specify the design variables, their types, and their bounds or options - a way to correct a set of design vectors such that they satisfy all design space hierarchy constraints - a way to query which design variables are acting for a set of design vectors - a way to impute a set of design vectors such that non-acting design variables are assigned some default value - a way to sample n valid design vectors from the design space

If you want to actually define a design space, use the DesignSpace class!

Note that the correction, querying, and imputation mechanisms should all be implemented in one function (correct_get_acting), as usually these operations are tightly related.

Attributes:

design_variables

is_all_cont

Whether or not the space is continuous

is_cat_mask

Boolean mask specifying for each design variable whether it is a categorical variable

is_conditionally_acting

Boolean mask specifying for each design variable whether it is conditionally acting (can be non-acting)

n_dv

Get the number of design variables

Methods

correct_get_acting(x)	Correct the given matrix of design vectors and return the corrected vectors and the is_acting matrix.
decode_values(x[, i_dv])	Return decoded values: converts ordinal and categorical back to their original values.
fold_x(x[, is_acting, fold_mask])	Fold x and optionally is_acting.
get_num_bounds()	Get bounds for the design space.
get_unfolded_num_bounds()	Get bounds for the unfolded continuous space.
get_x_limits()	Returns the variable limit definitions in SMT < 2.0 style
sample_valid_x(n[, unfolded, random_state])	Sample n design vectors and additionally return the is_acting matrix.
unfold_x(x[, is_acting, fold_mask])	Unfold x and optionally is_acting.

_get_design_variables() → List[DesignVariable]

Return the design variables defined in this design space if not provided upon initialization of the class

_is_conditionally_acting() → ndarray

Return for each design variable whether it is conditionally acting or not. A design variable is conditionally acting if it MAY be non-acting.

Returns:

is_conditionally_acting: np.ndarray [dim]

• Boolean vector specifying for each design variable whether it is conditionally acting

_correct_get_acting(x: ndarray) → Tuple[ndarray, ndarray]

Correct the given matrix of design vectors and return the corrected vectors and the is_acting matrix.

Parameters:

x: np.ndarray [n_obs, dim]

• Input variables

Returns:

x_corrected: np.ndarray [n_obs, dim]

• Corrected and imputed input variables

is_acting: np.ndarray [n_obs, dim]

• Boolean matrix specifying for each variable whether it is acting or non-acting

_sample_valid_x(n: int, random_state=None) → Tuple[ndarray, ndarray]

Sample n design vectors and additionally return the is_acting matrix.

Returns:

x: np.ndarray [n, dim]

• Valid design vectors

is_acting: np.ndarray [n, dim]

• Boolean matrix specifying for each variable whether it is acting or non-acting