culebra.solution.feature_selection.Solution class

class Solution(species: Species, fitness_cls: type[Fitness], features: Sequence[int] | None = None)

Bases: Solution

Construct a default solution.

Parameters:

• species (Species) – The species the solution will belong to

• fitness_cls (type[Fitness]) – The solution’s fitness class

• features (Sequence[int]) – Initial features

Raises:

• TypeError – If species is not a valid species

• TypeError – If fitness_cls is not a valid fitness class

Class attributes

Solution.species_cls = <class 'culebra.solution.feature_selection.Species'>

Class for the species used by the Solution class to constrain all its instances.

Class methods

classmethod Solution.load(filename: str) → Base

Load a serialized object from a file.

Parameters:

filename (str) – The file name.

Returns:

The loaded object

Raises:

• TypeError – If filename is not a valid file name

• ValueError – If the filename extension is not SERIALIZED_FILE_EXTENSION

Properties

abstract property Solution.features: ndarray[int]

Features selected by the solution

This property must be overridden by subclasses to return a correct value.

Returns:

Return an ordered sequence with the indices of the selected features

Return type:

ndarray[int]

Setter:

Set a new set of selected features

Parameters:

values (Sequence[int]) – The indices of the new features

Raises:

• ValueError – If set to new feature indices values which do not meet the species constraints.

• NotImplementedError – If has not been overridden

property Solution.fitness: Fitness

Solution’s fitness.

Return type:

Fitness

Setter:

Set a new Fitness

Parameters:

value (Fitness) – The new fitness

abstract property Solution.max_feat: int | None

Maximum feature index selected by the solution.

This property must be overridden by subclasses to return a correct value.

Returns:

The maximum feature index or None if no feature has been selected

Return type:

int

Raises:

NotImplementedError – If has not been overridden

abstract property Solution.min_feat: int | None

Minimum feature index selected by the solution.

This property must be overridden by subclasses to return a correct value.

Returns:

The minimum feature index or None if no feature has been selected

Return type:

int

Raises:

NotImplementedError – If has not been overridden

abstract property Solution.num_feats: int

Number of features selected by the solution.

This property must be overridden by subclasses to return a correct value.

Return type:

int

Raises:

NotImplementedError – If has not been overridden

property Solution.species: Species

Solution’s species.

Return type:

Species

Methods

Solution.delete_fitness() → None

Delete the solution’s fitness.

Solution.dominates(other: Solution) → bool

Dominate operator.

Parameters:

other (Solution) – Other solution

Returns:

True if each objective of the solution is not strictly worse than the corresponding objective of other and at least one objective is strictly better.

Return type:

bool

Solution.dump(filename: str) → None

Serialize this object and save it to a file.

Parameters:

filename (str) – The file name.

Raises:

• TypeError – If filename is not a valid file name

• ValueError – If the filename extension is not SERIALIZED_FILE_EXTENSION

Private methods

abstract Solution._setup() → None

Init the features of this solution randomly.

This method must be overridden by subclasses.

Raises:

NotImplementedError – If has not been overridden

Dunder methods

Intended to compare (lexicographically) two solutions according to their fitness.

Solution.__eq__(other: Solution) → bool

Equality test.

Parameters:

other (Solution) – Other solution

Returns:

True if other codes the same solution, or False otherwise

Return type:

bool

Solution.__ge__(other: Solution) → bool

Greater than or equal to operator.

Parameters:

other (Solution) – Other solution

Returns:

True if the solution’s fitness is greater than or equal to the other’s fitness

Return type:

bool

Solution.__gt__(other: Solution) → bool

Greater than operator.

Parameters:

other (Solution) – Other solution

Returns:

True if the solution’s fitness is greater than the other’s fitness

Return type:

bool

Solution.__hash__() → int

Return the hash number for this solution.

The hash number is used for equality comparisons. Currently is implemented as the hash of the solution’s string representation.

Return type:

int

Solution.__le__(other: Solution) → bool

Less than or equal to operator.

Parameters:

other (Solution) – Other solution

Returns:

True if the solution’s fitness is less than or equal to the other’s fitness

Return type:

bool

Solution.__lt__(other: Solution) → bool

Less than operator.

Parameters:

other (Solution) – Other solution

Returns:

True if the solution’s fitness is less than the other’s fitness

Return type:

bool

Solution.__ne__(other: Solution) → bool

Not equality test.

Parameters:

other (Solution) – Other solution

Returns:

False if other codes the same solutions, or True otherwise

Return type:

bool

Solution.__str__() → str

Solution as a string.

Return type:

str