fishy.models.evolutionary.gp

Genetic Programming (GP) engine for multi-tree evolution. Standardized to provide a scikit-learn compatible interface.

Classes

class fishy.models.evolutionary.gp.GP(generations: int = 10, population_size: int = 100, crossover_rate: float = 0.8, mutation_rate: float = 0.2, elitism: float = 0.1, max_depth: int = 6, random_state: int = 42)

Bases: BaseEstimator, ClassifierMixin

Scikit-learn compatible wrapper for Multi-tree Genetic Programming.

__init__(generations: int = 10, population_size: int = 100, crossover_rate: float = 0.8, mutation_rate: float = 0.2, elitism: float = 0.1, max_depth: int = 6, random_state: int = 42)

fit(X: ndarray, y: ndarray)

predict(X: ndarray) → ndarray

predict_proba(X: ndarray) → ndarray

set_score_request(*, sample_weight: bool | None | str = '$UNCHANGED$') → GP

Configure whether metadata should be requested to be passed to the score method.

Note that this method is only relevant when this estimator is used as a sub-estimator within a meta-estimator and metadata routing is enabled with enable_metadata_routing=True (see sklearn.set_config()). Please check the User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to score if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to score.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Parameters:

sample_weight (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_weight parameter in score.

Returns:

self – The updated object.

Return type:

object

Functions

fishy.models.evolutionary.gp.compileMultiTree(expr: List[deap.gp.PrimitiveTree], pset: deap.gp.PrimitiveSetTyped, X: ndarray) → ndarray

Compiles a multi-tree individual into a feature matrix.

fishy.models.evolutionary.gp.normalized_distances(X: ndarray, y: ndarray) → Tuple[float, float]

Computes normalized means of intraclass and interclass distances.

fishy.models.evolutionary.gp.quick_evaluate(expr: deap.gp.PrimitiveTree, pset: deap.gp.PrimitiveSetTyped, data: ndarray, prefix: str = 'ARG') → ndarray

Optimized stack-based evaluation of GP trees.

fishy.models.evolutionary.gp.staticLimit(key: Any, max_value: int) → Any

Depth limit decorator for multi-tree genetic operators.

fishy.models.evolutionary.gp.wrapper_fitness(X: ndarray, y: ndarray) → float

Calculates weighted fitness based on accuracy and distance metrics.

fishy.models.evolutionary.gp.xmate(ind1: List[deap.gp.PrimitiveTree], ind2: List[deap.gp.PrimitiveTree]) → Tuple[List[deap.gp.PrimitiveTree], List[deap.gp.PrimitiveTree]]

Reproduction operator for multi-tree GP individuals.

fishy.models.evolutionary.gp.xmut(individual: List[deap.gp.PrimitiveTree], expr: Any, pset: deap.gp.PrimitiveSetTyped) → Tuple[List[deap.gp.PrimitiveTree]]

Mutation operator for multi-tree GP individuals.

s