opytimizer.optimizers.evolutionary.gp

Genetic Programming.

class opytimizer.optimizers.evolutionary.gp.GP(params: Optional[Dict[str, Any]] = None)

A GP class, inherited from Optimizer.

This is the designed class to define GP-related variables and methods.

References

  1. Koza. Genetic programming: On the programming of computers by means of natural selection (1992).
__init__(params: Optional[Dict[str, Any]] = None) → None

Initialization method.

Parameters:params – Contains key-value parameters to the meta-heuristics.
p_reproduction

Probability of reproduction.

p_mutation

Probability of mutation.

p_crossover

Probability of crossover.

prunning_ratio

Nodes’ prunning ratio.

_prune_nodes(n_nodes: int) → int

Prunes the amount of possible nodes used for mutation and crossover.

Parameters:n_nodes – Number of current nodes.
Returns:Amount of prunned nodes.
Return type:(int)
_reproduction(space: opytimizer.spaces.tree.TreeSpace) → None

Reproducts a number of individuals pre-selected through a tournament procedure (p. 99).

Parameters:space – A TreeSpace object.
_mutation(space: opytimizer.spaces.tree.TreeSpace) → None

Mutates a number of individuals pre-selected through a tournament procedure.

Parameters:space – A TreeSpace object.
_mutate(space: opytimizer.spaces.tree.TreeSpace, tree: opytimizer.core.node.Node, max_nodes: int) → opytimizer.core.node.Node

Actually performs the mutation on a single tree (p. 105).

Parameters:
  • space – A TreeSpace object.
  • tree – A Node instance to be mutated.
  • max_nodes – Maximum number of nodes to be searched.
Returns:

A mutated tree.

Return type:

(Node)

_crossover(space: opytimizer.spaces.tree.TreeSpace) → None

Crossover a number of individuals pre-selected through a tournament procedure (p. 101).

Parameters:space – A TreeSpace object.
_cross(father: opytimizer.core.node.Node, mother: opytimizer.core.node.Node, max_father: int, max_mother: int) → Tuple[opytimizer.core.node.Node, opytimizer.core.node.Node]

Actually performs the crossover over a father and mother nodes.

Parameters:
  • father – A father’s node to be crossed.
  • mother – A mother’s node to be crossed.
  • max_father – Maximum of nodes from father to be used.
  • max_mother – Maximum of nodes from mother to be used.
Returns:

Two offsprings based on the crossover operator.

Return type:

(Tuple[Node, Node])

evaluate(space: opytimizer.core.space.Space, function: opytimizer.core.function.Function) → None

Evaluates the search space according to the objective function.

Parameters:
  • space – A TreeSpace object.
  • function – A Function object that will be used as the objective function.
update(space: opytimizer.core.space.Space) → None

Wraps Genetic Programming over all trees and variables.

Parameters:space – TreeSpace containing agents and update-related information.