Certara.RDarwin (version 1.1.1)

pyDarwinOptionsGA: Create options for the Genetic Algorithm (GA) in pyDarwin.

Description

This function allows you to set various options specific to the Genetic Algorithm (GA) in pyDarwin.

Usage

pyDarwinOptionsGA(
  elitist_num = 2,
  crossover_rate = 0.95,
  mutation_rate = 0.95,
  sharing_alpha = 0.1,
  selection = "tournament",
  selection_size = 2,
  crossover_operator = "cxOnePoint",
  mutate = "flipBit",
  attribute_mutation_probability = 0.1,
  niche_penalty = 20
)

Value

An object of class "pyDarwinOptionsGA" containing the specified GA options.

Arguments

elitist_num

A positive integer specifying the number of best models from any generation to carry over, unchanged, to the next generation. Functions like the Hall of Fame in DEAP. Default: 2

crossover_rate

A real value (between 0.0 and 1.0) specifying the fraction of mating pairs that will undergo crossover. Default: 0.95

mutation_rate

A real value (between 0.0 and 1.0) specifying the probability that at least one bit in the genome will be “flipped”, 0 to 1, or 1 to 0. Default: 0.95

sharing_alpha

A real value specifying the parameter of the niche penalty calculation. Default: 0.1

selection

A string specifying the selection algorithm for the GA. Currently, only "tournament" is available. Default: "tournament"

selection_size

A positive integer specifying the number of “parents” to enter in the selection. 2 is highly recommended, experience with other values is very limited. Default: 2

crossover_operator

A string specifying the algorithm for crossover. Only "cxOnePoint" (single-point crossover) is available. Default: "cxOnePoint"

mutate

A string specifying the algorithm for mutation. Currently, only "flipBit" is available. Default: "flipBit"

attribute_mutation_probability

A real value specifying the probability of any bit being mutated (real value between 0.0 and 1.0). Default: 0.1

niche_penalty

A positive real value used for the calculation of the crowding penalty. The niche penalty is calculated by first finding the “distance matrix”, the pair-wise Mikowski distance from the present model to all other models. The “crowding” quantity is then calculated as the sum of: (distance/niche_radius)^sharing_alpha for all other models in the generation for which the Mikowski distance is less than the niche radius. Finally, the penalty is calculated as: exp((crowding-1)*niche_penalty)-1. The objective of using a niche penalty is to maintain diversity of models, to avoid premature convergence of the search by penalizing when models are too similar to other models in the current generation. Default: 20

Examples

# Create GA options with default values
options <- pyDarwinOptionsGA()

# Create GA options with custom values
options <-
  pyDarwinOptionsGA(elitist_num = 4,
                    crossover_rate = 0.9,
                    mutation_rate = 0.8,
                    sharing_alpha = 0.2)