easy_vic_build.tools.calibrate_func.algorithm_NSGAII

NSGA-II base implementation for calibration workflows.

This module defines NSGAII_Base, a reusable DEAP-based NSGA-II runner with checkpoint persistence, generation history tracking, and optional Pareto front plotting.

Classes

NSGAII_Base([algParams, save_path])

A class that implements the NSGA-II (Non-dominated Sorting Genetic Algorithm II) for multi-objective optimization.

class easy_vic_build.tools.calibrate_func.algorithm_NSGAII.NSGAII_Base(algParams={'cxProb': 0.7, 'maxGen': 250, 'mutateProb': 0.2, 'popSize': 40}, save_path='checkpoint.pkl')[source]

Bases: object

A class that implements the NSGA-II (Non-dominated Sorting Genetic Algorithm II) for multi-objective optimization. This class provides methods for setting up the genetic algorithm, evaluating individuals, applying genetic operators (crossover, mutation), selecting the next generation, and saving/loading the algorithm state.

popSize

The population size for each generation.

Type:: int

maxGen

The maximum number of generations to run the algorithm.

Type:: int

save_path

The path to save and load the algorithm state.

Type:: str

history

A history of the population and non-dominated fronts at each generation.

Type:: list

current_generation

The current generation number.

Type:: int

initial_population

The initial population before any genetic operators are applied.

Type:: list

population

The current population of individuals.

Type:: list

__init__(algParams, save_path):: Initializes the NSGA-II algorithm with the given parameters and attempts to load the state.

get_obs():: Placeholder function to get the observed values (to be designed for specific use cases).

get_sim():: Placeholder function to get the simulated values (to be designed for specific use cases).

set_algorithm_params(popSize, maxGen, cxProb, mutateProb):: Sets the parameters for the genetic algorithm (population size, max generations, crossover probability, and mutation probability).

createFitness():: Creates the fitness function for individuals.

createInd():: Creates an individual representation (a list).

samplingInd():: Samples a new individual by generating random values for its elements.

registerInd():: Registers the individual sampling function with the toolbox.

registerPop():: Registers the population initialization function with the toolbox.

evaluate(ind):: A placeholder function for evaluating an individual’s fitness (to be customized for specific use cases).

registerEvaluate():: Registers the evaluation function with the toolbox.

evaluatePop(population):: Evaluates the fitness of the entire population.

operatorMate(parent1, parent2):: Defines the crossover operation for mating two individuals (using a two-point crossover).

operatorMutate(ind):: Defines the mutation operation for an individual (using a bit-flip mutation).

operatorSelect(population):: Defines the selection operation (using tournament selection).

registerOperators():: Registers the genetic operators (mate, mutate, and select) with the toolbox.

apply_genetic_operators(offspring):: Applies the genetic operators (crossover and mutation) to the offspring.

select_next_generation(combined):: Selects the next generation by sorting individuals based on Pareto dominance and applying crowding distance.

print_results(population):: Prints the results of the best individual from the final population.

load_state():: Loads the algorithm state from the specified save path (if a saved state exists).

save_state():: Saves the current algorithm state (current generation, population, and history) to the specified save path.

run():: Runs the NSGA-II algorithm for the specified number of generations, applying genetic operators and selecting the next generation.

Initializes the NSGA-II algorithm with the given parameters.

Parameters:

algParams (dict) – Dictionary containing the algorithm parameters: - popSize: The population size for each generation. - maxGen: The maximum number of generations to run the algorithm. - cxProb: The crossover probability. - mutateProb: The mutation probability.
save_path (str, optional) – The path to save and load the algorithm state (default is “checkpoint.pkl”).

__init__(algParams={'cxProb': 0.7, 'maxGen': 250, 'mutateProb': 0.2, 'popSize': 40}, save_path='checkpoint.pkl')[source]

Initializes the NSGA-II algorithm with the given parameters.

Parameters:

algParams (dict) – Dictionary containing the algorithm parameters: - popSize: The population size for each generation. - maxGen: The maximum number of generations to run the algorithm. - cxProb: The crossover probability. - mutateProb: The mutation probability.
save_path (str, optional) – The path to save and load the algorithm state (default is “checkpoint.pkl”).

set_algorithm_params(popSize=None, maxGen=None, cxProb=None, mutateProb=None, **kwargs)[source]

Sets the parameters for the genetic algorithm.

Parameters:

popSize (int, optional) – The population size for each generation (default is 40).
maxGen (int, optional) – The maximum number of generations to run the algorithm (default is 250).
cxProb (float, optional) – The crossover probability (default is 0.7).
mutateProb (float, optional) – The mutation probability (default is 0.2).

get_obs()[source]

Placeholder function to get observed values (to be designed for specific use cases).

Return type:: None

get_sim()[source]

Placeholder function to get simulated values (to be designed for specific use cases).

Return type:: None

createFitness()[source]: Creates the fitness function for individuals.

createInd()[source]: Creates an individual representation (a list).

samplingInd()[source]

Samples a new individual by generating random values for its elements.

Returns:: A new individual sampled with random elements.
Return type:: Individual

evaluate(ind)[source]

A placeholder function for evaluating an individual’s fitness.

Parameters:: ind (Individual) – The individual to evaluate.
Returns:: A tuple containing the fitness values.
Return type:: tuple

static operatorMate(parent1, parent2)[source]

Defines the crossover operation for mating two individuals.

Parameters:

parent1 (Individual) – The first parent individual.
parent2 (Individual) – The second parent individual.

Returns:

A tuple containing the offspring resulting from the crossover.

Return type:

tuple

static operatorMutate(ind)[source]

Defines the mutation operation for an individual.

Parameters:: ind (Individual) – The individual to mutate.
Returns:: A tuple containing the mutated individual.
Return type:: tuple

static operatorSelect(population)[source]

Defines the selection operation for choosing individuals from the population.

Parameters:: population (list of Individual) – The population from which to select individuals.
Returns:: A list of selected individuals.
Return type:: list

registerInd()[source]: Registers the individual sampling function with the toolbox.

registerPop()[source]: Registers the population initialization function with the toolbox.

registerEvaluate()[source]: Registers the evaluation function with the toolbox.

registerOperators()[source]: Registers the genetic operators (mate, mutate, and select) with the toolbox.

evaluatePop(population)[source]

Evaluates the fitness of the entire population.

Parameters:: population (list of Individual) – The population to evaluate.

apply_genetic_operators(offspring)[source]

Applies the genetic operators (crossover and mutation) to the offspring.

Parameters:: offspring (list of Individual) – The offspring to apply the genetic operators to.

select_next_generation(combined)[source]

Selects the next generation from the combined population and offspring.

Parameters:: combined (list of Individual) – The combined population of parents and offspring.
Returns:: The selected next generation.
Return type:: list

load_state()[source]: Loads the algorithm state from the specified save path if a saved state exists.

save_state()[source]: Saves the current algorithm state (current generation, population, and history) to the specified save path.

print_results(population)[source]

Prints the results of the best individual from the final population.

Parameters:: population (list of Individual) – The final population.

plot_front_pairwise(population, front, gen, names_plot=None, plot_dir='pareto_progress', transform_func=None)[source]

Plot pairwise objective scatter for full population and first front.

Parameters:

population (list) – Population to visualize (typically parent + offspring).
front (list) – First non-dominated front.
gen (int) – Generation index used in figure title.
names_plot (list of str, optional) – Objective axis labels. If None, labels are generated as obj1, obj2, …
plot_dir (str, optional) – Directory used to store figure output.
transform_func (callable, optional) – Optional transformation applied to objective arrays before plotting.

run(plot_progress=False, plot_dir='pareto_progress', names_plot=None, transform_func=None)[source]

Runs the NSGA-II algorithm for the specified number of generations.

This method applies genetic operators, evaluates individuals, selects the next generation, and stores the results.

Parameters:

plot_progress (bool, optional) – Whether to save a pairwise Pareto plot each generation.
plot_dir (str, optional) – Output directory for progress plots.
names_plot (list of str, optional) – Objective labels passed to plot_front_pairwise().
transform_func (callable, optional) – Optional transform function passed to plot_front_pairwise().

Returns:

The final population after all generations.

Return type:

list