bluepyopt.deapext.optimisations

Optimisation class

class bluepyopt.deapext.optimisations.DEAPOptimisation(evaluator=None, use_scoop=False, seed=1, offspring_size=10, eta=10, mutpb=1.0, cxpb=1.0, map_function=None, hof=None, selector_name=None)[source]

DEAP Optimisation class

Constructor

Parameters
  • evaluator (Evaluator) – Evaluator object

  • seed (float) – Random number generator seed

  • offspring_size (int) – Number of offspring individuals in each generation

  • eta (float) – Parameter that controls how far the crossover and mutation operator disturbe the original individuals

  • mutpb (float) – Mutation probability

  • cxpb (float) – Crossover probability

  • map_function (function) – Function used to map (parallelise) the evaluation function calls

  • hof (hof) – Hall of Fame object

  • selector_name (str) – The selector used in the evolutionary algorithm, possible values are ‘IBEA’ or ‘NSGA2’

run(max_ngen=10, offspring_size=None, continue_cp=False, cp_filename=None, cp_frequency=1, parent_population=None)[source]

Run optimisation

setup_deap()[source]

Set up optimisation

class bluepyopt.deapext.optimisations.IBEADEAPOptimisation(*args, **kwargs)[source]

IBEA DEAP class

Constructor

class bluepyopt.deapext.optimisations.WSListIndividual(*args, **kwargs)[source]

Individual consisting of list with weighted sum field

Constructor

class bluepyopt.deapext.optimisations.WeightedSumFitness(values=(), obj_size=None)[source]

Fitness that compares by weighted sum

property sum

Weighted sum of values

property weighted_sum

Weighted sum of wvalues