Created on 21 okt. 2012
Helper module with functions used by the model ensemble when perfomring
an optimization.
Code author: jhkwakkel <j.h.kwakkel (at) tudelft (dot) nl>
-
ema_optimization.mut_polynomial_bounded(individual, eta, policy_levers, keys, indpb)
Polynomial mutation as implemented in original NSGA-II algorithm in
C by Deb. Modified to cope with categories, next to continuous variables.
| Parameters: |
- individual – Individual to be mutated.
- eta – Crowding degree of the mutation. A high eta will produce
a mutant resembling its parent, while a small eta will
produce a solution much more different.
- policy_levers –
- keys –
|
|---|
| Returns: | A tuple of one individual.
|
|---|
-
class ema_optimization.NSGA2StatisticsCallback(weights=(), nr_of_generations=None, crossover_rate=None, mutation_rate=None, pop_size=None, caching=False)
Helper class for tracking statistics about the progression of the
optimization
-
ema_optimization.generate_individual_outcome(icls, attr_list, keys)
Helper function for generating an individual in case of outcome
optimization
| Parameters: |
- icls – class of the individual
- attr_list – list of initializers for each attribute
- keys – the name of each attribute
|
|---|
| Returns: | an instantiated individual
|
|---|
-
ema_optimization.generate_individual_robust(icls, attr_list, keys)
Helper function for generating an individual in case of robust optimization
| Parameters: |
- icls – class of the individual
- attr_list – list of initializers for each attribute
- keys – the name of each attribute
|
|---|
| Returns: | an instantiated individual
|
|---|
-
ema_optimization.evaluate_population_outcome(population, ri, toolbox, ensemble)
Helper function for evaluating a population in case of outcome optimization
| Parameters: |
- population – the population to evaluate
- ri – reporting interval
- toolbox – deap toolbox instance
- ensemble – the ensemble instance running the optimization
|
|---|
-
ema_optimization.evaluate_population_robust(population, ri, toolbox, ensemble, cases=None, **kwargs)
Helper function for evaluating a population in case of robust optimization
| Parameters: |
- population – the population to evaluate
- ri – reporinting interval
- toolbox – deap toolbox instance
- ensemble – the ensemble instance running the optimization
- cases – the cases to use in the robust optimization
|
|---|
-
ema_optimization.closest_multiple_of_four(number)
Helper function for transforming the population size to the closest
multiple of four. Is necessary because of implementation issues of the
NSGA2 algorithm in deap.
