Model¶

class pymoo.core.algorithm.Algorithm(termination=None, output=None, display=None, callback=None, archive=None, return_least_infeasible=False, save_history=False, verbose=False, seed=None, evaluator=None, **kwargs)¶

Attributes

n_gen

Methods

advance
ask
finalize
has_next
infill
next
result
run
setup
tell

class pymoo.core.sampling.Sampling¶

This abstract class represents any sampling strategy that can be used to create an initial population or an initial search point.

Methods

do(problem, n_samples, **kwargs)

Sample new points with problem information if necessary.

do(problem, n_samples, **kwargs)¶

Sample new points with problem information if necessary.

Parameters

problemProblem: The problem to which points should be sampled. (lower and upper bounds, discrete, binary, …)
n_samplesint: Number of samples

Returns

popPopulation: The output population after sampling

class pymoo.core.selection.Selection(**kwargs)¶

This class is used to select parents for the mating or other evolutionary operators. Several strategies can be used to increase the selection pressure.

Methods

do(problem, pop, n_select, n_parents[, to_pop])

Choose from the population new individuals to be selected.

do(problem, pop, n_select, n_parents, to_pop=True, **kwargs)¶

Choose from the population new individuals to be selected.

Parameters

problem: class: The problem to be solved. Provides information such as lower and upper bounds or feasibility conditions for custom crossovers.
popPopulation: The population which should be selected from. Some criteria from the design or objective space might be used for the selection. Therefore, only the number of individual might be not enough.
n_selectint: Number of individuals to select.
n_parentsint: Number of parents needed to create an offspring.
to_popbool: Whether IF(!) the implementation returns only indices, it should be converted to individuals.

Returns

parentslist: List of parents to be used in the crossover

class pymoo.core.mutation.Mutation(prob=1.0, prob_var=None, **kwargs)¶

Methods

do
get_prob_var

class pymoo.core.crossover.Crossover(n_parents, n_offsprings, prob=0.9, **kwargs)¶

Methods

do

class pymoo.core.survival.Survival(filter_infeasible=True)¶

Methods

do

class pymoo.core.termination.Termination¶

Methods

update(algorithm)

Provide the termination criterion a current status of the algorithm to update the perc.

do_continue
has_terminated
terminate

update(algorithm)¶

Provide the termination criterion a current status of the algorithm to update the perc.

Parameters

algorithmobject: The algorithm object which is used to determine whether a run has terminated.

class pymoo.core.indicator.Indicator(**kwargs)¶

Methods

__call__(F, *args, **kwargs)

Call self as a function.

do

class pymoo.core.population.Population(individuals=[])¶

Methods

apply
collect
create
empty
get
has
merge
new
set

class pymoo.core.individual.Individual(config: Optional[dict] = None, **kwargs: Any)¶

Constructor for the Invididual class.

Parameters

configdict, None: A dictionary of configuration metadata. If None, use a class-dependent default configuration.
kwargsAny: Additional keyword arguments containing data which is to be stored in the Individual.

Attributes

CV: Get the constraint violation vector for an individual by either reading it from the cache or calculating it.
F: Get the objective function vector for an individual.
FEAS: Get whether an individual is feasible for each constraint.
G: Get the inequality constraint vector for an individual.
H: Get the equality constraint vector for an individual.
X: Get the decision vector for an individual.
cv: Convenience property.
dF: Get the objective function vector first derivatives for an individual.
dG: Get the inequality constraint(s) first derivatives for an individual.
dH: Get the equality constraint(s) first derivatives for an individual.
ddF: Get the objective function vector second derivatives for an individual.
ddG: Get the inequality constraint(s) second derivatives for an individual.
ddH: Get the equality constraint(s) second derivatives for an individual.
f: Convenience property.
feas: Convenience property.
feasible: Deprecated.
x: Convenience property.

Methods

`copy`([other, deep])	Copy an individual.
`default_config`()	Get default constraint violation configuration settings.
`duplicate`(key, new_key)	Duplicate a key to a new key.
`get`(*keys)	Get the values for one or more keys for an individual.
`has`(key)	Determine whether an individual has a provided key or not.
`new`()	Create a new instance of this class.
`reset`([data])	Reset the value of objective(s), inequality constraint(s), equality constraint(s), their first and second derivatives, the constraint violation, and the metadata to empty values.
`set`(key, value)	Set an individual’s data or metadata based on a key and value.
`set_by_dict`(**kwargs)	Set an individual’s data or metadata using values in a dictionary.

property CV¶

Get the constraint violation vector for an individual by either reading it from the cache or calculating it.

Returns

outnp.ndarray: The constraint violation vector for an individual.

property F¶

Get the objective function vector for an individual.

Returns

outnp.ndarray: The objective function vector for the individual.

property FEAS¶

Get whether an individual is feasible for each constraint.

Returns

outnp.ndarray: An array containing whether each constraint is feasible for an individual.

property G¶

Get the inequality constraint vector for an individual.

Returns

outnp.ndarray: The inequality constraint vector for the individual.

property H¶

Get the equality constraint vector for an individual.

Returns

outnp.ndarray: The equality constraint vector for the individual.

property X¶

Get the decision vector for an individual.

Returns

outnp.ndarray: The decision variable for the individual.

copy(other: Optional[pymoo.core.individual.Individual] = None, deep: bool = True) → pymoo.core.individual.Individual ¶

Copy an individual.

Parameters

otherIndividual, None: The individual to copy. If None, assumed to be self.
deepbool: Whether to deep copy the individual.

Returns

outIndividual: A copy of the individual.

property cv¶

Convenience property. Get the first constraint violation value for an individual by either reading it from the cache or calculating it.

Returns

outfloat, None: The constraint violation vector for an individual.

property dF¶

Get the objective function vector first derivatives for an individual.

Returns

outnp.ndarray: The objective function vector first derivatives for the individual.

property dG¶

Get the inequality constraint(s) first derivatives for an individual.

Returns

outnp.ndarray: The inequality constraint(s) first derivatives for the individual.

property dH¶

Get the equality constraint(s) first derivatives for an individual.

Returns

outnp.ndarray: The equality constraint(s) first derivatives for the individual.

property ddF¶

Get the objective function vector second derivatives for an individual.

Returns

outnp.ndarray: The objective function vector second derivatives for the individual.

property ddG¶

Get the inequality constraint(s) second derivatives for an individual.

Returns

outnp.ndarray: The inequality constraint(s) second derivatives for the individual.

property ddH¶

Get the equality constraint(s) second derivatives for an individual.

Returns

outnp.ndarray: The equality constraint(s) second derivatives for the individual.

default_config() → dict ¶

Get default constraint violation configuration settings.

Returns

outdict: A dictionary of default constraint violation settings.

duplicate(key: str, new_key: str) → None ¶

Duplicate a key to a new key.

Parameters

keystr: Name of the key to duplicated.
new_keystr: Name of the key to which to duplicate the original key.

property f¶

Convenience property. Get the first objective function value for an individual.

Returns

outfloat: The first objective function value for the individual.

property feas¶

Convenience property. Get whether an individual is feasible for the first constraint.

Returns

outbool: Whether an individual is feasible for the first constraint.

property feasible¶

Deprecated. Get whether an individual is feasible for each constraint.

Returns

outnp.ndarray: An array containing whether each constraint is feasible for an individual.

get(*keys: str) → Union[tuple, object]¶

Get the values for one or more keys for an individual.

Parameters

keysstr: Keys for which to get values.

Returns

outtuple, object: If more than one key provided, return a tuple of retrieved values. If a single key provided, return the retrieved value.

has(key: str) → bool ¶

Determine whether an individual has a provided key or not.

Parameters

keystr: The key for which to test.

Returns

outbool: Whether the Individual has the provided key.

new() → pymoo.core.individual.Individual ¶

Create a new instance of this class.

Returns

outIndividual: A new instance of an Individual.

reset(data: bool = True) → None ¶

Reset the value of objective(s), inequality constraint(s), equality constraint(s), their first and second derivatives, the constraint violation, and the metadata to empty values.

Parameters

databool: Whether to reset metadata associated with the Individiual.

set(key: str, value: object) → pymoo.core.individual.Individual ¶

Set an individual’s data or metadata based on a key and value.

Parameters

keystr: Key of the data for which to set.
valueobject: Value of the data for which to set.

Returns

outIndividual: A reference to the Individual for which values were set.

set_by_dict(**kwargs: Any) → None ¶

Set an individual’s data or metadata using values in a dictionary.

Parameters

kwargsAny: Keyword arguments defining the data to set.

property x¶

Convenience property. Get the decision vector for an individual.

Returns

outnp.ndarray: The decision variable for the individual.

class pymoo.core.result.Result¶

The resulting object of an optimization run.

Attributes

cv
f
feas