Model

class pymoo.core.algorithm.Algorithm(termination: Termination | str | tuple[str, ...] | None = None, output: str | None = None, display: Display | None = None, callback: Callback | None = None, archive: Archive | None = None, return_least_infeasible: bool = False, save_history: bool = False, verbose: bool = False, seed: int | None = None, evaluator: Evaluator | None = None, **kwargs)

class pymoo.core.sampling.Sampling

Initialize a Sampling operator.

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

class pymoo.core.selection.Selection(**kwargs: Any)

Initialize selection operator.

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

Base class for mutation operators.

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

class pymoo.core.survival.Survival(filter_infeasible: bool = True)

Initialize survival operator.

Parameters:

filter_infeasible – Whether to separate feasible from infeasible solutions.

class pymoo.core.termination.Termination

Initialize termination criterion.

do_continue() → bool

Check whether optimization should continue.

has_terminated() → bool

Check whether termination criterion is satisfied.

terminate() → None

Force termination.

update(algorithm: object) → float

Update progress and check termination condition.

Parameters:

algorithm – Algorithm instance.

Returns:

Progress value in [0, 1].

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

class pymoo.core.population.Population(individuals: Individual | list[Individual] | None = None)

class pymoo.core.individual.Individual(config: dict | None = None, **kwargs: Any)

Constructor for the Individual class.

Parameters:

• config – A dictionary of configuration metadata. If None, uses a class-dependent default.

• kwargs – Additional keyword arguments containing data to store in the Individual.

property CV: ndarray

Get the constraint violation vector from cache or calculate it.

Returns:

The constraint violation vector for the individual.

Return type:

np.ndarray

property F: ndarray

Get the objective function vector for an individual.

Returns:

The objective function vector for the individual.

Return type:

np.ndarray

property FEAS: ndarray

Get whether an individual is feasible for each constraint.

Returns:

An array of feasibility flags for each constraint.

Return type:

np.ndarray

property G: ndarray

Get the inequality constraint vector for an individual.

Returns:

The inequality constraint vector for the individual.

Return type:

np.ndarray

property H: ndarray

Get the equality constraint vector for an individual.

Returns:

The equality constraint vector for the individual.

Return type:

np.ndarray

property X: ndarray

Get the decision vector for an individual.

Returns:

The decision variable for the individual.

Return type:

np.ndarray

copy(other: Individual | None = None, deep: bool = True) → Individual

Copy an individual.

Parameters:

• other – The individual to copy. If None, assumed to be self.

• deep – Whether to deep copy the individual.

Returns:

A copy of the individual.

Return type:

Individual

property cv: float | None

Get the first constraint violation value for an individual.

Returns:

The first constraint violation value for the individual.

Return type:

float or None

property dF: ndarray

Get the objective function first derivatives for an individual.

Returns:

The first derivatives of the objective function vector.

Return type:

np.ndarray

property dG: ndarray

Get the inequality constraint first derivatives for an individual.

Returns:

The first derivatives of the inequality constraints.

Return type:

np.ndarray

property dH: ndarray

Get the equality constraint first derivatives for an individual.

Returns:

The first derivatives of the equality constraints.

Return type:

np.ndarray

property ddF: ndarray

Get the objective function second derivatives for an individual.

Returns:

The second derivatives of the objective function vector.

Return type:

np.ndarray

property ddG: ndarray

Get the inequality constraint second derivatives for an individual.

Returns:

The second derivatives of the inequality constraints.

Return type:

np.ndarray

property ddH: ndarray

Get the equality constraint second derivatives for an individual.

Returns:

The second derivatives of the equality constraints.

Return type:

np.ndarray

default_config() → dict

Get default constraint violation configuration settings.

Returns:

A dictionary of default constraint violation settings.

Return type:

dict

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

Duplicate a key to a new key.

Parameters:

• key – Name of the key to duplicate.

• new_key – Name of the key to which to duplicate the original key.

property f: float

Get the first objective function value for an individual.

Returns:

The first objective function value for the individual.

Return type:

float

property feas: bool

Get whether an individual is feasible for the first constraint.

Returns:

Whether the individual is feasible for the first constraint.

Return type:

bool

property feasible: ndarray

Get whether an individual is feasible for each constraint.

Deprecated. Use FEAS instead.

Returns:

An array of feasibility flags for each constraint.

Return type:

np.ndarray

get(*keys: str) → tuple | object

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

Parameters:

keys – Keys for which to get values.

Returns:

If more than one key provided, returns a tuple of values.

If a single key provided, returns the retrieved value.

Return type:

tuple or object

has(key: str) → bool

Determine whether an individual has a provided key.

Parameters:

key – The key for which to test.

Returns:

Whether the Individual has the provided key.

Return type:

bool

new() → Individual

Create a new instance of this class.

Returns:

A new instance of an Individual.

Return type:

Individual

reset(data: bool = True) → None

Reset objectives, constraints, derivatives, and violation values.

Resets all objective values, inequality/equality constraints, their first and second derivatives, constraint violations, and metadata to empty values.

Parameters:

data – Whether to reset metadata associated with the Individual.

set(key: str, value: object) → Individual

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

Parameters:

• key – Key of the data for which to set.

• value – Value of the data for which to set.

Returns:

A reference to the Individual for which values were set.

Return type:

Individual

set_by_dict(**kwargs: Any) → None

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

Parameters:

kwargs – Keyword arguments defining the data to set.

property x: ndarray

Get the decision vector for an individual.

Returns:

The decision variable for the individual.

Return type:

np.ndarray

class pymoo.core.result.Result

The resulting object of an optimization run.