metaheuristic_designer.operators.factories package#

Submodules#

Module contents#

Operator factory package, provides registries and creation functions.

list_operators()[source]#

Return a list of all registered operator keys.

Each key is formatted as "registry.operator_name" and can be passed to create_operator().

Return type:

list[str]

Returns:
list of str

Fully qualified operator names.

add_operator_entry(operator_fn, operator_name, operator_registry='custom', preserves_order=False)[source]#

Register a new operator so it can be created by create_operator().

Parameters:
operator_fncallable

A callable that follows the operator signature expected by OperatorFromLambda. Usually wrapped with OperatorFnDef, OperatorRandomDef, etc.

operator_namestr

Key under which the operator is registered.

operator_registrystr, optional

Registry name (default "custom"). If the registry does not exist, it is created.

preserves_orderbool, optional

If True, the operator is marked as order-preserving, meaning individuals retain their position when applying it. Default False.

Parameters:

  • operator_fn (callable)

  • operator_name (str)

  • operator_registry (str)

create_crossover_operator(method, encoding=None, rng=None, name=None, **kwargs)[source]#

Create a crossover operator by name.

Return type:

OperatorFromLambda

Parameters:
methodstr

Key into crossover_ops_map (e.g., "one_point", "uniform").

encodingEncoding, optional

Encoding applied to the genotype after crossover.

rngRNGLike, optional

Random number generator.

namestr, optional

Display name; defaults to method.

**kwargs

Additional parameters forwarded to the operator function (e.g., k, crossover_prob, pairing_method).

Returns:
OperatorFromLambda

The wrapped crossover operator.

Parameters:

  • method (str)

  • encoding (Encoding | None)

  • rng (int | Generator | None)

  • name (str | None)

create_debug_operator(method, encoding=None, name=None, **kwargs)[source]#

Create a debug operator by name.

Return type:

OperatorFromLambda

Parameters:
methodstr

Key into debug_ops_map.

encodingEncoding, optional

Encoding applied to the genotype.

namestr, optional

Display name; defaults to method.

**kwargs

Forwarded to the operator function.

Returns:
OperatorFromLambda

The wrapped debug operator.

Parameters:

  • method (str)

  • encoding (Encoding | None)

  • name (str | None)

create_differential_evolution_operator(method, encoding=None, vectorized=True, name=None, **kwargs)[source]#

Create a DE operator by name.

Return type:

OperatorFromLambda

Parameters:
methodstr

DE variant string, e.g., "de/rand/1".

encodingEncoding, optional

Encoding applied to the genotype.

vectorizedbool, optional

Unused; kept for interface compatibility.

namestr, optional

Display name; defaults to method.

**kwargs

Forwarded to the DE operator function (e.g., F, Cr).

Returns:
OperatorFromLambda

The wrapped DE operator.

Parameters:

  • method (str)

  • encoding (Encoding | None)

  • vectorized (bool)

  • name (str | None)

create_mutation_operator(method, encoding=None, name=None, **kwargs)[source]#

Create a mutation operator by name.

Return type:

OperatorFromLambda

Parameters:
methodstr

Key into mutation_ops_map.

encodingEncoding, optional

Encoding applied to the genotype after mutation.

namestr, optional

Display name; defaults to method.

**kwargs

Parameters forwarded to the mutation function (e.g., N, F, distribution).

Returns:
OperatorFromLambda

The wrapped mutation operator.

Parameters:

  • method (str)

  • encoding (Encoding | None)

  • name (str | None)

create_operator(method, encoding=None, rng=None, name=None, **kwargs)[source]#

Create an operator by name from any registry.

The method string can be a simple key (e.g., "gauss") or dot-separated "registry.key" (e.g., "crossover.one_point").

Return type:

OperatorFromLambda

Parameters:
methodstr

Operator key, possibly with registry prefix.

encodingEncoding, optional

Encoding applied to the genotype after the operator runs.

rngRNGLike, optional

Random number generator.

namestr, optional

Display name; defaults to method.

**kwargs

Parameters forwarded to the underlying operator function.

Returns:
OperatorFromLambda

The wrapped operator.

Raises:
ValueError

If the operator cannot be found.

Parameters:

  • method (str)

  • encoding (Encoding | None)

  • rng (int | Generator | None)

  • name (str | None)

create_permutation_operator(method, encoding=None, name=None, **kwargs)[source]#

Create a permutation operator by name.

Return type:

OperatorFromLambda

Parameters:
methodstr

Key into permutation_ops_map.

encodingEncoding, optional

Encoding applied to the genotype.

namestr, optional

Display name; defaults to method.

**kwargs

Forwarded to the operator function.

Returns:
OperatorFromLambda

The wrapped permutation operator.

Parameters:

  • method (str)

  • encoding (Encoding | None)

  • name (str | None)

create_random_operator(method, initializer, encoding=None, name=None, **kwargs)[source]#

Create a random operator that uses an Initializer for fresh values.

Return type:

OperatorFromLambda

Parameters:
methodstr

Key into random_ops_map.

encodingEncoding, optional

Encoding applied to the genotype.

namestr, optional

Display name; defaults to method.

**kwargs

Forwarded to the operator function.

Returns:
OperatorFromLambda

The wrapped random operator.

Parameters:
create_swarm_operator(method, name=None, **kwargs)[source]#

Create a swarm operator by name.

Return type:

OperatorFromLambda

Parameters:
methodstr

Key into swarm_ops_map (e.g., "pso").

namestr, optional

Display name; defaults to method.

**kwargs

Forwarded to the operator function.

Returns:
OperatorFromLambda

The wrapped swarm operator.

Parameters:

  • method (str)

  • name (str | None)