josiann.parallel.ParallelArgument

class josiann.parallel.ParallelArgument(positions, nb_evaluations, args=_Nothing.NOTHING)

Object passed to parallel cost functions which holds instructions on what should be computed.

Parameters:

• positions (ndarray[Any, dtype[Union[float64, int64]]]) – matrix of position vectors at current iteration

• nb_evaluations (ndarray[Any, dtype[int64]]) – array indicating the number of evaluations to compute per position vector

• args (tuple[ndarray[Any, dtype[Any]], ...] (default: NOTHING)) – parallel arguments

Method generated by attrs for class ParallelArgument.

Attributes

ParallelArgument.positions	matrix of position vectors at current iteration
ParallelArgument.nb_evaluations	array indicating the number of evaluations to compute per position vector
ParallelArgument.args	parallel arguments
ParallelArgument.result	Array of costs of length equal to the total number of function evaluations required at current iteration.
ParallelArgument.where_evaluations	Convenience attribute for getting formatted tuples of (position vector, parallel arguments ...) repeated as many times as required by nb_evaluations.

Methods

ParallelArgument.__init__	Method generated by attrs for class ParallelArgument.
ParallelArgument.result_iter	rtype: Generator[ndarray[Any, dtype[float64]], None, None]

Attributes

ParallelArgument.positions: ndarray[Any, dtype[Union[float64, int64]]]

matrix of position vectors at current iteration

ParallelArgument.nb_evaluations: ndarray[Any, dtype[int64]]

array indicating the number of evaluations to compute per position vector

ParallelArgument.args: tuple[ndarray[Any, dtype[Any]], ...]

parallel arguments

ParallelArgument.result

Array of costs of length equal to the total number of function evaluations required at current iteration.

ParallelArgument.where_evaluations

Convenience attribute for getting formatted tuples of (position vector, parallel arguments …) repeated as many times as required by nb_evaluations.

Methods

ParallelArgument.__init__(positions, nb_evaluations, args=_Nothing.NOTHING)

Method generated by attrs for class ParallelArgument.

Parameters:

• positions (ndarray[Any, dtype[float64 | int64]]) –

• nb_evaluations (ndarray[Any, dtype[int64]]) –

• args (tuple[numpy.ndarray[Any, numpy.dtype[Any]], ...]) –

Return type:

None

ParallelArgument.result_iter()

Return type:

Generator[ndarray[Any, dtype[float64]], None, None]