ribs.emitters.opt.PyCMAEvolutionStrategy¶

class ribs.emitters.opt.PyCMAEvolutionStrategy(sigma0: float | floating, solution_dim: int | integer, batch_size: int | None = None, seed: int | integer | None = None, dtype: numpy.typing.DTypeLike = numpy.float64, lower_bounds: float | floating | ndarray | None = None, upper_bounds: float | floating | ndarray | None = None, opts: dict | None = None)[source]¶

Wrapper around the pycma CMAEvolutionStrategy.

Note

This optimizer requires the pycma package, which can be installed with pip install cma or conda install cma.

Parameters:¶

sigma0: float | floating¶

Initial step size.

batch_size: int | None = None¶

Number of solutions to evaluate at a time. This is passed directly as popsize in opts.

solution_dim: int | integer¶

Size of the solution space.

seed: int | integer | None = None¶

Seed for the random number generator.

dtype: numpy.typing.DTypeLike = numpy.float64¶

Data type of solutions.

lower_bounds: float | floating | ndarray | None = None¶

scalar or (solution_dim,) array indicating lower bounds of the solution space. Scalars specify the same bound for the entire space, while arrays specify a bound for each dimension. Pass -np.inf in the array or scalar to indicated unbounded space.

upper_bounds: float | floating | ndarray | None = None¶

Same as above, but for upper bounds (and pass np.inf instead of -np.inf).

opts: dict | None = None¶

Additional options for pycma. Note that popsize, bounds, randn, and seed are overwritten by us and thus should not be provided in this dict. We also make verbose default to -9, but you can also pass in a custom value here.

Methods

ask([batch_size])	Samples new solutions from the Gaussian distribution.
check_stop(ranking_values)	Checks if the optimization should stop and be reset.
reset(x0)	Resets the optimizer to start at x0.
tell(ranking_indices, ranking_values, ...)	Passes the solutions back to the ES.

Attributes

batch_size

Number of solutions per iteration.

ask(batch_size: int | integer | None = None) → ndarray[source]¶

Samples new solutions from the Gaussian distribution.

Parameters:¶

batch_size: int | integer | None = None¶

batch size of the sample. Defaults to self.batch_size.

check_stop(ranking_values: ndarray) → bool[source]¶

Checks if the optimization should stop and be reset.

Parameters:¶

ranking_values: ndarray¶

Not used.

Returns:¶

Output of cma.CMAEvolutionStrategy.stop

reset(x0: ndarray) → None[source]¶

Resets the optimizer to start at x0.

Parameters:¶

x0: ndarray¶

Initial mean.

tell(ranking_indices: ndarray, ranking_values: ndarray, num_parents: int | integer) → None[source]¶

Passes the solutions back to the ES.

Parameters:¶

ranking_indices: ndarray¶

Integer indices that are used to rank the solutions returned in ask(). Note that these are NOT the ranks of the solutions. Rather, they are indices such that solutions[ranking_indices] will correctly rank the solutions (think of an argsort).

ranking_values: ndarray¶

Array of values that were used to rank the solutions. Shape can be either (batch_size,) or (batch_size, n_values)``, where batch_size is the number of solutions and n_values is the number of values that the ranker used. Note that we assume a descending sort, i.e., higher values should come first.

num_parents: int | integer¶

Number of top solutions to select from the ranked solutions.

property batch_size : int | integer¶

Number of solutions per iteration.

Only valid after a call to reset().