deephyper.skopt.utils.point_aslist#

deephyper.skopt.utils.point_aslist(search_space, point_as_dict)[source]#

Convert a dictionary representation of a point from a search space to the list representation. The list of values is created from the values of the dictionary, sorted by the names of dimensions used as keys.

See also

deephyper.skopt.utils.point_asdict

Parameters:

  • search_space (dict) – Represents search space. The keys are dimension names (strings) and values are instances of classes that inherit from the class deephyper.skopt.space.Dimension (Real, Integer or Categorical)

  • point_as_dict (dict) – dict with parameter names as keys to which corresponding parameter values are assigned.

Returns:

point_as_list – list with point values.The order of parameters in the list is given by sorted(params_space.keys()).

Return type:

list

Examples

>>> from deephyper.skopt.space.space import Real, Integer
>>> from deephyper.skopt.utils import point_aslist
>>> search_space = {'name1': Real(0,1),
...                 'name2': Integer(2,4), 'name3': Real(-1,1)}
>>> point_as_dict = {'name1': 0.66, 'name2': 3, 'name3': -0.15}
>>> point_aslist(search_space, point_as_dict)
[0.66, 3, -0.15]