Ordinal

OrdinalParameter (BaseParameter)

Ordinal parameter that can take discrete values of any type.

Source code in parameterspace/parameters/ordinal.py

class OrdinalParameter(BaseParameter):
    """Ordinal parameter that can take discrete values of any type."""

    @store_init_arguments
    def __init__(
        self,
        name: str,
        values: list,
        *,
        prior: Union[list, np.ndarray, None] = None,
        transformation: Optional[BaseTransformation] = None,
        inactive_numerical_value: Optional[float] = np.nan
    ):
        """
        Initialize with options for ordinal parameter.

        Args:
            name: Name of the parameter.
            values: Allowed values in ascending order for this parameter.
            prior: List of (unnormalized) probablities for each value.
                Default puts equal probablity for each value.
            transformation: The only supported transformation right now
                is [parameterspace.transformations.categorical.Cat2Num][],
                which is also the default.
            inactive_numerical_value: Placeholder value for this parameter
                in case it is not active. Default is NaN.
        """
        self.values = values
        transformation = Cat2Num(values) if transformation is None else transformation
        prior = Uniform() if prior is None else prior
        super().__init__(
            name,
            prior,
            transformation,
            is_continuous=False,
            is_ordered=True,
            num_values=len(values),
            inactive_numerical_value=inactive_numerical_value,
        )

    def check_value(self, value):
        """Check if value is valid."""
        return value in self.values

__init__(self, name, values, *, prior=None, transformation=None, inactive_numerical_value=nan) special

Initialize with options for ordinal parameter.

Parameters:

Name	Type	Description	Default
name	str	Name of the parameter.	required
values	list	Allowed values in ascending order for this parameter.	required
prior	Union[list, numpy.ndarray]	List of (unnormalized) probablities for each value. Default puts equal probablity for each value.	None
transformation	Optional[parameterspace.transformations.base.BaseTransformation]	The only supported transformation right now is parameterspace.transformations.categorical.Cat2Num, which is also the default.	None
inactive_numerical_value	Optional[float]	Placeholder value for this parameter in case it is not active. Default is NaN.	nan

Source code in parameterspace/parameters/ordinal.py

@store_init_arguments
def __init__(
    self,
    name: str,
    values: list,
    *,
    prior: Union[list, np.ndarray, None] = None,
    transformation: Optional[BaseTransformation] = None,
    inactive_numerical_value: Optional[float] = np.nan
):
    """
    Initialize with options for ordinal parameter.

    Args:
        name: Name of the parameter.
        values: Allowed values in ascending order for this parameter.
        prior: List of (unnormalized) probablities for each value.
            Default puts equal probablity for each value.
        transformation: The only supported transformation right now
            is [parameterspace.transformations.categorical.Cat2Num][],
            which is also the default.
        inactive_numerical_value: Placeholder value for this parameter
            in case it is not active. Default is NaN.
    """
    self.values = values
    transformation = Cat2Num(values) if transformation is None else transformation
    prior = Uniform() if prior is None else prior
    super().__init__(
        name,
        prior,
        transformation,
        is_continuous=False,
        is_ordered=True,
        num_values=len(values),
        inactive_numerical_value=inactive_numerical_value,
    )

check_value(self, value)

Check if value is valid.

Source code in parameterspace/parameters/ordinal.py

def check_value(self, value):
    """Check if value is valid."""
    return value in self.values