models.log_score()

Mean log score -(1/n) sum log f(y_i) of the outcomes under the

Usage

models.log_score(
    family, y, mu, dispersion=1.0, weights=None, theta=None, power=None
)

family’s predictive distribution; lower is better.

Parameters

family: str

y: list of float

mu: list of float

dispersion: float = 1.0

weights: list of float = None

theta: float = None

power: float = None

Returns

float

Examples

>>> from prospicio.models import log_score
>>> round(log_score("poisson", [0.0], [1.0]), 12)

1.0