Enum Link 

pub enum Link {
    Log,
    Identity,
}

The link function g of a Generalized Linear Model, mapping the mean mu to the linear predictor eta = g(mu) (and back via the inverse link h, mu = h(eta)).

Mirrors the link carried by sklearn’s loss classes (sklearn/linear_model/_glm/glm.py:119-131): HalfPoissonLoss, HalfGammaLoss and HalfTweedieLoss use the log link (y_pred = exp(X @ coef + intercept)); HalfSquaredError and HalfTweedieLossIdentity use the identity link (y_pred = X @ coef + intercept).

Variants

Log

Log link: g(mu) = ln(mu), inverse h(eta) = exp(eta).

Used by Poisson, Gamma and Tweedie-with-power > 0 losses.

Identity

Identity link: g(mu) = mu, inverse h(eta) = eta.

Used by the Normal/least-squares loss and Tweedie-with-power <= 0.

Trait Implementations

impl Clone for Link

fn clone(&self) -> Link

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Copy for Link

impl Debug for Link

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Eq for Link

impl PartialEq for Link

fn eq(&self, other: &Link) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StructuralPartialEq for Link