Struct GenLinearModel 

pub struct GenLinearModel<C: Criterion> { /* private fields */ }

The Generalized Linear Model

The model is generic over a Criterion which specifies the distribution family and the link function.

Implementations

impl<C: Criterion> GenLinearModel<C>

pub fn new(criterion: C) -> GenLinearModel<C>

Constructs a new Generalized Linear Model.

Takes a Criterion which fully specifies the family and the link function used by the GLM.

use rusty_machine::learning::glm::GenLinearModel;
use rusty_machine::learning::glm::Bernoulli;

let glm = GenLinearModel::new(Bernoulli);

Trait Implementations

impl<C: Debug + Criterion> Debug for GenLinearModel<C>

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

Formats the value using the given formatter.

impl<C: Criterion> SupModel<Matrix<f64>, Vector<f64>> for GenLinearModel<C>

Supervised model trait for the GLM.

Predictions are made from the model by computing g^-1(Xb).

The model is trained using Iteratively Re-weighted Least Squares.

fn predict(&self, inputs: &Matrix<f64>) -> LearningResult<Vector<f64>>

Predict output from inputs.

fn train( &mut self, inputs: &Matrix<f64>, targets: &Vector<f64>, ) -> LearningResult<()>

Train the model using inputs and targets.