Struct MultinomialLogistic 

pub struct MultinomialLogistic {
    pub classes: Vec<usize>,
    /* private fields */
}

A fitted softmax classifier. weights is n_classes × k row-major (k = p + intercept).

Fields

classes: Vec<usize>

Implementations

impl MultinomialLogistic

pub fn fit( x: &[f64], y: &[usize], n: usize, p: usize, fit_intercept: bool, lr: f64, l2: f64, max_iter: usize, ) -> Result<Self, LearningError>

Fit by gradient ascent. lr learning rate, l2 ridge penalty (≥ 0), max_iter iterations. Fails closed on shape mismatch / a single class.

pub fn predict_proba_row(&self, x_row: &[f64]) -> Vec<f64>

Class probabilities for one row, aligned with classes.

pub fn predict_row(&self, x_row: &[f64]) -> usize

Predicted class label (argmax probability).

Trait Implementations

impl Clone for MultinomialLogistic

fn clone(&self) -> MultinomialLogistic

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for MultinomialLogistic

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

Formats the value using the given formatter.