Struct LogisticRegression

pub struct LogisticRegression {
    pub coefficients: Vec<f64>,
    pub fit_intercept: bool,
    pub learning_rate: f64,
    pub max_iterations: usize,
    pub tolerance: f64,
    pub lambda: f64,
}

A simple logistic regression model with L2 regularization (optional). It uses batch gradient descent for training.

Fields

coefficients: Vec<f64>

Coefficients, including intercept as the first element: coefficients[0] = intercept coefficients[1..] = feature weights

fit_intercept: bool

learning_rate: f64

max_iterations: usize

tolerance: f64

lambda: f64

L2 regularization parameter (0.0 = no regularization).

Implementations

impl LogisticRegression

pub fn new( fit_intercept: bool, learning_rate: f64, max_iterations: usize, tolerance: f64, lambda: f64, ) -> Self

Creates a new LogisticRegression model with default settings.

• fit_intercept: whether to fit an intercept term
• learning_rate: step size for gradient descent
• max_iterations: maximum number of gradient descent steps
• tolerance: stopping criterion on coefficient updates
• lambda: L2 regularization strength (0.0 = no regularization)

pub fn fit(&mut self, features: &[Vec<f64>], labels: &[f64])

Fit the logistic regression model on the given data and binary labels (0 or 1).

Arguments

• features: NxD data, N samples, D features each
• labels: Nx1 binary labels (0.0 or 1.0)

Panics

• If features is empty or labels length does not match features length.
• If any feature row has different length from the others.
• If labels are not 0.0 or 1.0 (though minor floating tolerances are allowed).

pub fn predict_proba_one(&self, features: &[f64]) -> f64

Predict the probability of label=1 for a single feature vector.

Panics

• If features dimension doesn’t match the trained model.

pub fn predict_proba_batch(&self, features: &[Vec<f64>]) -> Vec<f64>

Predict probabilities for multiple rows of features.

pub fn predict_one(&self, features: &[f64]) -> f64

Predict a binary label (0 or 1) for a single feature vector, using threshold=0.5.

pub fn predict_batch(&self, features: &[Vec<f64>]) -> Vec<f64>

Predict binary labels for multiple rows.

Trait Implementations

impl Clone for LogisticRegression

fn clone(&self) -> LogisticRegression

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for LogisticRegression

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

Formats the value using the given formatter.