Struct ClassBalancedSampler 

pub struct ClassBalancedSampler {
    pub samples_per_class: usize,
    pub seed: u64,
}

Class-balanced sampling for imbalanced datasets.

Ensures equal representation of all classes during training.

Fields

samples_per_class: usize

Number of samples per class

seed: u64

Random seed

Implementations

impl ClassBalancedSampler

pub fn new(samples_per_class: usize, seed: u64) -> Self

Create a new class-balanced sampler.

pub fn sample(&self, labels: &Array1<f64>) -> TrainResult<Vec<usize>>

Sample balanced batches from each class.

Arguments

• labels - Class labels

Returns

Sampled indices

pub fn compute_class_weights( &self, labels: &Array1<f64>, ) -> TrainResult<HashMap<i32, f64>>

Compute class weights for weighted sampling.

Trait Implementations

impl Clone for ClassBalancedSampler

fn clone(&self) -> ClassBalancedSampler

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ClassBalancedSampler

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

Formats the value using the given formatter.