Struct ImportanceSampler 

pub struct ImportanceSampler {
    pub num_samples: usize,
    pub seed: u64,
}

Importance sampling based on sample scores.

Samples examples with probability proportional to their importance scores. Useful for focusing on informative examples.

Fields

num_samples: usize

Number of samples to draw

seed: u64

Random seed for reproducibility

Implementations

impl ImportanceSampler

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

Create a new importance sampler.

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

Sample indices based on importance scores.

Arguments

• scores - Importance scores for each sample (higher = more important)

Returns

Sampled indices

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

Sample with replacement allowed.

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

Sample without replacement (unique indices).

Trait Implementations

impl Clone for ImportanceSampler

fn clone(&self) -> ImportanceSampler

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ImportanceSampler

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

Formats the value using the given formatter.