Struct ParallelFeatureExtractor 

pub struct ParallelFeatureExtractor { /* private fields */ }

Parallel Feature Extractor for High-Performance Computing

This extractor provides parallel computation of various feature types across data chunks, optimized for large datasets and multi-core systems.

Features extracted include:

• Statistical features computed in parallel across columns and rows
• Distance matrix features with parallel distance computation
• Correlation features with parallel correlation computation
• Configurable chunk size and thread count for optimal performance

Parameters

• chunk_size - Size of data chunks for parallel processing
• num_threads - Number of threads to use (None for automatic)
• include_distance_matrix - Whether to compute distance matrix features
• include_statistical_features - Whether to compute statistical features
• include_correlation_features - Whether to compute correlation features
• distance_metric - Distance metric for distance matrix computation

Examples

let data = Array2::from_shape_vec((1000, 50), (0..50000).map(|x| x as f64).collect())?;

let extractor = ParallelFeatureExtractor::new()
    .chunk_size(100)
    .num_threads(4)
    .include_statistical_features(true)
    .include_distance_matrix(true);

let features = extractor.extract_features(&data.view())?;
assert!(features.len() > 0);

Implementations

impl ParallelFeatureExtractor

pub fn new() -> Self

Create a new parallel feature extractor

pub fn chunk_size(self, size: usize) -> Self

Set the chunk size for parallel processing

pub fn num_threads(self, threads: usize) -> Self

Set the number of threads to use

pub fn include_distance_matrix(self, include: bool) -> Self

Set whether to compute distance matrix features

pub fn include_statistical_features(self, include: bool) -> Self

Set whether to compute statistical features

pub fn include_correlation_features(self, include: bool) -> Self

Set whether to compute correlation features

pub fn distance_metric(self, metric: &str) -> Self

Set the distance metric for distance matrix computation

pub fn extract_features( &self, data: &ArrayView2<'_, Float>, ) -> SklResult<Array1<Float>>

Extract features using parallel processing

Trait Implementations

impl Clone for ParallelFeatureExtractor

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ParallelFeatureExtractor

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

Formats the value using the given formatter.

impl Default for ParallelFeatureExtractor

fn default() -> Self

Returns the “default value” for a type.