Struct RandomProjectionFeatures 

pub struct RandomProjectionFeatures { /* private fields */ }

Random Projection Feature Extractor

Performs dimensionality reduction using random projections while preserving approximate distances between data points (Johnson-Lindenstrauss lemma). Useful for high-dimensional data preprocessing and approximate similarity search.

Parameters

• n_components - Number of dimensions in the projected space
• random_state - Random seed for reproducibility
• density - Density of random projection matrix (for sparse projections)
• eps - Distortion parameter for auto-sizing components

Examples

use sklears_feature_extraction::streaming_projection::RandomProjectionFeatures;
use scirs2_core::ndarray::Array2;

let X = Array2::from_elem((100, 1000), 1.0);
let mut rp = RandomProjectionFeatures::new()
    .n_components(50)
    .random_state(42);

let X_projected = rp.fit_transform(&X.view())?;
assert_eq!(X_projected.ncols(), 50);

Implementations

impl RandomProjectionFeatures

pub fn new() -> Self

Create a new random projection feature extractor

pub fn n_components(self, n_components: usize) -> Self

Set number of components in projected space

pub fn random_state(self, random_state: u64) -> Self

Set random state for reproducibility

pub fn density(self, density: Float) -> Self

Set density for sparse random projections

pub fn eps(self, eps: Float) -> Self

Set distortion parameter for auto-sizing

pub fn fit(&mut self, X: &ArrayView2<'_, Float>) -> SklResult<()>

Fit the random projection to data

pub fn transform(&self, X: &ArrayView2<'_, Float>) -> SklResult<Array2<Float>>

Transform data using fitted random projection

pub fn fit_transform( &mut self, X: &ArrayView2<'_, Float>, ) -> SklResult<Array2<Float>>

Fit and transform data in one step

Trait Implementations

impl Clone for RandomProjectionFeatures

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 RandomProjectionFeatures

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

Formats the value using the given formatter.

impl Default for RandomProjectionFeatures

fn default() -> Self

Returns the “default value” for a type.