Struct PLSSVD 

pub struct PLSSVD<F> { /* private fields */ }

PLS via Singular Value Decomposition of the cross-covariance matrix.

This is the simplest PLS variant. It computes the weight matrices by taking the leading left and right singular vectors of X^T Y after optional centring and scaling.

Unlike PLSRegression, PLSSVD does not iterate; it is a single matrix decomposition. It cannot predict Y from X — use PLSRegression if you need a predict method.

Type Parameters

• F: The floating-point scalar type.

Examples

use ferrolearn_decomp::cross_decomposition::PLSSVD;
use ferrolearn_core::traits::{Fit, Transform};
use ndarray::array;

let x = array![[1.0, 2.0], [3.0, 4.0], [5.0, 6.0], [7.0, 8.0]];
let y = array![[1.0], [2.0], [3.0], [4.0]];
let svd = PLSSVD::<f64>::new(1);
let fitted = svd.fit(&x, &y).unwrap();
let scores = fitted.transform(&x).unwrap();
assert_eq!(scores.ncols(), 1);

Implementations

impl<F: Float + Send + Sync + 'static> PLSSVD<F>

pub fn new(n_components: usize) -> Self

Create a new PLSSVD that extracts n_components components.

pub fn with_scale(self, scale: bool) -> Self

Set whether to scale X and Y to unit variance (default: true).

pub fn n_components(&self) -> usize

Return the number of components.

Trait Implementations

impl<F: Clone> Clone for PLSSVD<F>

fn clone(&self) -> PLSSVD<F>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<F: Debug> Debug for PLSSVD<F>

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

Formats the value using the given formatter.

impl<F: Float + Send + Sync + 'static> Fit<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>, ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>> for PLSSVD<F>

fn fit( &self, x: &Array2<F>, y: &Array2<F>, ) -> Result<FittedPLSSVD<F>, FerroError>

Fit PLSSVD by computing the SVD of the cross-covariance matrix X^T Y.

Errors

• FerroError::InvalidParameter if n_components is zero or exceeds min(n_features_x, n_features_y).
• FerroError::InsufficientSamples if there are fewer than 2 samples.
• FerroError::ShapeMismatch if X and Y have different numbers of rows.

type Fitted = FittedPLSSVD<F>

The fitted model type returned by fit.

type Error = FerroError

The error type returned by fit.