Struct linfa_pls::PlsRegression

pub struct PlsRegression<F: Float>(_);

Implementations

impl<F: Float> PlsRegression<F>

pub fn params(n_components: usize) -> PlsRegressionParams<F>

pub fn weights(&self) -> (&Array2<F>, &Array2<F>)

Singular vectors of the cross-covariance matrices

pub fn loadings(&self) -> (&Array2<F>, &Array2<F>)

Loadings of records and targets

pub fn rotations(&self) -> (&Array2<F>, &Array2<F>)

Projection matrices used to transform records and targets

pub fn coefficients(&self) -> &Array2<F>

The coefficients of the linear model such that Y is approximated as Y = X.coefficients

pub fn inverse_transform(
    &self,
    dataset: DatasetBase<ArrayBase<impl Data<Elem = F>, Ix2>, ArrayBase<impl Data<Elem = F>, Ix2>>
) -> DatasetBase<Array2<F>, Array2<F>>

Transform the given dataset in the projected space back to the original space.

Trait Implementations

impl<F: Float, D: Data<Elem = F>> PredictInplace<ArrayBase<D, Dim<[usize; 2]>>, ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>> for PlsRegression<F>

fn predict_inplace<'a>(&'a self, x: &ArrayBase<D, Ix2>, y: &mut Array2<F>)

Given an input matrix X, with shape (n_samples, n_features), predict returns the target variable according to [<Pls $name>] method learned from the training data distribution.

fn default_target(&self, x: &ArrayBase<D, Ix2>) -> Array2<F>

Create targets that predict_inplace works with.

impl<F: Float, D: Data<Elem = F>> Transformer<DatasetBase<ArrayBase<D, Dim<[usize; 2]>>, ArrayBase<D, Dim<[usize; 2]>>>, DatasetBase<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>, ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>>> for PlsRegression<F>

fn transform(
    &self,
    dataset: DatasetBase<ArrayBase<D, Ix2>, ArrayBase<D, Ix2>>
) -> DatasetBase<Array2<F>, Array2<F>>

Apply dimension reduction to the given dataset