Struct PCA 

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

Principal Component Analysis configuration.

Holds the n_components hyperparameter. Calling Fit::fit centres the data, computes the eigendecomposition of the covariance matrix, and returns a FittedPCA that can project new data.

Implementations

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

pub fn new(n_components: usize) -> Self

Create a new PCA that retains n_components principal components.

Panics

Does not panic. Validation of n_components against the data dimensions happens during Fit::fit.

pub fn n_components(&self) -> usize

Return the number of components this PCA is configured to retain.

Trait Implementations

impl<F: Clone> Clone for PCA<F>

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

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<F: Debug> Debug for PCA<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]>>, ()> for PCA<F>

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

Fit PCA by centring the data and eigendecomposing the covariance matrix.

Errors

• FerroError::InvalidParameter if n_components is zero or exceeds the number of features.
• FerroError::InsufficientSamples if there are fewer than 2 samples.
• FerroError::ConvergenceFailure if the Jacobi eigendecomposition does not converge.

type Fitted = FittedPCA<F>

The fitted model type returned by fit.

type Error = FerroError

The error type returned by fit.

impl<F: Float + Send + Sync + 'static> PipelineTransformer<F> for PCA<F>

fn fit_pipeline( &self, x: &Array2<F>, _y: &Array1<F>, ) -> Result<Box<dyn FittedPipelineTransformer<F>>, FerroError>

Fit PCA using the pipeline interface.

The y argument is ignored; PCA is unsupervised.

Errors

Propagates errors from Fit::fit.