Struct LLE 

pub struct LLE { /* private fields */ }

Locally Linear Embedding configuration.

Holds hyperparameters for the LLE algorithm. Call Fit::fit to compute the embedding and obtain a FittedLLE.

Implementations

impl LLE

pub fn new(n_components: usize) -> Self

Create a new LLE with n_components embedding dimensions.

Defaults: n_neighbors = 5, reg = 1e-3.

pub fn with_n_neighbors(self, k: usize) -> Self

Set the number of nearest neighbors.

pub fn with_reg(self, reg: f64) -> Self

Set the regularization parameter.

pub fn n_components(&self) -> usize

Return the configured number of components.

pub fn n_neighbors(&self) -> usize

Return the configured number of neighbors.

pub fn reg(&self) -> f64

Return the configured regularization parameter.

Trait Implementations

impl Clone for LLE

fn clone(&self) -> LLE

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for LLE

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

Formats the value using the given formatter.

impl Fit<ArrayBase<OwnedRepr<f64>, Dim<[usize; 2]>>, ()> for LLE

fn fit(&self, x: &Array2<f64>, _y: &()) -> Result<FittedLLE, FerroError>

Fit LLE by computing reconstruction weights and finding the bottom eigenvectors of (I - W)^T (I - W).

Errors

• FerroError::InvalidParameter if n_components is zero, n_neighbors is zero, or parameters are out of range.
• FerroError::InsufficientSamples if there are fewer than n_neighbors + 1 samples.
• FerroError::NumericalInstability if a local covariance matrix is singular.

type Fitted = FittedLLE

The fitted model type returned by fit.

type Error = FerroError

The error type returned by fit.