Struct SparseTensorDecomposition 

pub struct SparseTensorDecomposition<State = Untrained> {
    pub n_factors: usize,
    pub max_iter: usize,
    pub tol: Float,
    pub sparsity_penalty: Float,
    pub regularization: Float,
    pub sparsity_threshold: Float,
    /* private fields */
}

Sparse Tensor Decomposition

Decomposes a sparse tensor using CP decomposition with sparsity constraints. Handles tensors with many zero entries efficiently and can enforce sparsity in the factor matrices through L1 regularization.

Examples

use scirs2_core::ndarray::Array3;
use sklears_cross_decomposition::SparseTensorDecomposition;
use sklears_core::traits::Fit;

let tensor = Array3::zeros((20, 15, 10));
let sparse_decomp = SparseTensorDecomposition::new(5).sparsity_penalty(0.1);
let fitted = sparse_decomp.fit(&tensor, &()).unwrap();

Fields

n_factors: usize

Number of factors

max_iter: usize

Maximum number of iterations

tol: Float

Convergence tolerance

sparsity_penalty: Float

L1 sparsity penalty

regularization: Float

L2 regularization

sparsity_threshold: Float

Sparsity threshold (values below this are set to zero)

Implementations

impl SparseTensorDecomposition<Untrained>

pub fn new(n_factors: usize) -> Self

Create a new sparse tensor decomposition

pub fn sparsity_penalty(self, penalty: Float) -> Self

Set sparsity penalty (L1 regularization)

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

Set L2 regularization

pub fn sparsity_threshold(self, threshold: Float) -> Self

Set sparsity threshold

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

Set maximum iterations

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

Set tolerance

impl SparseTensorDecomposition<Trained>

pub fn factor_matrices(&self) -> &Vec<Array2<Float>>

Get the factor matrices

pub fn sparsity_levels(&self) -> &Array1<Float>

Get the sparsity levels for each mode

pub fn reconstruction_error(&self) -> Float

Get the reconstruction error

pub fn n_iter(&self) -> usize

Get the number of iterations

Trait Implementations

impl<State: Clone> Clone for SparseTensorDecomposition<State>

fn clone(&self) -> SparseTensorDecomposition<State>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<State: Debug> Debug for SparseTensorDecomposition<State>

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

Formats the value using the given formatter.

impl Estimator for SparseTensorDecomposition<Untrained>

type Config = ()

Configuration type for the estimator

type Error = SklearsError

Error type for the estimator

type Float = f64

The numeric type used by this estimator

fn config(&self) -> &Self::Config

Get estimator configuration

fn validate_config(&self) -> Result<(), SklearsError>

Validate estimator configuration with detailed error context

fn check_compatibility( &self, n_samples: usize, n_features: usize, ) -> Result<(), SklearsError>

Check if estimator is compatible with given data dimensions

fn metadata(&self) -> EstimatorMetadata

Get estimator metadata

impl Fit<ArrayBase<OwnedRepr<f64>, Dim<[usize; 3]>>, ()> for SparseTensorDecomposition<Untrained>

type Fitted = SparseTensorDecomposition<Trained>

The fitted model type

fn fit(self, tensor: &Array3<Float>, _target: &()) -> Result<Self::Fitted>

Fit the model to the provided data with validation

fn fit_with_validation( self, x: &X, y: &Y, _x_val: Option<&X>, _y_val: Option<&Y>, ) -> Result<(Self::Fitted, FitMetrics), SklearsError>

where Self: Sized,

Fit with custom validation and early stopping

impl Transform<ArrayBase<OwnedRepr<f64>, Dim<[usize; 3]>>> for SparseTensorDecomposition<Trained>

fn transform(&self, tensor: &Array3<Float>) -> Result<Array3<Float>>

Reconstruct tensor using sparse factors