Struct nalgebra_lapack::SVD

pub struct SVD<N: Scalar, R: DimMin<C>, C: Dim> where
    DefaultAllocator: Allocator<N, R, R> + Allocator<N, DimMinimum<R, C>> + Allocator<N, C, C>,  {
    pub u: MatrixN<N, R>,
    pub vt: MatrixN<N, C>,
    pub singular_values: VectorN<N, DimMinimum<R, C>>,
}

The SVD decomposition of a general matrix.

Fields

u: MatrixN<N, R>

The left-singular vectors U of this SVD.

vt: MatrixN<N, C>

The right-singular vectors V^t of this SVD.

singular_values: VectorN<N, DimMinimum<R, C>>

The singular values of this SVD.

Methods

impl<N: SVDScalar<R, C>, R: DimMin<C>, C: Dim> SVD<N, R, C> where
    DefaultAllocator: Allocator<N, R, R> + Allocator<N, R, C> + Allocator<N, DimMinimum<R, C>> + Allocator<N, C, C>, 

pub fn new(m: MatrixMN<N, R, C>) -> Option<Self>

Computes the Singular Value Decomposition of matrix.

impl<R: DimMin<C>, C: Dim> SVD<f32, R, C> where
    DefaultAllocator: Allocator<f32, R, C> + Allocator<f32, C, R> + Allocator<f32, U1, R> + Allocator<f32, U1, C> + Allocator<f32, R, R> + Allocator<f32, DimMinimum<R, C>> + Allocator<f32, DimMinimum<R, C>, R> + Allocator<f32, DimMinimum<R, C>, C> + Allocator<f32, R, DimMinimum<R, C>> + Allocator<f32, C, C>, 

pub fn recompose(self) -> MatrixMN<f32, R, C>

Reconstructs the matrix from its decomposition.

Useful if some components (e.g. some singular values) of this decomposition have been manually changed by the user.

pub fn pseudo_inverse(&self, epsilon: f32) -> MatrixMN<f32, C, R>

Computes the pseudo-inverse of the decomposed matrix.

All singular value below epsilon will be set to zero instead of being inverted.

pub fn rank(&self, epsilon: f32) -> usize

The rank of the decomposed matrix.

This is the number of singular values that are not too small (i.e. greater than the given epsilon).

impl<R: DimMin<C>, C: Dim> SVD<f64, R, C> where
    DefaultAllocator: Allocator<f64, R, C> + Allocator<f64, C, R> + Allocator<f64, U1, R> + Allocator<f64, U1, C> + Allocator<f64, R, R> + Allocator<f64, DimMinimum<R, C>> + Allocator<f64, DimMinimum<R, C>, R> + Allocator<f64, DimMinimum<R, C>, C> + Allocator<f64, R, DimMinimum<R, C>> + Allocator<f64, C, C>, 

pub fn recompose(self) -> MatrixMN<f64, R, C>

Reconstructs the matrix from its decomposition.

Useful if some components (e.g. some singular values) of this decomposition have been manually changed by the user.

pub fn pseudo_inverse(&self, epsilon: f64) -> MatrixMN<f64, C, R>

Computes the pseudo-inverse of the decomposed matrix.

All singular value below epsilon will be set to zero instead of being inverted.

pub fn rank(&self, epsilon: f64) -> usize

The rank of the decomposed matrix.

This is the number of singular values that are not too small (i.e. greater than the given epsilon).

Trait Implementations

impl<N: Clone + Scalar, R: Clone + DimMin<C>, C: Clone + Dim> Clone for SVD<N, R, C> where
    DefaultAllocator: Allocator<N, R, R> + Allocator<N, DimMinimum<R, C>> + Allocator<N, C, C>, 

fn clone(&self) -> SVD<N, R, C>

Returns a copy of the value.

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

Performs copy-assignment from source.

impl<N: Debug + Scalar, R: Debug + DimMin<C>, C: Debug + Dim> Debug for SVD<N, R, C> where
    DefaultAllocator: Allocator<N, R, R> + Allocator<N, DimMinimum<R, C>> + Allocator<N, C, C>, 

fn fmt(&self, __arg_0: &mut Formatter) -> Result

Formats the value using the given formatter.

impl<N: Scalar, R: DimMin<C>, C: Dim> Copy for SVD<N, R, C> where
    DefaultAllocator: Allocator<N, C, C> + Allocator<N, R, R> + Allocator<N, DimMinimum<R, C>>,
    MatrixMN<N, R, R>: Copy,
    MatrixMN<N, C, C>: Copy,
    VectorN<N, DimMinimum<R, C>>: Copy,