Function svd 

pub fn svd<F>(
    a: &ArrayView2<'_, F>,
    fullmatrices: bool,
) -> LinalgResult<SVDDecomposition<F>>
where
    F: Float + NumAssign + ScalarOperand + Sum + Send + Sync + 'static,

Performs singular value decomposition.

Arguments

• a - Input matrix
• full_matrices - Whether to return full U and V^T matrices

Returns

• SVD decomposition result

Examples

use scirs2_core::ndarray::{array, ScalarOperand};
use scirs2_linalg::lapack::svd;

let a = array![[1.0, 2.0], [3.0, 4.0], [5.0, 6.0]];
let svd_result = svd(&a.view(), false).unwrap();

// Check that A = U*diag(S)*V^T
// (implementation dependent, so not shown here)