Skip to main content

Module sve

Module sve 

Source
Expand description

Singular value expansion (SVE) of the kernels, from sparse-ir-sve

Structs§

CentrosymmSVE
Centrosymmetric SVE computation
KernelBlock
A discretized, weighted kernel matrix A[i, j] = K(x_i, y_j) √w_i √w'_j (Gauss points x_i, y_j, weights w_i, w'_j) whose entries are evaluated on demand
NonCentrosymmSVE
Non-centrosymmetric SVE computation
SVEOptions
Options of compute_sve_with
SVEResult
Result of Singular Value Expansion computation
SamplingSVE
Sampling-based SVE computation

Enums§

Error
Error returned by the fallible public functions of this crate
ErrorKind
Category of an Error: what the caller can do about it
TworkType
Working precision type for SVE computations

Constants§

ACCURACY_CUTOFF_FACTOR
Cutoff of SVEOptions::cutoff_from_epsilon relative to the required accuracy ε

Traits§

SVEStrategy
Trait for SVE computation strategies

Functions§

compute_sve
Main SVE computation function
compute_sve_dense
compute_sve by the dense CPQR-based SVD of the fully formed blocks (SVEStrategy::matrices and compute_svd_dtensor), as before the ACA (and as compute_sve still does in f64): the reference the ACA path is tested against
compute_sve_with
The SVE of kernel with options
half_domain_segments
The half-domain segment boundaries the SVE of a centrosymmetric kernel discretizes
mirror_segments_to_full_domain
Mirror half-domain segment boundaries onto the full domain
sve_residual
The residual of the SVE sve of kernel, relative to the largest singular value: max |K(x, y) - Σ_l s_l u_l(x) v_l(y)| / s_0