dspk-ops 0.0.9

Signal-processing operations for dspk, written once in CubeCL: filters, resampling, linear algebra, statistics, peak finding, spatial operators
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

//! Linear algebra on the device: matrix products ([`matmul`](fn@matmul)), covariance, symmetric
//! eigendecomposition, Cholesky; and the models built on them: [`PcaModel`], [`TopComponents`]
//! (the leading principal components of many rows, streamed in batches), [`PpcaModel`]
//! (probabilistic PCA), [`FastIcaModel`] (independent components), and [`DeviceProjection`]
//! (applying a fitted projection to device buffers).

pub mod cholesky;
pub mod covariance;
pub mod eigen;
pub mod matmul;
pub mod projection;
pub mod pca;
pub mod subspace;
pub mod tridiagonal;
pub mod ppca;
pub mod ica;
pub mod kernels;

pub use cholesky::{cholesky, cholesky_solve, spd_inverse_logdet};
pub use covariance::{covariance, covariance_of_host, SecondMomentAccumulator};
pub use matmul::{matmul, MatrixView};
pub use eigen::{symmetric_eigen, symmetric_eigen_batched, symmetric_eigen_host, EigenOptions, SymmetricEigen};
pub use pca::PcaModel;
pub use tridiagonal::symmetric_eigen_cpu;
pub use subspace::{DeviceRows, HostRows, IndexedRows, RowSource, TopComponents, TopComponentsOptions};
pub use projection::DeviceProjection;
pub use ppca::PpcaModel;
pub use ica::{FastIcaModel, IcaContrast};