numr 0.5.2

High-performance numerical computing with multi-backend GPU acceleration (CPU/CUDA/WebGPU)
Documentation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
//! Linear algebra CUDA kernel launchers
//!
//! This module re-exports all linalg launcher functions from the split submodules
//! in `linalg_launchers/`. The actual implementations are organized by category:
//!
//! - `basic` - trace, diag, diagflat, identity, copy, transpose
//! - `solvers` - forward/backward substitution, determinant, permutation
//! - `decompositions` - LU, Cholesky, QR
//! - `svd` - SVD Jacobi algorithm
//! - `eigen` - eigendecomposition (symmetric, general, Schur)
//! - `advanced` - rsf2csf, QZ decomposition

// Re-export all launcher functions for backward compatibility
pub use super::linalg_launchers::*;