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
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72

//! Operation traits for tensor operations.
//!
//! This module contains trait definitions for various tensor operations.
//! Implementations are in the backend-specific modules (cpu/, cuda/, wgpu/).

mod activation;
mod advanced_random;
mod binary;
mod compare;
mod complex;
mod conditional;
mod conv;
mod cumulative;
mod distance;
mod einsum;
mod fp8_matmul;
mod gemm_epilogue;
mod indexing;
mod kernel;
mod linalg;
mod logical;
mod matmul;
pub mod multivariate;
mod normalization;
mod quasirandom;
mod random;
mod reduce;
mod scalar;
mod semiring_matmul;
mod shape;
mod sorting;
#[cfg(feature = "sparse")]
mod sparse_24;
mod statistics;
mod tensor_ops;
mod type_conversion;
mod unary;
mod utility;

pub use activation::ActivationOps;
pub use advanced_random::AdvancedRandomOps;
pub use binary::BinaryOps;
pub use compare::CompareOps;
pub use complex::ComplexOps;
pub use conditional::ConditionalOps;
pub use conv::{ConvOps, PaddingMode};
pub use cumulative::CumulativeOps;
pub use distance::{DistanceMetric, DistanceOps};
pub use einsum::EinsumOps;
pub use fp8_matmul::Fp8MatmulOps;
pub use gemm_epilogue::{GemmActivation, GemmEpilogueOps};
pub use indexing::{IndexingOps, ScatterReduceOp};
pub use kernel::Kernel;
pub use linalg::LinalgOps;
pub use logical::LogicalOps;
pub use matmul::MatmulOps;
pub use multivariate::MultivariateRandomOps;
pub use normalization::NormalizationOps;
pub use quasirandom::QuasiRandomOps;
pub use random::RandomOps;
pub use reduce::ReduceOps;
pub use scalar::ScalarOps;
pub use semiring_matmul::SemiringMatmulOps;
pub use shape::ShapeOps;
pub use sorting::SortingOps;
#[cfg(feature = "sparse")]
pub use sparse_24::Sparse24Ops;
pub use statistics::StatisticalOps;
pub use tensor_ops::TensorOps;
pub use type_conversion::TypeConversionOps;
pub use unary::UnaryOps;
pub use utility::{MeshgridIndexing, UtilityOps};