Crate rten_vecmath

Expand description

rten-vecmath provides portable SIMD types for implementing vectorized functions that work across different architectures.

§Portable SIMD types

The simd_vec module contains types and traits to support writing portable SIMD code, that works on stable Rust.

SIMD wrappers are provided for the following architectures:

There is also a scalar fallback that works on all platforms, but provides no performance benefit over non-SIMD code.

This crate contains SIMD-vectorized implementations of various math functions that are commonly used in neural networks.

For each function in this library there are multiple variants, which typically include:

A version that operates on scalars
A version that reads values from an input slice and writes to the corresponding position in an equal-length output slice. These have a vec_ prefix.
A version that reads values from a mutable input slice and writes the computed values back in-place. These have a vec_ prefix and _in_place suffix.

All variants use the same underlying implementation and should have the same accuracy.

See the source code for comments on accuracy.

erf
Computes the error function.
exp
Computes e^val. Functionally equivalent to f32::exp.
sigmoid
Computes the sigmoid function, aka. the standard logistic function, 1. / (1. + (-x).exp()).
tanh
vec_erf
Vectorized error function.
vec_erf_in_place
Variant of vec_erf that modifies elements in-place.
vec_exp
Vectorized exponential function.
vec_exp_in_place
Variant of vec_exp that modifies elements in-place.
vec_sigmoid
Vectorized sigmoid function.
vec_sigmoid_in_place
Variant of vec_sigmoid that modifies elements in-place.
vec_softmax
Computes the softmax function over a slice of floats.
vec_softmax_in_place
Computes the softmax function over a slice of floats.
vec_tanh
Vectorized tanh implementation.
vec_tanh_in_place