tract-linalg

Crates.iotract-linalg
lib.rstract-linalg
version0.21.5
sourcesrc
created_at2019-01-16 15:41:28.476283
updated_at2024-05-11 20:52:06.74735
descriptionTiny, no-nonsense, self contained, TensorFlow and ONNX inference
homepage
repositoryhttps://github.com/snipsco/tract
max_upload_size
id108938
size1,140,736
Mathieu Poumeyrol (kali)

documentation

README

tract-linalg

linalg stands for "linear algebra". This is a misnamer. This crates contains low-level, architecture dependant optimisations used by tract-core.

Functions

  • MatMatMul: Extended matrix*matrix product:
    • inspired by Gotoblass and BLIS micro kernel approach
    • extended for convolution friendly addressing (fused img2col)
    • fused output pipeline (min, max, and a few more simple, fast ops)
    • f32*f32 -> f32 (à la sgemm)
    • i8*i8 -> i32 accumulator -> i32 storage
    • i8*i8 -> i32 accumulator -> i8 (with channel zeropoint and scale, and re-quantization pipeline)
  • f32 sigmoid and f32 tanh: at f32 precision, by a rationale function (no exponentiation)
  • byte-to-byte lookup table

Implementations

generic fallback armv6, vfp armv7 neon armv8 simd x64 FMA
MatMatMul f32 4x4 8x4 8x8 16x6
MatMatMul i8->i8 8x4 8x8
MatMatMul i8->i32 8x8
sigmoid f32 4n 4n
tanh f32 4n 4n
byte lookup
Commit count: 5821

cargo fmt