onnxruntime/onnxruntime/core/mlas
Tracy Sharpe 79e27d937a
MLAS: add sgemm weight prepacking (#5183)
Add support to MLAS to prepack weights for the float GEMM. Support for prepacking has been added to MatMul and Attention for this release.
2020-09-16 08:36:27 -07:00
..
inc MLAS: add sgemm weight prepacking (#5183) 2020-09-16 08:36:27 -07:00
lib MLAS: add sgemm weight prepacking (#5183) 2020-09-16 08:36:27 -07:00
README.md Initial bootstrap commit. 2018-11-19 16:48:22 -08:00

About MLAS

MLAS is a compute library containing processor optimized GEMM kernels and platform specific threading code.

Unit tests for MLAS

Unit tests for the SGEMM kernels are available under onnxruntime\test\mlas. These tests run over a range of inputs that then execute the various special cases for aligned and unaligned outputs. The tests have failed if any "mismatch" strings are printed.