mkadavil d37c91dffa Quantization (scale + zero point) support for BF16 LPGEMM api. ...

-Quantization of f32 to bf16 (bf16 = (f32 * scale_factor) + zero_point)
instead of just type conversion in aocl_gemm_bf16bf16f32obf16.
-Support for multiple scale/sum/matrix_add/bias post-ops in a single
LPGEMM api call.
-Post-ops mask related fixes in lpgemv kernels .
-Additional scale post-ops sanity checks.

AMD-Internal: [SWLCSG-2945]
Change-Id: I3b35cc413c176bb50bfdbd6acd4839a5ba7e94bb

2024-07-18 05:32:51 -04:00

..

aocl_gemm

Quantization (scale + zero point) support for BF16 LPGEMM api.

2024-07-18 05:32:51 -04:00

gemmd

Code cleanup: spelling corrections

2023-11-09 00:16:30 -05:00

CMakeLists.txt

CMake: Enable builds for both static and shared builds for Linux.

2024-03-14 10:32:51 -04:00