Add support for mixed 4-bit/8-bit data types GEMM (#1413)

* Add support for mixed 4-bit/8-bit data types GEMM

* fix ( and )

---------

Co-authored-by: Aleksandar Samardžić <asamardzic@matf.bg.ac.rs>
Co-authored-by: Haicheng Wu <haichengw@nvidia.com>
This commit is contained in:
Aleksandar Samardžić
2024-08-30 05:11:06 +02:00
committed by GitHub
parent f7b19de32c
commit e1976daacc
15 changed files with 960 additions and 14 deletions

View File

@@ -57,6 +57,7 @@ void initialize_gemm_reference_operations_fp8in_fp32out(Manifest &manifest);
void initialize_gemm_reference_operations_fp32out(Manifest &manifest);
void initialize_gemm_reference_operations_fp_other(Manifest &manifest);
void initialize_gemm_reference_operations_fp_mixed_input(Manifest &manifest);
void initialize_gemm_reference_operations_int_mixed_input(Manifest &manifest);
void initialize_conv2d_reference_operations(Manifest &manifest);
void initialize_conv3d_reference_operations(Manifest &manifest);
@@ -85,6 +86,8 @@ void initialize_reference_operations(Manifest &manifest) {
initialize_gemm_reference_operations_fp_other(manifest);
initialize_gemm_reference_operations_fp_mixed_input(manifest);
initialize_gemm_reference_operations_int_mixed_input(manifest);
}
///////////////////////////////////////////////////////////////////////////////////////////////////