Update example/15_grouped_gemm/grouped_gemm_multi_abd_xdl_fixed_nk_bias_fp16.cpp

Co-authored-by: Bartłomiej Kocot <barkocot@amd.com>

Update example/15_grouped_gemm/grouped_gemm_multi_abd_xdl_fixed_nk_bias_fp16.cpp
Co-authored-by: Bartłomiej Kocot <barkocot@amd.com>
fa995966 · zjing14 · GitHub · dacef7b5 · fa995966
Unverified Commit fa995966 authored Nov 15, 2023 by zjing14 Committed by GitHub Nov 15, 2023
Hide whitespace changes
Inline Side-by-side

Showing with 4 additions and 1 deletion

example/15_grouped_gemm/grouped_gemm_multi_abd_xdl_fixed_nk_bias_fp16.cpp ...ed_gemm/grouped_gemm_multi_abd_xdl_fixed_nk_bias_fp16.cpp +4 -1

No files found.
--- a/example/15_grouped_gemm/grouped_gemm_multi_abd_xdl_fixed_nk_bias_fp16.cpp
+++ b/example/15_grouped_gemm/grouped_gemm_multi_abd_xdl_fixed_nk_bias_fp16.cpp
@@ -201,8 +201,11 @@ bool run_grouped_gemm(const ProblemSize& problem_size, const ExecutionConfig& co
        d0_tensors[i].GenerateTensorValue(GeneratorTensor_3<D0DataType>{-0.5, 0.5});
    }
+    constexpr index_t NumATensor = 1;
+    constexpr index_t NumBTensor = 1;
+    constexpr index_t NumDTensor = 0;
    using GroupedGemmKernelArgument =
-        ck::tensor_operation::device::GroupedGemmMultiABDKernelArgument<2, 1, 1>;
+        ck::tensor_operation::device::GroupedGemmMultiABDKernelArgument<NumATensor , NumBTensor, NumDTensor>;
    std::vector<GroupedGemmKernelArgument> grouped_gemm_kernel_args_;
    grouped_gemm_kernel_args_.reserve(group_count);