leavelet/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 28 Packages Projects Releases Wiki Activity

Softmax (#546)

Browse Source 
* add test layernorm g-mem version

* Delete include/configure directory

* Delete examples/test_layernorm directory

* Update gemm_with_softmax.h

* Update gemm_softmax.cu

* Update linear_combination.h

* Update fast_math.h

* remove redundant vars

Co-authored-by: yujia.zhai <yujia.zhai@bytedance.com>
Co-authored-by: yuzhai <yuzhai@nvidia.com>
This commit is contained in:
Yujia Zhai
2022-07-02 01:19:18 -04:00
committed by GitHub
co-authored by yujia.zhai yuzhai
parent e45e773436
commit 04a9777b87
4 changed files with 428 additions and 376 deletions
Download Patch File Download Diff File Expand all files Collapse all files
include/cutlass/epilogue/thread/linear_combination.h
+1 -1
View File
@@ -71,7 +71,7 @@ public:
using ElementCompute = ElementCompute_;
static int const kCount = Count;
static const ScaleType::Kind kScale = Scale;
using FragmentOutput = Array<ElementOutput, kCount>;
using FragmentAccumulator = Array<ElementAccumulator, kCount>;
using ComputeFragment = Array<ElementCompute, kCount>;