leavelet/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 25 Packages Projects Releases Wiki Activity

CUTLASS 3.1 (#915)

Browse Source 
Co-authored-by: Aniket Shivam <ashivam@nvidia.com>
This commit is contained in:
ANIKET SHIVAM
2023-04-14 23:19:34 -04:00
committed by GitHub
co-authored by Aniket Shivam
parent 9b8166e3f0
commit d572cc1aab
482 changed files with 37175 additions and 16410 deletions
Download Patch File Download Diff File Expand all files Collapse all files
test/unit/conv/device/conv2d_testbed_interleaved.h
+2 -1
View File
@@ -410,6 +410,7 @@ public:
LayoutC,
ElementCompute,
ElementAccumulator,
ElementC,
cutlass::NumericConverterClamp<ElementC, ElementCompute>
>(
kConvolutionalOperator,
@@ -517,7 +518,7 @@ public:
/////////////////////////////////////////////////////////////////////////////////////////////////////////
// TestAllConv: Runs cutlass::conv::device::ImplicitGemmConvolution operator and compares it with reference
// TestAllConv runs conv operator on default conv problem sizes from test::conv::device::TestbedConv2dProblemSizes
// Additionally, each conv2d test can provide conv problem sizes (conv_test_sizes) and blacklist of sizes
// Additionally, each conv2d test can provide conv problem sizes (conv_test_sizes) and blacklist of sizes
// (conv_blacklist_sizes)
/////////////////////////////////////////////////////////////////////////////////////////////////////////////
template <typename ImplicitGemm, int InterleavedK>