adding implicit gemm v4r3

This commit is contained in:
Chao Liu
2019-07-15 17:42:18 -05:00
parent 61faf02b52
commit 5f82fdd9d3
5 changed files with 632 additions and 3 deletions

View File

@@ -15,6 +15,7 @@
#include "device_convolution_implicit_gemm_v3_nchw_cyxk_nkhw.hpp"
#include "device_convolution_implicit_gemm_v4r1_nchw_kcyx_nkhw.hpp"
#include "device_convolution_implicit_gemm_v4r2_nchw_kcyx_nkhw.hpp"
#include "device_convolution_implicit_gemm_v4r3_nchw_kcyx_nkhw.hpp"
struct GeneratorTensor_1
{
@@ -537,7 +538,7 @@ int main(int argc, char* argv[])
ConvStrides{},
ConvDilations{},
nrepeat);
#elif 1
#elif 0
device_convolution_implicit_gemm_v4r2_nchw_kcyx_nkhw(in_nchw_desc,
in_nchw,
wei_kcyx_desc,
@@ -547,6 +548,16 @@ int main(int argc, char* argv[])
ConvStrides{},
ConvDilations{},
nrepeat);
#elif 1
device_convolution_implicit_gemm_v4r3_nchw_kcyx_nkhw(in_nchw_desc,
in_nchw,
wei_kcyx_desc,
wei_kcyx,
out_nkhw_desc,
out_nkhw_device,
ConvStrides{},
ConvDilations{},
nrepeat);
#elif 0
device_implicit_gemm_convolution_1_chwn_cyxk_khwn_padded(in_nchw_desc,
in_nchw,