IntenseWebs/openvino
3
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[GPU] Optimize conv_b_fs_zyx_fsv16_imad (#18356)

Browse Source 
+ Make this kernel do not support asymmetric weight.
This commit is contained in:
Jade Cho
2023-07-06 16:28:40 +09:00
committed by GitHub
parent dfba702c74
commit ebf1199733
1 changed files with 0 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/plugins/intel_gpu/src/kernel_selector/kernels/convolution/convolution_kernel_b_fs_zyx_fsv16_imad.cpp
View File

@@ -372,8 +372,6 @@ ParamsKey Convolution_kernel_b_fs_zyx_fsv16_imad::GetSupportedKey() const {
k.EnableGroupedConvolution();
k.EnableQuantization(QuantizationType::SYMMETRIC);
k.EnableQuantization(QuantizationType::ASYMMETRIC_DATA);
k.EnableQuantization(QuantizationType::ASYMMETRIC_WEIGHTS);
k.EnableQuantization(QuantizationType::ASYMMETRIC_DATA_AND_WEIGHTS);
k.EnableDilation();
return k;
}