ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-18 12:00:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

Rangify constructor of HostTensorDescriptor & Tensor<> (#445)

Browse Source 
* Rangify STL algorithms

This commit adapts rangified std::copy(), std::fill() & std::transform()

* Rangify check_err()

By rangifying check_err(), we can not only compare values between
std::vector<>s, but also compare any ranges which have same value
type.

* Allow constructing Tensor<> like a HostTensorDescriptor

* Simplify Tensor<> object construction logics

* Remove more unnecessary 'HostTensorDescriptor' objects

* Re-format example code

* Re-write more HostTensorDescriptor ctor call

[ROCm/composable_kernel commit: 4a2a56c22f]
This commit is contained in:
Po Yen Chen
2022-11-12 01:36:01 +08:00
committed by GitHub
parent a4d6101e99
commit f2dd2e5b09
103 changed files with 657 additions and 649 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
example/27_layernorm/layernorm_blockwise.cpp
View File

@@ -17,6 +17,7 @@
#include "ck/library/utility/host_common_util.hpp"
#include "ck/library/utility/host_tensor.hpp"
#include "ck/library/utility/host_tensor_generator.hpp"
#include "ck/library/utility/literals.hpp"
#include "ck/library/reference_tensor_operation/cpu/reference_layernorm.hpp"
using XDataType = ck::half_t;
@@ -60,13 +61,13 @@ int main()
ck::index_t Stride = N;
auto f_host_tensor_descriptor1d = [](std::size_t len, std::size_t stride) {
return HostTensorDescriptor(std::vector<std::size_t>({len}),
std::vector<std::size_t>({stride}));
return HostTensorDescriptor({len}, {stride});
};
auto f_host_tensor_descriptor2d = [](std::size_t row, std::size_t col, std::size_t stride) {
return HostTensorDescriptor(std::vector<std::size_t>({row, col}),
std::vector<std::size_t>({stride, 1}));
using namespace ck::literals;
return HostTensorDescriptor({row, col}, {stride, 1_uz});
};
Tensor<XDataType> x(f_host_tensor_descriptor2d(M, N, Stride));
@@ -132,8 +133,7 @@ int main()
ref_invoker.Run(ref_argument);
y_dev.FromDevice(y.mData.data());
pass &=
ck::utils::check_err(y.mData, host_y.mData, "Error: Incorrect results d1", 1e-3, 1e-3);
pass &= ck::utils::check_err(y, host_y, "Error: Incorrect results d1", 1e-3, 1e-3);
}
return (pass ? 0 : 1);
}