mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-05-05 22:22:27 +00:00
500fa99512
* convnd_fwd fp16 example * update example * update example * update instance * updating refernce conv * update reference conv * update conv fwd profiler * update conv 1d and 3d instance * update include path * clean * update profiler for conv bwd data and weight * update conv bwd weight * clean * update conv example * update profiler for conv bwd weight * update ckprofiler for conv bwd data * fix reference conv bwd data bug; update conv bwd data test * update examples * fix initialization issue * update test for conv fwd * clean * clean * remove test case too sensitive to error threshhold * fix test * clean * fix build * adding conv multiple d * adding conv multiple D * add matrix padder * add gemm padding to convnd * adding group conv * update gemm multi-d * refactor * refactor * refactor * clean * clean * refactor * refactor * reorg * add ds * add bias * clean * add G * adding group * adding group * adding group * update Tensor * clean * update example * update DeviceGemmMultipleD_Xdl_CShuffle * update conv bwd-data and bwd-weight * upate contraction example * update gemm and batch gemm with e permute * fix example build * instance for grouped conv1d * update example * adding group conv instance * update gemm bilinear instance * update gemm+add+add+fastgelu instance * update profiler * update profiler * update test * update test and client example * clean * add grouped conv into profiler * update profiler * clean * add test grouped conv, update all conv test to gtest * update test
242 lines
12 KiB
C++
242 lines
12 KiB
C++
// SPDX-License-Identifier: MIT
|
|
// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
|
|
|
|
#include <cstdlib>
|
|
#include <iostream>
|
|
#include <initializer_list>
|
|
#include <vector>
|
|
#include <gtest/gtest.h>
|
|
|
|
#include "profiler/include/profile_conv_fwd_impl.hpp"
|
|
|
|
class TestConvndFwd : public ::testing::Test
|
|
{
|
|
protected:
|
|
std::vector<ck::utils::conv::ConvParam> conv_params;
|
|
};
|
|
|
|
// 1d
|
|
TEST_F(TestConvndFwd, Conv1dFwd)
|
|
{
|
|
conv_params.clear();
|
|
conv_params.push_back({1, 1, 128, 128, 256, {1}, {14}, {2}, {1}, {0}, {0}});
|
|
conv_params.push_back({1, 1, 128, 128, 256, {3}, {28}, {1}, {1}, {1}, {1}});
|
|
conv_params.push_back({1, 1, 128, 128, 256, {1}, {3}, {1}, {1}, {0}, {0}});
|
|
|
|
for(auto& param : conv_params)
|
|
{
|
|
bool pass;
|
|
|
|
// fp32
|
|
pass = ck::profiler::profile_conv_fwd_impl<1,
|
|
ck::tensor_layout::convolution::NWC,
|
|
ck::tensor_layout::convolution::KXC,
|
|
ck::tensor_layout::convolution::NWK,
|
|
float,
|
|
float,
|
|
float>(true, // do_verification
|
|
1, // init_method
|
|
false, // do_log
|
|
false, // time_kernel
|
|
param);
|
|
|
|
EXPECT_TRUE(pass);
|
|
|
|
// fp16
|
|
pass = ck::profiler::profile_conv_fwd_impl<1,
|
|
ck::tensor_layout::convolution::NWC,
|
|
ck::tensor_layout::convolution::KXC,
|
|
ck::tensor_layout::convolution::NWK,
|
|
ck::half_t,
|
|
ck::half_t,
|
|
ck::half_t>(true, // do_verification
|
|
1, // init_method
|
|
false, // do_log
|
|
false, // time_kernel
|
|
param);
|
|
|
|
EXPECT_TRUE(pass);
|
|
|
|
// bf16
|
|
pass = ck::profiler::profile_conv_fwd_impl<1,
|
|
ck::tensor_layout::convolution::NWC,
|
|
ck::tensor_layout::convolution::KXC,
|
|
ck::tensor_layout::convolution::NWK,
|
|
ck::bhalf_t,
|
|
ck::bhalf_t,
|
|
ck::bhalf_t>(true, // do_verification
|
|
1, // init_method
|
|
false, // do_log
|
|
false, // time_kernel
|
|
param);
|
|
|
|
EXPECT_TRUE(pass);
|
|
|
|
// int8
|
|
pass = ck::profiler::profile_conv_fwd_impl<1,
|
|
ck::tensor_layout::convolution::NWC,
|
|
ck::tensor_layout::convolution::KXC,
|
|
ck::tensor_layout::convolution::NWK,
|
|
int8_t,
|
|
int8_t,
|
|
int8_t>(true, // do_verification
|
|
1, // init_method
|
|
false, // do_log
|
|
false, // time_kernel
|
|
param);
|
|
|
|
EXPECT_TRUE(pass);
|
|
}
|
|
}
|
|
|
|
// 2d
|
|
TEST_F(TestConvndFwd, Conv2dFwd)
|
|
{
|
|
conv_params.clear();
|
|
conv_params.push_back({2, 1, 128, 128, 256, {1, 1}, {7, 7}, {2, 2}, {1, 1}, {0, 0}, {0, 0}});
|
|
conv_params.push_back({2, 1, 128, 128, 256, {3, 3}, {14, 14}, {1, 1}, {1, 1}, {1, 1}, {1, 1}});
|
|
conv_params.push_back({2, 1, 128, 128, 256, {1, 1}, {3, 3}, {1, 1}, {1, 1}, {0, 0}, {0, 0}});
|
|
|
|
for(auto& param : conv_params)
|
|
{
|
|
bool pass;
|
|
|
|
// fp32
|
|
pass = ck::profiler::profile_conv_fwd_impl<2,
|
|
ck::tensor_layout::convolution::NHWC,
|
|
ck::tensor_layout::convolution::KYXC,
|
|
ck::tensor_layout::convolution::NHWK,
|
|
float,
|
|
float,
|
|
float>(true, // do_verification
|
|
1, // init_method
|
|
false, // do_log
|
|
false, // time_kernel
|
|
param);
|
|
|
|
EXPECT_TRUE(pass);
|
|
|
|
// fp16
|
|
pass = ck::profiler::profile_conv_fwd_impl<2,
|
|
ck::tensor_layout::convolution::NHWC,
|
|
ck::tensor_layout::convolution::KYXC,
|
|
ck::tensor_layout::convolution::NHWK,
|
|
ck::half_t,
|
|
ck::half_t,
|
|
ck::half_t>(true, // do_verification
|
|
1, // init_method
|
|
false, // do_log
|
|
false, // time_kernel
|
|
param);
|
|
|
|
EXPECT_TRUE(pass);
|
|
|
|
// bf16
|
|
pass = ck::profiler::profile_conv_fwd_impl<2,
|
|
ck::tensor_layout::convolution::NHWC,
|
|
ck::tensor_layout::convolution::KYXC,
|
|
ck::tensor_layout::convolution::NHWK,
|
|
ck::bhalf_t,
|
|
ck::bhalf_t,
|
|
ck::bhalf_t>(true, // do_verification
|
|
1, // init_method
|
|
false, // do_log
|
|
false, // time_kernel
|
|
param);
|
|
|
|
EXPECT_TRUE(pass);
|
|
|
|
// int8
|
|
pass = ck::profiler::profile_conv_fwd_impl<2,
|
|
ck::tensor_layout::convolution::NHWC,
|
|
ck::tensor_layout::convolution::KYXC,
|
|
ck::tensor_layout::convolution::NHWK,
|
|
int8_t,
|
|
int8_t,
|
|
int8_t>(true, // do_verification
|
|
1, // init_method
|
|
false, // do_log
|
|
false, // time_kernel
|
|
param);
|
|
|
|
EXPECT_TRUE(pass);
|
|
}
|
|
}
|
|
|
|
// 3d
|
|
TEST_F(TestConvndFwd, Conv3dFwd)
|
|
{
|
|
conv_params.clear();
|
|
conv_params.push_back(
|
|
{3, 1, 128, 128, 256, {1, 1, 1}, {7, 7, 7}, {2, 2, 2}, {1, 1, 1}, {0, 0, 0}, {0, 0, 0}});
|
|
conv_params.push_back(
|
|
{3, 1, 128, 128, 256, {3, 3, 3}, {14, 14, 3}, {1, 1, 1}, {1, 1, 1}, {1, 1, 1}, {1, 1, 1}});
|
|
conv_params.push_back(
|
|
{3, 1, 128, 128, 256, {1, 1, 1}, {3, 3, 3}, {1, 1, 1}, {1, 1, 1}, {0, 0, 0}, {0, 0, 0}});
|
|
|
|
for(auto& param : conv_params)
|
|
{
|
|
bool pass;
|
|
|
|
// fp32
|
|
pass = ck::profiler::profile_conv_fwd_impl<3,
|
|
ck::tensor_layout::convolution::NDHWC,
|
|
ck::tensor_layout::convolution::KZYXC,
|
|
ck::tensor_layout::convolution::NDHWK,
|
|
float,
|
|
float,
|
|
float>(true, // do_verification
|
|
1, // init_method
|
|
false, // do_log
|
|
false, // time_kernel
|
|
param);
|
|
|
|
EXPECT_TRUE(pass);
|
|
|
|
// fp16
|
|
pass = ck::profiler::profile_conv_fwd_impl<3,
|
|
ck::tensor_layout::convolution::NDHWC,
|
|
ck::tensor_layout::convolution::KZYXC,
|
|
ck::tensor_layout::convolution::NDHWK,
|
|
ck::half_t,
|
|
ck::half_t,
|
|
ck::half_t>(true, // do_verification
|
|
1, // init_method
|
|
false, // do_log
|
|
false, // time_kernel
|
|
param);
|
|
|
|
EXPECT_TRUE(pass);
|
|
|
|
// bf16
|
|
pass = ck::profiler::profile_conv_fwd_impl<3,
|
|
ck::tensor_layout::convolution::NDHWC,
|
|
ck::tensor_layout::convolution::KZYXC,
|
|
ck::tensor_layout::convolution::NDHWK,
|
|
ck::bhalf_t,
|
|
ck::bhalf_t,
|
|
ck::bhalf_t>(true, // do_verification
|
|
1, // init_method
|
|
false, // do_log
|
|
false, // time_kernel
|
|
param);
|
|
|
|
EXPECT_TRUE(pass);
|
|
|
|
// int8
|
|
pass = ck::profiler::profile_conv_fwd_impl<3,
|
|
ck::tensor_layout::convolution::NDHWC,
|
|
ck::tensor_layout::convolution::KZYXC,
|
|
ck::tensor_layout::convolution::NDHWK,
|
|
int8_t,
|
|
int8_t,
|
|
int8_t>(true, // do_verification
|
|
1, // init_method
|
|
false, // do_log
|
|
false, // time_kernel
|
|
param);
|
|
|
|
EXPECT_TRUE(pass);
|
|
}
|
|
}
|