1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
|
/*
* Copyright (c) 2017 ARM Limited.
*
* SPDX-License-Identifier: MIT
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to
* deal in the Software without restriction, including without limitation the
* rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
* sell copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in all
* copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#include "Globals.h"
#include "NEON/Helper.h"
#include "NEON/NEAccessor.h"
#include "TensorLibrary.h"
#include "benchmark/Datasets.h"
#include "benchmark/Profiler.h"
#include "benchmark/WallClockTimer.h"
#include "arm_compute/core/Helpers.h"
#include "arm_compute/core/Types.h"
#include "arm_compute/runtime/NEON/functions/NENormalizationLayer.h"
#include "arm_compute/runtime/Tensor.h"
#include "arm_compute/runtime/TensorAllocator.h"
#include "benchmark/benchmark_api.h"
using namespace arm_compute;
using namespace arm_compute::test;
using namespace arm_compute::test::benchmark;
using namespace arm_compute::test::neon;
#include "benchmark/common/NormalizationLayer.h"
namespace
{
using NormalizationLayerAlexNetF32 = NormalizationLayer<AlexNetNormalizationLayerDataset, Tensor, NEAccessor, NENormalizationLayer>;
using NormalizationLayerAlexNetQS8 = NormalizationLayer<AlexNetNormalizationLayerDataset, Tensor, NEAccessor, NENormalizationLayer, DataType::QS8>;
using NormalizationLayerGoogLeNet = NormalizationLayer<GoogLeNetNormalizationLayerDataset, Tensor, NEAccessor, NENormalizationLayer>;
} // namespace
// F32
BENCHMARK_DEFINE_F(NormalizationLayerAlexNetF32, neon_alexnet)
(::benchmark::State &state)
{
while(state.KeepRunning())
{
// Run function
profiler.start();
norm_layer->run();
profiler.stop();
}
}
BENCHMARK_REGISTER_F(NormalizationLayerAlexNetF32, neon_alexnet)
->Threads(1)
->Apply(DataSetArgBatched<AlexNetNormalizationLayerDataset, 0, 1, 4, 8>);
BENCHMARK_REGISTER_F(NormalizationLayerAlexNetF32, neon_alexnet)
->Threads(1)
->Apply(DataSetArgBatched<AlexNetNormalizationLayerDataset, 1, 1, 4, 8>);
// QS8
BENCHMARK_DEFINE_F(NormalizationLayerAlexNetQS8, neon_alexnet)
(::benchmark::State &state)
{
while(state.KeepRunning())
{
// Run function
profiler.start();
norm_layer->run();
profiler.stop();
}
}
BENCHMARK_REGISTER_F(NormalizationLayerAlexNetQS8, neon_alexnet)
->Threads(1)
->Apply(DataSetArgBatched<AlexNetNormalizationLayerDataset, 0, 1, 4, 8>);
BENCHMARK_REGISTER_F(NormalizationLayerAlexNetQS8, neon_alexnet)
->Threads(1)
->Apply(DataSetArgBatched<AlexNetNormalizationLayerDataset, 1, 1, 4, 8>);
BENCHMARK_DEFINE_F(NormalizationLayerGoogLeNet, neon_googlenet)
(::benchmark::State &state)
{
while(state.KeepRunning())
{
// Run function
profiler.start();
norm_layer->run();
profiler.stop();
}
}
BENCHMARK_REGISTER_F(NormalizationLayerGoogLeNet, neon_googlenet)
->Threads(1)
->Apply(DataSetArgBatched<GoogLeNetNormalizationLayerDataset, 0, 1, 4, 8>);
BENCHMARK_REGISTER_F(NormalizationLayerGoogLeNet, neon_googlenet)
->Threads(1)
->Apply(DataSetArgBatched<GoogLeNetNormalizationLayerDataset, 1, 1, 4, 8>);
|
