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
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
|
//
// Copyright © 2017 Arm Ltd. All rights reserved.
// SPDX-License-Identifier: MIT
//
#pragma once
#include "QuantizeHelper.hpp"
#include "WorkloadTestUtils.hpp"
#include <armnn/ArmNN.hpp>
#include <armnn/Tensor.hpp>
#include <armnn/TypesUtils.hpp>
#include <backendsCommon/CpuTensorHandle.hpp>
#include <backendsCommon/IBackendInternal.hpp>
#include <backendsCommon/WorkloadFactory.hpp>
#include <test/TensorHelpers.hpp>
#include <algorithm>
template<armnn::DataType ArmnnType, std::size_t n, typename T = armnn::ResolveType<ArmnnType>>
LayerTestResult<T, n> SimpleSoftmaxBaseTestImpl(
armnn::IWorkloadFactory& workloadFactory,
const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager,
float beta,
const armnn::TensorShape& inputShape,
const std::vector<float>& outputData)
{
using std::exp;
const float qScale = 1.f / 256.f;
const int qOffset = 0;
armnn::TensorInfo inputTensorInfo;
armnn::TensorInfo outputTensorInfo;
inputTensorInfo = armnn::TensorInfo(inputShape, ArmnnType);
inputTensorInfo.SetQuantizationScale(qScale);
inputTensorInfo.SetQuantizationOffset(qOffset);
outputTensorInfo = armnn::TensorInfo(inputShape, ArmnnType);
outputTensorInfo.SetQuantizationScale(qScale);
outputTensorInfo.SetQuantizationOffset(qOffset);
LayerTestResult<T, n> ret(outputTensorInfo);
// Each row is independently softmax'd.
auto input = MakeTensor<T, n>(inputTensorInfo, std::vector<T>(
QuantizedVector<T>(qScale, qOffset, {
0.f, 1.f, 0.f, 0.f,
.5f, 0.f, 0.f, 0.f,
})));
std::unique_ptr<armnn::ITensorHandle> inputHandle = workloadFactory.CreateTensorHandle(inputTensorInfo);
std::unique_ptr<armnn::ITensorHandle> outputHandle = workloadFactory.CreateTensorHandle(outputTensorInfo);
armnn::SoftmaxQueueDescriptor data;
data.m_Parameters.m_Beta = beta;
armnn::WorkloadInfo info;
AddInputToWorkload(data, info, inputTensorInfo, inputHandle.get());
AddOutputToWorkload(data, info, outputTensorInfo, outputHandle.get());
std::unique_ptr<armnn::IWorkload> workload = workloadFactory.CreateSoftmax(data, info);
inputHandle->Allocate();
outputHandle->Allocate();
CopyDataToITensorHandle(inputHandle.get(), input.origin());
BOOST_ASSERT(workload);
ExecuteWorkload(*workload, memoryManager);
CopyDataFromITensorHandle(ret.output.origin(), outputHandle.get());
std::vector<T> expectedOutput = std::vector<T>(
QuantizedVector<T>(qScale, qOffset, outputData));
ret.outputExpected = MakeTensor<T, n>(outputTensorInfo, expectedOutput);
return ret;
}
template<armnn::DataType ArmnnType, typename T = armnn::ResolveType<ArmnnType>>
LayerTestResult<T, 2> SimpleSoftmaxTestImpl(
armnn::IWorkloadFactory& workloadFactory,
const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager,
float beta)
{
using std::exp;
const armnn::TensorShape inputShape{ 2, 4 };
float x0[4] = { exp((0.f - 1.0f) * beta), exp((1.0f - 1.0f) * beta),
exp((0.0f - 1.0f) * beta), exp((0.0f - 1.0f) * beta) };
float sum0 = x0[0] + x0[1] + x0[2] + x0[3];
float x1[4] = { exp((0.5f - 0.5f) * beta), exp((0.0f - 0.5f) * beta),
exp((0.0f - 0.5f) * beta), exp((0.0f - 0.5f) * beta) };
float sum1 = x1[0] + x1[1] + x1[2] + x1[3];
const std::vector<float> outputData = { x0[0] / sum0, x0[1] / sum0, x0[2] / sum0, x0[3] / sum0,
x1[0] / sum1, x1[1] / sum1, x1[2] / sum1, x1[3] / sum1 };
return SimpleSoftmaxBaseTestImpl<ArmnnType, 2>(workloadFactory, memoryManager, beta, inputShape, outputData);
}
template<armnn::DataType ArmnnType, typename T = armnn::ResolveType<ArmnnType>>
LayerTestResult<T, 3> Simple3dSoftmaxTestImpl(
armnn::IWorkloadFactory& workloadFactory,
const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager,
float beta)
{
const armnn::TensorShape inputShape{ 1, 8, 1 };
const std::vector<float> outputData = { 0.0964599f, 0.26220518f, 0.0964599f, 0.0964599f,
0.15903549f, 0.0964599f, 0.0964599f, 0.0964599f };
return SimpleSoftmaxBaseTestImpl<ArmnnType, 3>(workloadFactory, memoryManager, beta, inputShape, outputData);
}
template<armnn::DataType ArmnnType, typename T = armnn::ResolveType<ArmnnType>>
LayerTestResult<T, 4> Simple4dSoftmaxTestImpl(
armnn::IWorkloadFactory& workloadFactory,
const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager,
float beta)
{
const armnn::TensorShape inputShape{ 1, 8, 1, 1 };
const std::vector<float> outputData = { 0.0964599f, 0.26220518f, 0.0964599f, 0.0964599f,
0.15903549f, 0.0964599f, 0.0964599f, 0.0964599f };
return SimpleSoftmaxBaseTestImpl<ArmnnType, 4>(workloadFactory, memoryManager, beta, inputShape, outputData);
}
template<armnn::DataType ArmnnType, typename T = armnn::ResolveType<ArmnnType>>
LayerTestResult<T, 2> CompareSoftmaxTestImpl(
armnn::IWorkloadFactory& workloadFactory,
const armnn::IBackendInternal::IMemoryManagerSharedPtr& memoryManager,
armnn::IWorkloadFactory& refWorkloadFactory,
float beta)
{
const int batchSize = 20;
const int channels = 30;
armnn::TensorInfo inputTensorInfo;
armnn::TensorInfo outputTensorInfo;
unsigned int inputShape[] = { batchSize, channels };
inputTensorInfo = armnn::TensorInfo(2, inputShape, ArmnnType);
outputTensorInfo = armnn::TensorInfo(2, inputShape, ArmnnType);
float qScale = 1.f / 256.f;
int qOffset = 0;
inputTensorInfo.SetQuantizationScale(qScale);
inputTensorInfo.SetQuantizationOffset(qOffset);
outputTensorInfo.SetQuantizationScale(qScale);
outputTensorInfo.SetQuantizationOffset(qOffset);
LayerTestResult<T, 2> ret(outputTensorInfo);
auto input = MakeRandomTensor<T, 2>(inputTensorInfo, 0xF00D, 0.0f, 1.0f);
std::unique_ptr<armnn::ITensorHandle> inputHandle = workloadFactory.CreateTensorHandle(inputTensorInfo);
std::unique_ptr<armnn::ITensorHandle> outputHandle = workloadFactory.CreateTensorHandle(outputTensorInfo);
armnn::SoftmaxQueueDescriptor data;
data.m_Parameters.m_Beta = beta;
armnn::WorkloadInfo info;
AddInputToWorkload(data, info, inputTensorInfo, inputHandle.get());
AddOutputToWorkload(data, info, outputTensorInfo, outputHandle.get());
std::unique_ptr<armnn::ITensorHandle> outputHandleRef = refWorkloadFactory.CreateTensorHandle(outputTensorInfo);
std::unique_ptr<armnn::ITensorHandle> inputHandleRef = refWorkloadFactory.CreateTensorHandle(inputTensorInfo);
armnn::SoftmaxQueueDescriptor refData = data;
armnn::WorkloadInfo refInfo = info;
SetWorkloadInput(refData, refInfo, 0, inputTensorInfo, inputHandleRef.get());
SetWorkloadOutput(refData, refInfo, 0, outputTensorInfo, outputHandleRef.get());
std::unique_ptr<armnn::IWorkload> workload = workloadFactory.CreateSoftmax(data, info);
std::unique_ptr<armnn::IWorkload> workloadRef = refWorkloadFactory.CreateSoftmax(refData, refInfo);
outputHandleRef->Allocate();
inputHandleRef->Allocate();
inputHandle->Allocate();
outputHandle->Allocate();
CopyDataToITensorHandle(inputHandle.get(), &input[0][0]);
CopyDataToITensorHandle(inputHandleRef.get(), &input[0][0]);
ExecuteWorkload(*workload, memoryManager);
workloadRef->Execute();
CopyDataFromITensorHandle(&ret.output[0][0], outputHandle.get());
CopyDataFromITensorHandle(&ret.outputExpected[0][0], outputHandleRef.get());
return ret;
}
|
