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
|
// Copyright (C) 2018-2019 Intel Corporation
// SPDX-License-Identifier: Apache-2.0
//
#include "ext_list.hpp"
#include "ext_base.hpp"
#include <cmath>
#include <string>
#include <vector>
#include <cassert>
#include <algorithm>
#include "ie_parallel.hpp"
namespace InferenceEngine {
namespace Extensions {
namespace Cpu {
class ReverseSequenceImpl: public ExtLayerBase {
public:
explicit ReverseSequenceImpl(const CNNLayer* layer) {
try {
if (layer->insData.size() != 2 || layer->outData.size() != 1)
THROW_IE_EXCEPTION << layer->name << " Incorrect number of input/output edges!";
src_dims = layer->insData[REVERSESEQUENCE_DATA].lock()->getTensorDesc().getDims();
SizeVector seq_lengths_dims = layer->insData[REVERSESEQUENCE_LENGTHS].lock()->getTensorDesc().getDims();
if (layer->insData[REVERSESEQUENCE_LENGTHS].lock()->getTensorDesc().getPrecision() != Precision::I32 &&
layer->insData[REVERSESEQUENCE_LENGTHS].lock()->getTensorDesc().getPrecision() != Precision::FP32)
THROW_IE_EXCEPTION << layer->name << " Incorrect 'seq_lengths' input precision. Only FP32 and I32 are supported!";
if (seq_lengths_dims.size() > 1)
THROW_IE_EXCEPTION << layer->name << " Seq_lengths vector should be 1 dimension";
SizeVector dst_dims = layer->outData[0]->getTensorDesc().getDims();
if (src_dims.size() != dst_dims.size())
THROW_IE_EXCEPTION << layer->name << " Incorrect number of input/output sizes!";
for (size_t i = 0; i < dst_dims.size(); i++) {
if (src_dims[i] != dst_dims[i])
THROW_IE_EXCEPTION << layer->name << " Incorrect number of input/output dimension!";
}
seq_axis = layer->GetParamAsInt("seq_axis", 1);
if (seq_axis < 0)
seq_axis += src_dims.size();
if (seq_axis < 0 || seq_axis >= static_cast<int>(src_dims.size()))
THROW_IE_EXCEPTION << layer->name << " Incorrect 'seq_axis' parameters dimensions and axis number!";
batch_axis = layer->GetParamAsInt("batch_axis", 0);
if (batch_axis < 0)
batch_axis += src_dims.size();
if (batch_axis < 0 || batch_axis >= static_cast<int>(src_dims.size()))
THROW_IE_EXCEPTION << layer->name << " Incorrect 'batch_axis' parameters dimensions and axis number!";
if (seq_lengths_dims[0] != dst_dims[batch_axis])
THROW_IE_EXCEPTION << layer->name << " Incorrect 'seq_lengths_dims' parameters dimension!";
srcStrides = layer->insData[REVERSESEQUENCE_DATA].lock()->getTensorDesc().getBlockingDesc().getStrides();
work_amount_dst = srcStrides[0] * src_dims[0];
addConfig(layer, { DataConfigurator(ConfLayout::PLN), DataConfigurator(ConfLayout::PLN) }, { DataConfigurator(ConfLayout::PLN) });
} catch (InferenceEngine::details::InferenceEngineException &ex) {
errorMsg = ex.what();
}
}
StatusCode execute(std::vector<Blob::Ptr>& inputs, std::vector<Blob::Ptr>& outputs, ResponseDesc *resp) noexcept override {
size_t i;
const float *src_data = inputs[REVERSESEQUENCE_DATA]->cbuffer().as<const float *>() +
inputs[REVERSESEQUENCE_DATA]->getTensorDesc().getBlockingDesc().getOffsetPadding();
float* dst_data = outputs[0]->cbuffer().as<float *>() +
outputs[0]->getTensorDesc().getBlockingDesc().getOffsetPadding();
switch (inputs[REVERSESEQUENCE_LENGTHS]->getTensorDesc().getPrecision()) {
case Precision::FP32: {
float *seq_lengths_data = inputs[REVERSESEQUENCE_LENGTHS]->cbuffer().as<float *>() +
inputs[REVERSESEQUENCE_LENGTHS]->getTensorDesc().getBlockingDesc().getOffsetPadding();
for (i = 0; i < src_dims[batch_axis]; i++) {
if (static_cast<int32_t>(seq_lengths_data[i]) > static_cast<int>(src_dims[seq_axis])) {
if (resp) {
std::string errorMsg = "Incorrect input 'seq_lengths' values!";
errorMsg.copy(resp->msg, sizeof(resp->msg) - 1);
}
return PARAMETER_MISMATCH;
}
}
parallel_nt(0, [&](const int ithr, const int nthr) {
size_t i, start = 0, end = 0, src_idx = 0;
SizeVector counters(src_dims.size(), 0);
splitter(work_amount_dst, nthr, ithr, start, end);
for (int j = src_dims.size() - 1, i = start; j >= 0; j--) {
counters[j] = i % src_dims[j];
i /= src_dims[j];
}
for (size_t iwork = start; iwork < end; ++iwork) {
for (i = 0, src_idx = 0; i < src_dims.size(); ++i) {
size_t idx = counters[i];
if (static_cast<int>(i) == seq_axis &&
static_cast<int>(idx) < static_cast<int32_t>(seq_lengths_data[counters[batch_axis]])) {
idx = static_cast<int32_t>(seq_lengths_data[counters[batch_axis]]) - idx - 1;
}
src_idx += idx * srcStrides[i];
}
dst_data[iwork] = src_data[src_idx];
for (int j = src_dims.size() - 1; j >= 0; j--) {
counters[j] = (counters[j] + 1) % src_dims[j];
if (counters[j] != 0) break;
}
}
});
}
break;
case Precision::I32: {
int32_t *seq_lengths_data = inputs[REVERSESEQUENCE_LENGTHS]->cbuffer().as<int32_t *>() +
inputs[REVERSESEQUENCE_LENGTHS]->getTensorDesc().getBlockingDesc().getOffsetPadding();
for (i = 0; i < src_dims[batch_axis]; i++) {
if (seq_lengths_data[i] > static_cast<int>(src_dims[seq_axis])) {
if (resp) {
std::string errorMsg = "Incorrect input 'seq_lengths' values!";
errorMsg.copy(resp->msg, sizeof(resp->msg) - 1);
}
return PARAMETER_MISMATCH;
}
}
parallel_nt(0, [&](const int ithr, const int nthr) {
size_t i, start = 0, end = 0, src_idx = 0;
SizeVector counters(src_dims.size(), 0);
splitter(work_amount_dst, nthr, ithr, start, end);
for (int j = src_dims.size() - 1, i = start; j >= 0; j--) {
counters[j] = i % src_dims[j];
i /= src_dims[j];
}
for (size_t iwork = start; iwork < end; ++iwork) {
for (i = 0, src_idx = 0; i < src_dims.size(); ++i) {
size_t idx = counters[i];
if (static_cast<int>(i) == seq_axis &&
static_cast<int>(idx) < seq_lengths_data[counters[batch_axis]]) {
idx = seq_lengths_data[counters[batch_axis]] - idx - 1;
}
src_idx += idx * srcStrides[i];
}
dst_data[iwork] = src_data[src_idx];
for (int j = src_dims.size() - 1; j >= 0; j--) {
counters[j] = (counters[j] + 1) % src_dims[j];
if (counters[j] != 0) break;
}
}
});
}
break;
default:
return GENERAL_ERROR;
}
return OK;
}
private:
const size_t REVERSESEQUENCE_DATA = 0;
const size_t REVERSESEQUENCE_LENGTHS = 1;
int seq_axis;
int batch_axis;
SizeVector src_dims;
SizeVector srcStrides;
size_t work_amount_dst;
};
REG_FACTORY_FOR(ImplFactory<ReverseSequenceImpl>, ReverseSequence);
} // namespace Cpu
} // namespace Extensions
} // namespace InferenceEngine
|