nnet-parse.cc
8.42 KB
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
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
// nnet3/nnet-parse.cc
// Copyright 2015 Johns Hopkins University (author: Daniel Povey)
// See ../../COPYING for clarification regarding multiple authors
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// THIS CODE IS PROVIDED *AS IS* BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
// KIND, EITHER EXPRESS OR IMPLIED, INCLUDING WITHOUT LIMITATION ANY IMPLIED
// WARRANTIES OR CONDITIONS OF TITLE, FITNESS FOR A PARTICULAR PURPOSE,
// MERCHANTABLITY OR NON-INFRINGEMENT.
// See the Apache 2 License for the specific language governing permissions and
// limitations under the License.
#include <iterator>
#include <sstream>
#include <iomanip>
#include "nnet3/nnet-parse.h"
#include "cudamatrix/cu-vector.h"
#include "cudamatrix/cu-matrix.h"
namespace kaldi {
namespace nnet3 {
bool DescriptorTokenize(const std::string &input,
std::vector<std::string> *tokens) {
KALDI_ASSERT(tokens != NULL);
size_t start = input.find_first_not_of(" \t"), size = input.size();
tokens->clear();
while (start < size) {
KALDI_ASSERT(!isspace(input[start]));
if (input[start] == '(' || input[start] == ')' || input[start] == ',') {
tokens->push_back(std::string(input, start, 1));
start = input.find_first_not_of(" \t", start + 1);
} else {
size_t found = input.find_first_of(" \t(),", start);
KALDI_ASSERT(found != start);
if (found == std::string::npos) {
std::string str(input, start, input.size() - start);
BaseFloat tmp;
if (!IsValidName(str) && !ConvertStringToReal(str, &tmp)) {
KALDI_WARN << "Could not tokenize line " << ErrorContext(std::string(input, start));
return false;
}
tokens->push_back(str);
break;
} else {
if (input[found] == '(' || input[found] == ')' || input[found] == ',') {
std::string str(input, start, found - start);
BaseFloat tmp;
if (!IsValidName(str) && !ConvertStringToReal(str, &tmp)) {
KALDI_WARN << "Could not tokenize line " << ErrorContext(std::string(input, start));
return false;
}
tokens->push_back(str);
start = found;
} else {
std::string str(input, start, found - start);
BaseFloat tmp;
if (!IsValidName(str) && !ConvertStringToReal(str, &tmp)) {
KALDI_WARN << "Could not tokenize line " << ErrorContext(std::string(input, start));
return false;
}
tokens->push_back(str);
start = input.find_first_not_of(" \t", found);
}
}
}
}
return true;
}
std::string ErrorContext(std::istream &is) {
if (!is.good()) return "end of line";
char buf[21];
is.read(buf, 21);
if (is) {
return (std::string(buf, 20) + "...");
}
return std::string(buf, is.gcount());
}
std::string ErrorContext(const std::string &str) {
if (str.size() == 0) return "end of line";
if (str.size() <= 20) return str;
return std::string(str, 0, 20) + "...";
}
static void PrintFloatSuccinctly(std::ostream &os, BaseFloat f) {
if (fabs(f) < 10000.0 && fabs(f) >= 10.0) {
os << std::fixed << std::setprecision(0) << f;
} else if (fabs(f) >= 0.995) {
os << std::fixed << std::setprecision(1) << f;
} else if (fabs(f) >= 0.01) {
os << std::fixed << std::setprecision(2) << f;
} else {
os << std::setprecision(1) << f;
}
os.unsetf(std::ios_base::floatfield);
os << std::setprecision(6); // Restore the default.
}
// Returns a string that summarizes a vector fairly succintly, for
// printing stats in info lines.
std::string SummarizeVector(const VectorBase<float> &vec) {
std::ostringstream os;
if (vec.Dim() < 10) {
os << "[ ";
for (int32 i = 0; i < vec.Dim(); i++) {
PrintFloatSuccinctly(os, vec(i));
os << ' ';
}
os << "]";
} else {
// print out mean and standard deviation, and some selected values.
BaseFloat mean = vec.Sum() / vec.Dim(),
stddev = sqrt(VecVec(vec, vec) / vec.Dim() - mean * mean);
std::string percentiles_str = "0,1,2,5 10,20,50,80,90 95,98,99,100";
std::vector<int32> percentiles;
bool ans = SplitStringToIntegers(percentiles_str, ", ", false,
&percentiles);
KALDI_ASSERT(ans);
os << "[percentiles(" << percentiles_str << ")=(";
Vector<BaseFloat> vec_sorted(vec);
std::sort(vec_sorted.Data(), vec_sorted.Data() + vec_sorted.Dim());
int32 n = vec.Dim() - 1;
for (size_t i = 0; i < percentiles.size(); i++) {
int32 percentile = percentiles[i];
BaseFloat value = vec_sorted((n * percentile) / 100);
PrintFloatSuccinctly(os, value);
if (i + 1 < percentiles.size())
os << (i == 3 || i == 8 ? ' ' : ',');
}
os << std::setprecision(3);
os << "), mean=" << mean << ", stddev=" << stddev << "]";
}
return os.str();
}
std::string SummarizeVector(const VectorBase<double> &vec) {
Vector<float> vec_copy(vec);
return SummarizeVector(vec_copy);
}
std::string SummarizeVector(const CuVectorBase<BaseFloat> &cu_vec) {
Vector<float> vec(cu_vec);
return SummarizeVector(vec);
}
void PrintParameterStats(std::ostringstream &os,
const std::string &name,
const CuVectorBase<BaseFloat> ¶ms,
bool include_mean) {
os << std::setprecision(4);
os << ", " << name << '-';
if (include_mean) {
BaseFloat mean = params.Sum() / params.Dim(),
stddev = std::sqrt(VecVec(params, params) / params.Dim() - mean * mean);
os << "{mean,stddev}=" << mean << ',' << stddev;
} else {
BaseFloat rms = std::sqrt(VecVec(params, params) / params.Dim());
os << "rms=" << rms;
}
os << std::setprecision(6); // restore the default precision.
}
void PrintParameterStats(std::ostringstream &os,
const std::string &name,
const CuMatrix<BaseFloat> ¶ms,
bool include_mean,
bool include_row_norms,
bool include_column_norms,
bool include_singular_values) {
os << std::setprecision(4);
os << ", " << name << '-';
int32 dim = params.NumRows() * params.NumCols();
if (include_mean) {
BaseFloat mean = params.Sum() / dim,
stddev = std::sqrt(TraceMatMat(params, params, kTrans) / dim -
mean * mean);
os << "{mean,stddev}=" << mean << ',' << stddev;
} else {
BaseFloat rms = std::sqrt(TraceMatMat(params, params, kTrans) / dim);
os << "rms=" << rms;
}
os << std::setprecision(6); // restore the default precision.
if (include_row_norms) {
CuVector<BaseFloat> row_norms(params.NumRows());
row_norms.AddDiagMat2(1.0, params, kNoTrans, 0.0);
row_norms.ApplyPow(0.5);
Vector<BaseFloat> row_norms_cpu;
row_norms.Swap(&row_norms_cpu);
os << ", " << name << "-row-norms="
<< SummarizeVector(row_norms_cpu);
}
if (include_column_norms) {
CuVector<BaseFloat> col_norms(params.NumCols());
col_norms.AddDiagMat2(1.0, params, kTrans, 0.0);
col_norms.ApplyPow(0.5);
Vector<BaseFloat> col_norms_cpu;
col_norms.Swap(&col_norms_cpu);
os << ", " << name << "-col-norms="
<< SummarizeVector(col_norms_cpu);
}
if (include_singular_values) {
Matrix<BaseFloat> params_cpu(params);
Vector<BaseFloat> s(std::min(params.NumRows(), params.NumCols()));
params_cpu.Svd(&s);
std::string singular_values_str = SummarizeVector(s);
os << ", " << name << "-singular-values=" << singular_values_str;
std::ostringstream name_os;
}
}
void ParseConfigLines(const std::vector<std::string> &lines,
std::vector<ConfigLine> *config_lines) {
config_lines->resize(lines.size());
for (size_t i = 0; i < lines.size(); i++) {
bool ret = (*config_lines)[i].ParseLine(lines[i]);
if (!ret) {
KALDI_ERR << "Error parsing config line: " << lines[i];
}
}
}
bool NameMatchesPattern(const char *name, const char *pattern) {
if (*pattern == '*') {
return NameMatchesPattern(name, pattern + 1) ||
(*name != '\0' && NameMatchesPattern(name + 1, pattern));
} else if (*name == *pattern) {
return (*name == '\0' || NameMatchesPattern(name + 1, pattern + 1));
} else {
return false;
}
}
} // namespace nnet3
} // namespace kaldi