nnet3-shuffle-egs.cc
Go to the documentation of this file.
1 // nnet3bin/nnet3-shuffle-egs.cc
2 
3 // Copyright 2012-2015 Johns Hopkins University (author: Daniel Povey)
4 // 2014 Vimal Manohar
5 
6 // See ../../COPYING for clarification regarding multiple authors
7 //
8 // Licensed under the Apache License, Version 2.0 (the "License");
9 // you may not use this file except in compliance with the License.
10 // You may obtain a copy of the License at
11 //
12 // http://www.apache.org/licenses/LICENSE-2.0
13 //
14 // THIS CODE IS PROVIDED *AS IS* BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
15 // KIND, EITHER EXPRESS OR IMPLIED, INCLUDING WITHOUT LIMITATION ANY IMPLIED
16 // WARRANTIES OR CONDITIONS OF TITLE, FITNESS FOR A PARTICULAR PURPOSE,
17 // MERCHANTABLITY OR NON-INFRINGEMENT.
18 // See the Apache 2 License for the specific language governing permissions and
19 // limitations under the License.
20 
21 #include "base/kaldi-common.h"
22 #include "util/common-utils.h"
23 #include "hmm/transition-model.h"
24 #include "nnet3/nnet-example.h"
25 
26 int main(int argc, char *argv[]) {
27  try {
28  using namespace kaldi;
29  using namespace kaldi::nnet3;
30  typedef kaldi::int32 int32;
31  typedef kaldi::int64 int64;
32 
33  const char *usage =
34  "Copy examples (typically single frames or small groups of frames) for\n"
35  "neural network training, from the input to output, but randomly shuffle the order.\n"
36  "This program will keep all of the examples in memory at once, unless you\n"
37  "use the --buffer-size option\n"
38  "\n"
39  "Usage: nnet3-shuffle-egs [options] <egs-rspecifier> <egs-wspecifier>\n"
40  "\n"
41  "nnet3-shuffle-egs --srand=1 ark:train.egs ark:shuffled.egs\n";
42 
43  int32 srand_seed = 0;
44  int32 buffer_size = 0;
45  ParseOptions po(usage);
46  po.Register("srand", &srand_seed, "Seed for random number generator ");
47  po.Register("buffer-size", &buffer_size, "If >0, size of a buffer we use "
48  "to do limited-memory partial randomization. Otherwise, do "
49  "full randomization.");
50 
51  po.Read(argc, argv);
52 
53  srand(srand_seed);
54 
55  if (po.NumArgs() != 2) {
56  po.PrintUsage();
57  exit(1);
58  }
59 
60  std::string examples_rspecifier = po.GetArg(1),
61  examples_wspecifier = po.GetArg(2);
62 
63  int64 num_done = 0;
64 
65  std::vector<std::pair<std::string, NnetExample*> > egs;
66 
67  SequentialNnetExampleReader example_reader(examples_rspecifier);
68  NnetExampleWriter example_writer(examples_wspecifier);
69  if (buffer_size == 0) { // Do full randomization
70  // Putting in an extra level of indirection here to avoid excessive
71  // computation and memory demands when we have to resize the vector.
72 
73  for (; !example_reader.Done(); example_reader.Next())
74  egs.push_back(std::make_pair(example_reader.Key(),
75  new NnetExample(example_reader.Value())));
76 
77  std::random_shuffle(egs.begin(), egs.end());
78  } else {
79  KALDI_ASSERT(buffer_size > 0);
80  egs.resize(buffer_size,
81  std::pair<std::string, NnetExample*>("", NULL));
82  for (; !example_reader.Done(); example_reader.Next()) {
83  int32 index = RandInt(0, buffer_size - 1);
84  if (egs[index].second == NULL) {
85  egs[index] = std::make_pair(example_reader.Key(),
86  new NnetExample(example_reader.Value()));
87  } else {
88  example_writer.Write(egs[index].first, *(egs[index].second));
89  egs[index].first = example_reader.Key();
90  *(egs[index].second) = example_reader.Value();
91  num_done++;
92  }
93  }
94  }
95  for (size_t i = 0; i < egs.size(); i++) {
96  if (egs[i].second != NULL) {
97  example_writer.Write(egs[i].first, *(egs[i].second));
98  delete egs[i].second;
99  num_done++;
100  }
101  }
102 
103  KALDI_LOG << "Shuffled order of " << num_done
104  << " neural-network training examples "
105  << (buffer_size ? "using a buffer (partial randomization)" : "");
106 
107  return (num_done == 0 ? 1 : 0);
108  } catch(const std::exception &e) {
109  std::cerr << e.what() << '\n';
110  return -1;
111  }
112 }
NnetExample is the input data and corresponding label (or labels) for one or more frames of input...
Definition: nnet-example.h:111
This code computes Goodness of Pronunciation (GOP) and extracts phone-level pronunciation feature for...
Definition: chain.dox:20
void PrintUsage(bool print_command_line=false)
Prints the usage documentation [provided in the constructor].
A templated class for writing objects to an archive or script file; see The Table concept...
Definition: kaldi-table.h:368
kaldi::int32 int32
void Write(const std::string &key, const T &value) const
void Register(const std::string &name, bool *ptr, const std::string &doc)
The class ParseOptions is for parsing command-line options; see Parsing command-line options for more...
Definition: parse-options.h:36
A templated class for reading objects sequentially from an archive or script file; see The Table conc...
Definition: kaldi-table.h:287
int Read(int argc, const char *const *argv)
Parses the command line options and fills the ParseOptions-registered variables.
std::string GetArg(int param) const
Returns one of the positional parameters; 1-based indexing for argc/argv compatibility.
int NumArgs() const
Number of positional parameters (c.f. argc-1).
#define KALDI_ASSERT(cond)
Definition: kaldi-error.h:185
#define KALDI_LOG
Definition: kaldi-error.h:153
int32 RandInt(int32 min_val, int32 max_val, struct RandomState *state)
Definition: kaldi-math.cc:95
int main(int argc, char *argv[])