summaryrefslogtreecommitdiff
path: root/training/crf/mpi_extract_reachable.cc
diff options
context:
space:
mode:
authorAvneesh Saluja <asaluja@gmail.com>2013-03-28 18:28:16 -0700
committerAvneesh Saluja <asaluja@gmail.com>2013-03-28 18:28:16 -0700
commit3d8d656fa7911524e0e6885647173474524e0784 (patch)
tree81b1ee2fcb67980376d03f0aa48e42e53abff222 /training/crf/mpi_extract_reachable.cc
parentbe7f57fdd484e063775d7abf083b9fa4c403b610 (diff)
parent96fedabebafe7a38a6d5928be8fff767e411d705 (diff)
fixed conflicts
Diffstat (limited to 'training/crf/mpi_extract_reachable.cc')
-rw-r--r--training/crf/mpi_extract_reachable.cc163
1 files changed, 163 insertions, 0 deletions
diff --git a/training/crf/mpi_extract_reachable.cc b/training/crf/mpi_extract_reachable.cc
new file mode 100644
index 00000000..2a7c2b9d
--- /dev/null
+++ b/training/crf/mpi_extract_reachable.cc
@@ -0,0 +1,163 @@
+#include <iostream>
+#include <sstream>
+#include <vector>
+#include <cassert>
+
+#include "config.h"
+#ifdef HAVE_MPI
+#include <boost/mpi.hpp>
+#endif
+#include <boost/program_options.hpp>
+#include <boost/program_options/variables_map.hpp>
+
+#include "ff_register.h"
+#include "verbose.h"
+#include "filelib.h"
+#include "fdict.h"
+#include "decoder.h"
+#include "weights.h"
+
+using namespace std;
+namespace po = boost::program_options;
+
+bool InitCommandLine(int argc, char** argv, po::variables_map* conf) {
+ po::options_description opts("Configuration options");
+ opts.add_options()
+ ("training_data,t",po::value<string>(),"Training data corpus")
+ ("decoder_config,c",po::value<string>(),"Decoder configuration file")
+ ("weights,w", po::value<string>(), "(Optional) weights file; weights may affect what features are encountered in pruning configurations")
+ ("output_prefix,o",po::value<string>()->default_value("reachable"),"Output path prefix");
+ po::options_description clo("Command line options");
+ clo.add_options()
+ ("config", po::value<string>(), "Configuration file")
+ ("help,h", "Print this help message and exit");
+ po::options_description dconfig_options, dcmdline_options;
+ dconfig_options.add(opts);
+ dcmdline_options.add(opts).add(clo);
+
+ po::store(parse_command_line(argc, argv, dcmdline_options), *conf);
+ if (conf->count("config")) {
+ ifstream config((*conf)["config"].as<string>().c_str());
+ po::store(po::parse_config_file(config, dconfig_options), *conf);
+ }
+ po::notify(*conf);
+
+ if (conf->count("help") || !conf->count("training_data") || !conf->count("decoder_config")) {
+ cerr << "Decode an input set (optionally in parallel using MPI) and write\nout the inputs that produce reachable parallel parses.\n";
+ cerr << dcmdline_options << endl;
+ return false;
+ }
+ return true;
+}
+
+void ReadTrainingCorpus(const string& fname, int rank, int size, vector<string>* c) {
+ ReadFile rf(fname);
+ istream& in = *rf.stream();
+ string line;
+ int lc = 0;
+ while(in) {
+ getline(in, line);
+ if (!in) break;
+ if (lc % size == rank) c->push_back(line);
+ ++lc;
+ }
+}
+
+static const double kMINUS_EPSILON = -1e-6;
+
+struct ReachabilityObserver : public DecoderObserver {
+
+ virtual void NotifyDecodingStart(const SentenceMetadata&) {
+ reachable = false;
+ }
+
+ // compute model expectations, denominator of objective
+ virtual void NotifyTranslationForest(const SentenceMetadata&, Hypergraph* hg) {
+ }
+
+ // compute "empirical" expectations, numerator of objective
+ virtual void NotifyAlignmentForest(const SentenceMetadata& smeta, Hypergraph* hg) {
+ reachable = true;
+ }
+
+ bool reachable;
+};
+
+#ifdef HAVE_MPI
+namespace mpi = boost::mpi;
+#endif
+
+int main(int argc, char** argv) {
+#ifdef HAVE_MPI
+ mpi::environment env(argc, argv);
+ mpi::communicator world;
+ const int size = world.size();
+ const int rank = world.rank();
+#else
+ const int size = 1;
+ const int rank = 0;
+#endif
+ if (size > 1) SetSilent(true); // turn off verbose decoder output
+ register_feature_functions();
+
+ po::variables_map conf;
+ if (!InitCommandLine(argc, argv, &conf))
+ return false;
+
+ // load cdec.ini and set up decoder
+ ReadFile ini_rf(conf["decoder_config"].as<string>());
+ Decoder decoder(ini_rf.stream());
+ if (decoder.GetConf()["input"].as<string>() != "-") {
+ cerr << "cdec.ini must not set an input file\n";
+ abort();
+ }
+
+ if (FD::UsingPerfectHashFunction()) {
+ cerr << "Your configuration file has enabled a cmph hash function. Please disable.\n";
+ return 1;
+ }
+
+ // load optional weights
+ if (conf.count("weights"))
+ Weights::InitFromFile(conf["weights"].as<string>(), &decoder.CurrentWeightVector());
+
+ vector<string> corpus;
+ ReadTrainingCorpus(conf["training_data"].as<string>(), rank, size, &corpus);
+ assert(corpus.size() > 0);
+
+
+ if (rank == 0)
+ cerr << "Each processor is decoding ~" << corpus.size() << " training examples...\n";
+
+ size_t num_reached = 0;
+ {
+ ostringstream os;
+ os << conf["output_prefix"].as<string>() << '.' << rank << "_of_" << size;
+ WriteFile wf(os.str());
+ ostream& out = *wf.stream();
+ ReachabilityObserver observer;
+ for (int i = 0; i < corpus.size(); ++i) {
+ decoder.Decode(corpus[i], &observer);
+ if (observer.reachable) {
+ out << corpus[i] << endl;
+ ++num_reached;
+ }
+ corpus[i].clear();
+ }
+ cerr << "Shard " << rank << '/' << size << " finished, wrote "
+ << num_reached << " instances to " << os.str() << endl;
+ }
+
+ size_t total = 0;
+#ifdef HAVE_MPI
+ reduce(world, num_reached, total, std::plus<double>(), 0);
+#else
+ total = num_reached;
+#endif
+ if (rank == 0) {
+ cerr << "-----------------------------------------\n";
+ cerr << "TOTAL = " << total << " instances\n";
+ }
+ return 0;
+}
+