summaryrefslogtreecommitdiff
path: root/dpmert/mr_dpmert_map.cc
diff options
context:
space:
mode:
authorChris Dyer <cdyer@allegro.clab.cs.cmu.edu>2012-11-18 13:35:42 -0500
committerChris Dyer <cdyer@allegro.clab.cs.cmu.edu>2012-11-18 13:35:42 -0500
commit8aa29810bb77611cc20b7a384897ff6703783ea1 (patch)
tree8635daa8fffb3f2cd90e30b41e27f4f9e0909447 /dpmert/mr_dpmert_map.cc
parentfbdacabc85bea65d735f2cb7f92b98e08ce72d04 (diff)
major restructure of the training code
Diffstat (limited to 'dpmert/mr_dpmert_map.cc')
-rw-r--r--dpmert/mr_dpmert_map.cc112
1 files changed, 0 insertions, 112 deletions
diff --git a/dpmert/mr_dpmert_map.cc b/dpmert/mr_dpmert_map.cc
deleted file mode 100644
index d1efcf96..00000000
--- a/dpmert/mr_dpmert_map.cc
+++ /dev/null
@@ -1,112 +0,0 @@
-#include <sstream>
-#include <iostream>
-#include <fstream>
-#include <vector>
-
-#include <boost/program_options.hpp>
-#include <boost/program_options/variables_map.hpp>
-
-#include "ns.h"
-#include "ns_docscorer.h"
-#include "ces.h"
-#include "filelib.h"
-#include "stringlib.h"
-#include "sparse_vector.h"
-#include "mert_geometry.h"
-#include "inside_outside.h"
-#include "error_surface.h"
-#include "b64tools.h"
-#include "hg_io.h"
-
-using namespace std;
-namespace po = boost::program_options;
-
-void InitCommandLine(int argc, char** argv, po::variables_map* conf) {
- po::options_description opts("Configuration options");
- opts.add_options()
- ("reference,r",po::value<vector<string> >(), "[REQD] Reference translation (tokenized text)")
- ("source,s",po::value<string>(), "Source file (ignored, except for AER)")
- ("evaluation_metric,m",po::value<string>()->default_value("ibm_bleu"), "Evaluation metric being optimized")
- ("input,i",po::value<string>()->default_value("-"), "Input file to map (- is STDIN)")
- ("help,h", "Help");
- po::options_description dcmdline_options;
- dcmdline_options.add(opts);
- po::store(parse_command_line(argc, argv, dcmdline_options), *conf);
- bool flag = false;
- if (!conf->count("reference")) {
- cerr << "Please specify one or more references using -r <REF.TXT>\n";
- flag = true;
- }
- if (flag || conf->count("help")) {
- cerr << dcmdline_options << endl;
- exit(1);
- }
-}
-
-bool ReadSparseVectorString(const string& s, SparseVector<double>* v) {
-#if 0
- // this should work, but untested.
- std::istringstream i(s);
- i>>*v;
-#else
- vector<string> fields;
- Tokenize(s, ';', &fields);
- if (fields.empty()) return false;
- for (unsigned i = 0; i < fields.size(); ++i) {
- vector<string> pair(2);
- Tokenize(fields[i], '=', &pair);
- if (pair.size() != 2) {
- cerr << "Error parsing vector string: " << fields[i] << endl;
- return false;
- }
- v->set_value(FD::Convert(pair[0]), atof(pair[1].c_str()));
- }
- return true;
-#endif
-}
-
-int main(int argc, char** argv) {
- po::variables_map conf;
- InitCommandLine(argc, argv, &conf);
- const string evaluation_metric = conf["evaluation_metric"].as<string>();
- EvaluationMetric* metric = EvaluationMetric::Instance(evaluation_metric);
- DocumentScorer ds(metric, conf["reference"].as<vector<string> >());
- cerr << "Loaded " << ds.size() << " references for scoring with " << evaluation_metric << endl;
- Hypergraph hg;
- string last_file;
- ReadFile in_read(conf["input"].as<string>());
- istream &in=*in_read.stream();
- while(in) {
- string line;
- getline(in, line);
- if (line.empty()) continue;
- istringstream is(line);
- int sent_id;
- string file, s_origin, s_direction;
- // path-to-file (JSON) sent_ed starting-point search-direction
- is >> file >> sent_id >> s_origin >> s_direction;
- SparseVector<double> origin;
- ReadSparseVectorString(s_origin, &origin);
- SparseVector<double> direction;
- ReadSparseVectorString(s_direction, &direction);
- // cerr << "File: " << file << "\nDir: " << direction << "\n X: " << origin << endl;
- if (last_file != file) {
- last_file = file;
- ReadFile rf(file);
- HypergraphIO::ReadFromJSON(rf.stream(), &hg);
- }
- const ConvexHullWeightFunction wf(origin, direction);
- const ConvexHull hull = Inside<ConvexHull, ConvexHullWeightFunction>(hg, NULL, wf);
-
- ErrorSurface es;
- ComputeErrorSurface(*ds[sent_id], hull, &es, metric, hg);
- //cerr << "Viterbi envelope has " << ve.size() << " segments\n";
- // cerr << "Error surface has " << es.size() << " segments\n";
- string val;
- es.Serialize(&val);
- cout << 'M' << ' ' << s_origin << ' ' << s_direction << '\t';
- B64::b64encode(val.c_str(), val.size(), &cout);
- cout << endl << flush;
- }
- return 0;
-}