From 8aa29810bb77611cc20b7a384897ff6703783ea1 Mon Sep 17 00:00:00 2001 From: Chris Dyer Date: Sun, 18 Nov 2012 13:35:42 -0500 Subject: major restructure of the training code --- dpmert/mr_dpmert_map.cc | 112 ------------------------------------------------ 1 file changed, 112 deletions(-) delete mode 100644 dpmert/mr_dpmert_map.cc (limited to 'dpmert/mr_dpmert_map.cc') 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 -#include -#include -#include - -#include -#include - -#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 >(), "[REQD] Reference translation (tokenized text)") - ("source,s",po::value(), "Source file (ignored, except for AER)") - ("evaluation_metric,m",po::value()->default_value("ibm_bleu"), "Evaluation metric being optimized") - ("input,i",po::value()->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 \n"; - flag = true; - } - if (flag || conf->count("help")) { - cerr << dcmdline_options << endl; - exit(1); - } -} - -bool ReadSparseVectorString(const string& s, SparseVector* v) { -#if 0 - // this should work, but untested. - std::istringstream i(s); - i>>*v; -#else - vector fields; - Tokenize(s, ';', &fields); - if (fields.empty()) return false; - for (unsigned i = 0; i < fields.size(); ++i) { - vector 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(); - EvaluationMetric* metric = EvaluationMetric::Instance(evaluation_metric); - DocumentScorer ds(metric, conf["reference"].as >()); - cerr << "Loaded " << ds.size() << " references for scoring with " << evaluation_metric << endl; - Hypergraph hg; - string last_file; - ReadFile in_read(conf["input"].as()); - 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 origin; - ReadSparseVectorString(s_origin, &origin); - SparseVector 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(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; -} -- cgit v1.2.3