diff options
author | Chris Dyer <cdyer@allegro.clab.cs.cmu.edu> | 2012-11-18 13:35:42 -0500 |
---|---|---|
committer | Chris Dyer <cdyer@allegro.clab.cs.cmu.edu> | 2012-11-18 13:35:42 -0500 |
commit | 1b8181bf0d6e9137e6b9ccdbe414aec37377a1a9 (patch) | |
tree | 33e5f3aa5abff1f41314cf8f6afbd2c2c40e4bfd /dpmert/mr_dpmert_reduce.cc | |
parent | 7c4665949fb93fb3de402e4ce1d19bef67850d05 (diff) |
major restructure of the training code
Diffstat (limited to 'dpmert/mr_dpmert_reduce.cc')
-rw-r--r-- | dpmert/mr_dpmert_reduce.cc | 77 |
1 files changed, 0 insertions, 77 deletions
diff --git a/dpmert/mr_dpmert_reduce.cc b/dpmert/mr_dpmert_reduce.cc deleted file mode 100644 index 31512a03..00000000 --- a/dpmert/mr_dpmert_reduce.cc +++ /dev/null @@ -1,77 +0,0 @@ -#include <sstream> -#include <iostream> -#include <fstream> -#include <vector> - -#include <boost/program_options.hpp> -#include <boost/program_options/variables_map.hpp> - -#include "sparse_vector.h" -#include "error_surface.h" -#include "line_optimizer.h" -#include "b64tools.h" -#include "stringlib.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() - ("evaluation_metric,m",po::value<string>(), "Evaluation metric (IBM_BLEU, etc.)") - ("help,h", "Help"); - po::options_description dcmdline_options; - dcmdline_options.add(opts); - po::store(parse_command_line(argc, argv, dcmdline_options), *conf); - bool flag = conf->count("evaluation_metric") == 0; - if (flag || conf->count("help")) { - cerr << dcmdline_options << endl; - exit(1); - } -} - -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); - LineOptimizer::ScoreType opt_type = LineOptimizer::MAXIMIZE_SCORE; - if (metric->IsErrorMetric()) - opt_type = LineOptimizer::MINIMIZE_SCORE; - - vector<ErrorSurface> esv; - string last_key, line, key, val; - while(getline(cin, line)) { - size_t ks = line.find("\t"); - assert(string::npos != ks); - assert(ks > 2); - key = line.substr(2, ks - 2); - val = line.substr(ks + 1); - if (key != last_key) { - if (!last_key.empty()) { - float score; - double x = LineOptimizer::LineOptimize(metric, esv, opt_type, &score); - cout << last_key << "|" << x << "|" << score << endl; - } - last_key.swap(key); - esv.clear(); - } - if (val.size() % 4 != 0) { - cerr << "B64 encoding error 1! Skipping.\n"; - continue; - } - string encoded(val.size() / 4 * 3, '\0'); - if (!B64::b64decode(reinterpret_cast<const unsigned char*>(&val[0]), val.size(), &encoded[0], encoded.size())) { - cerr << "B64 encoding error 2! Skipping.\n"; - continue; - } - esv.push_back(ErrorSurface()); - esv.back().Deserialize(encoded); - } - if (!esv.empty()) { - float score; - double x = LineOptimizer::LineOptimize(metric, esv, opt_type, &score); - cout << last_key << "|" << x << "|" << score << endl; - } - return 0; -} |