From 671c21451542e2dd20e45b4033d44d8e8735f87b Mon Sep 17 00:00:00 2001
From: Chris Dyer <redpony@gmail.com>
Date: Thu, 3 Dec 2009 16:33:55 -0500
Subject: initial check in

---
 vest/mr_vest_reduce.cc | 80 ++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 80 insertions(+)
 create mode 100644 vest/mr_vest_reduce.cc

(limited to 'vest/mr_vest_reduce.cc')

diff --git a/vest/mr_vest_reduce.cc b/vest/mr_vest_reduce.cc
new file mode 100644
index 00000000..c1347065
--- /dev/null
+++ b/vest/mr_vest_reduce.cc
@@ -0,0 +1,80 @@
+#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 "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()
+        ("loss_function,l",po::value<string>()->default_value("ibm_bleu"), "Loss function being optimized")
+        ("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 (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 loss_function = conf["loss_function"].as<string>();
+  ScoreType type = ScoreTypeFromString(loss_function);
+  LineOptimizer::ScoreType opt_type = LineOptimizer::MAXIMIZE_SCORE;
+  if (type == TER)
+    opt_type = LineOptimizer::MINIMIZE_SCORE;
+  string last_key;
+  vector<ErrorSurface> esv;
+  while(cin) {
+    string line;
+    getline(cin, line);
+    if (line.empty()) continue;
+    size_t ks = line.find("\t");
+    assert(string::npos != ks);
+    assert(ks > 2);
+    string key = line.substr(2, ks - 2);
+    string val = line.substr(ks + 1);
+    if (key != last_key) {
+      if (!last_key.empty()) {
+	float score;
+        double x = LineOptimizer::LineOptimize(esv, opt_type, &score);
+	cout << last_key << "|" << x << "|" << score << endl;
+      }
+      last_key = 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(type, encoded);
+  }
+  if (!esv.empty()) {
+        cerr << "ESV=" << esv.size() << endl;
+        for (int i = 0; i < esv.size(); ++i) { cerr << esv[i].size() << endl; }
+    float score;
+    double x = LineOptimizer::LineOptimize(esv, opt_type, &score);
+    cout << last_key << "|" << x << "|" << score << endl;
+  }
+  return 0;
+}
-- 
cgit v1.2.3