From 38a5bee71f6b49515cd105a9467ff602ff9dee64 Mon Sep 17 00:00:00 2001
From: Chris Dyer <cdyer@cs.cmu.edu>
Date: Tue, 13 Sep 2011 13:25:46 +0100
Subject: optional support for doing perfect hashing of feature strings to save
 lots of memory

---
 utils/phmt.cc | 44 ++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 44 insertions(+)
 create mode 100644 utils/phmt.cc

(limited to 'utils/phmt.cc')

diff --git a/utils/phmt.cc b/utils/phmt.cc
new file mode 100644
index 00000000..1f59afaf
--- /dev/null
+++ b/utils/phmt.cc
@@ -0,0 +1,44 @@
+#include "config.h"
+
+#ifndef HAVE_CMPH
+int main() {
+  return 0;
+}
+#else
+
+#include <iostream>
+#include "weights.h"
+#include "fdict.h"
+
+using namespace std;
+
+int main(int argc, char** argv) {
+  if (argc != 2) { cerr << "Usage: " << argv[0] << " file.mphf\n"; return 1; }
+  FD::EnableHash(argv[1]);
+  cerr << "Number of keys: " << FD::NumFeats() << endl;
+  cerr << "LexFE = " << FD::Convert("LexFE") << endl;
+  cerr << "LexEF = " << FD::Convert("LexEF") << endl;
+  {
+    Weights w;
+    vector<weight_t> v(FD::NumFeats());
+    v[FD::Convert("LexFE")] = 1.0;
+    v[FD::Convert("LexEF")] = 0.5;
+    w.InitFromVector(v);
+    cerr << "Writing...\n";
+    w.WriteToFile("weights.bin");
+    cerr << "Done.\n";
+  }
+  {
+    Weights w;
+    vector<weight_t> v(FD::NumFeats());
+    cerr << "Reading...\n";
+    w.InitFromFile("weights.bin");
+    cerr << "Done.\n";
+    w.InitVector(&v);
+    assert(v[FD::Convert("LexFE")] == 1.0);
+    assert(v[FD::Convert("LexEF")] == 0.5);
+  }
+}
+
+#endif
+
-- 
cgit v1.2.3