initial checkin

git-svn-id: https://ws10smt.googlecode.com/svn/trunk@2 ec762483-ff6d-05da-a07a-a48fb63a330f

1 files changed, 11 insertions, 0 deletions

diff --git a/extools/simple-extract.sh b/extools/simple-extract.sh
new file mode 100755
index 00000000..7d9f439d
--- /dev/null
+++ b/extools/simple-extract.sh
@@ -0,0 +1,11 @@
+#!/bin/bash
+
+export LANG=C
+date
+./extractor -i $1 -d X -c 500000 -L 12 -b | sort -t $'\t' -k 1 | gzip > ex.output.gz
+date
+# -p = compute phrase marginals
+# -b = bidirectional rules (starting with F or E) were extracted
+gzcat ex.output.gz | ./mr_stripe_rule_reduce -p -b | sort -t $'\t' -k 1 | ./mr_stripe_rule_reduce | gzip > phrase-table.gz
+date
+