summaryrefslogtreecommitdiff
path: root/dtrain/hstreaming/hadoop-streaming-job.sh
diff options
context:
space:
mode:
authorPatrick Simianer <simianer@cl.uni-heidelberg.de>2011-09-27 00:10:48 +0200
committerPatrick Simianer <simianer@cl.uni-heidelberg.de>2011-09-27 00:10:48 +0200
commit40c7d43f8cd5ccaef22f098937b985d4101368f9 (patch)
tree6a039583398f18b4ec1295bba156e62078e53cb3 /dtrain/hstreaming/hadoop-streaming-job.sh
parent8bb00a2a2775442418f1cb7c041f7cba5d6e0d42 (diff)
some updates for hadoop streaming
Diffstat (limited to 'dtrain/hstreaming/hadoop-streaming-job.sh')
-rwxr-xr-xdtrain/hstreaming/hadoop-streaming-job.sh11
1 files changed, 5 insertions, 6 deletions
diff --git a/dtrain/hstreaming/hadoop-streaming-job.sh b/dtrain/hstreaming/hadoop-streaming-job.sh
index 2cf3f50a..9da0a6c3 100755
--- a/dtrain/hstreaming/hadoop-streaming-job.sh
+++ b/dtrain/hstreaming/hadoop-streaming-job.sh
@@ -5,16 +5,15 @@ JAR=contrib/streaming/hadoop-streaming-0.20.2-cdh3u1.jar
HSTREAMING="$HADOOP_HOME/bin/hadoop jar $HADOOP_HOME/$JAR"
IN=in/nc-wmt11-de-en-dyer-cs-joshua.tok.lc.fixamp1.loo.psg.dtrain.1400m
-OUT=out/nc-wmt11-de-en-dyer-cs-joshua.tok.lc.fixamp1.loo.psg.dtrain-weights-1400m
+OUT=out/nc-wmt11-de-en-dyer-cs-joshua.tok.lc.fixamp1.loo.psg.dtrain-weights-1400m-NEW
$HSTREAMING \
- -mapper "dtrain.sh" \
- -reducer "avgweights.rb" \
+ -mapper "dtrain -c dtrain.ini --hstreaming" \
+ -reducer "red-avg.rb" \
-input $IN \
-output $OUT \
- -file avgweights.rb \
- -file dtrain.sh \
- -file dtrain \
+ -file red-avg.rb \
+ -file ../dtrain \
-file dtrain.ini \
-file cdec.ini \
-file nc-wmt11.en.srilm.3.gz \