You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mahout.apache.org by sm...@apache.org on 2013/12/04 07:13:55 UTC

svn commit: r1547716 - /mahout/trunk/examples/bin/cluster-reuters.sh

Author: smarthi
Date: Wed Dec  4 06:13:54 2013
New Revision: 1547716

URL: http://svn.apache.org/r1547716
Log:
MAHOUT-1347: Added 'qualcluster' utility to get stats about quality of Streaming KMeans clustering

Modified:
    mahout/trunk/examples/bin/cluster-reuters.sh

Modified: mahout/trunk/examples/bin/cluster-reuters.sh
URL: http://svn.apache.org/viewvc/mahout/trunk/examples/bin/cluster-reuters.sh?rev=1547716&r1=1547715&r2=1547716&view=diff
==============================================================================
--- mahout/trunk/examples/bin/cluster-reuters.sh (original)
+++ mahout/trunk/examples/bin/cluster-reuters.sh Wed Dec  4 06:13:54 2013
@@ -185,8 +185,12 @@ elif [ "x$clustertype" == "xstreamingkme
     -dm org.apache.mahout.common.distance.SquaredEuclideanDistanceMeasure \
     -k 20 -km 200 -rskm -ow \
   && \
-  $MAHOUT seqdumper \
-    -i ${WORK_DIR}/reuters-streamingkmeans
+  $MAHOUT qualcluster \
+    -i ${WORK_DIR}/reuters-out-seqdir-sparse-streamingkmeans/tfidf-vectors/part-r-00000 \
+    -c ${WORK_DIR}/reuters-streamingkmeans/part-r-00000   \
+    -o ${WORK_DIR}/reuters-cluster-distance.csv \
+    && \
+  cat ${WORK_DIR}/reuters-cluster-distance.csv
 else 
   echo "unknown cluster type: $clustertype"
 fi