You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@systemml.apache.org by ac...@apache.org on 2017/09/11 20:39:02 UTC
systemml git commit: [MINOR] Update kmeans' number of centroids in perf scripts

Repository: systemml
Updated Branches:
  refs/heads/master 918e57937 -> 687e19c55


[MINOR] Update kmeans' number of centroids in perf scripts


Project: http://git-wip-us.apache.org/repos/asf/systemml/repo
Commit: http://git-wip-us.apache.org/repos/asf/systemml/commit/687e19c5
Tree: http://git-wip-us.apache.org/repos/asf/systemml/tree/687e19c5
Diff: http://git-wip-us.apache.org/repos/asf/systemml/diff/687e19c5

Branch: refs/heads/master
Commit: 687e19c5529b371f937a9ecb60075373f36fd4d6
Parents: 918e579
Author: Arvind Surve <ac...@yahoo.com>
Authored: Mon Sep 11 13:38:18 2017 -0700
Committer: Arvind Surve <ac...@yahoo.com>
Committed: Mon Sep 11 13:38:18 2017 -0700

----------------------------------------------------------------------
 scripts/perftest/genClusteringData.sh | 10 +++++-----
 scripts/perftest/runKmeans.sh         |  2 +-
 2 files changed, 6 insertions(+), 6 deletions(-)
----------------------------------------------------------------------


http://git-wip-us.apache.org/repos/asf/systemml/blob/687e19c5/scripts/perftest/genClusteringData.sh
----------------------------------------------------------------------
diff --git a/scripts/perftest/genClusteringData.sh b/scripts/perftest/genClusteringData.sh
index f8895b5..2e7463f 100755
--- a/scripts/perftest/genClusteringData.sh
+++ b/scripts/perftest/genClusteringData.sh
@@ -31,22 +31,22 @@ SPARSE_SP=0.01
 export HADOOP_CLIENT_OPTS="-Xmx2048m -Xms2048m -Xmn256m"
 
 #generate XS scenarios (80MB)
-${CMD} -f ../datagen/genRandData4Kmeans.dml $DASH-nvargs nr=10000 nf=1000 nc=50 dc=10.0 dr=1.0 fbf=100.0 cbf=100.0 X=$BASE/X10k_1k_dense C=$BASE/C10k_1k_dense Y=$BASE/y10k_1k_dense YbyC=$BASE/YbyC10k_1k_dense fmt=$FORMAT
+${CMD} -f ../datagen/genRandData4Kmeans.dml $DASH-nvargs nr=10000 nf=1000 nc=5 dc=10.0 dr=1.0 fbf=100.0 cbf=100.0 X=$BASE/X10k_1k_dense C=$BASE/C10k_1k_dense Y=$BASE/y10k_1k_dense YbyC=$BASE/YbyC10k_1k_dense fmt=$FORMAT
 ${CMD} -f extractTestData.dml $DASH-args $BASE/X10k_1k_dense $BASE/y10k_1k_dense $BASE/X10k_1k_dense_test $BASE/y10k_1k_dense_test $FORMAT
 
 #generate S scenarios (800MB)
-#${CMD} -f ../datagen/genRandData4Kmeans.dml $DASH-nvargs nr=100000 nf=1000 nc=50 dc=10.0 dr=1.0 fbf=100.0 cbf=100.0 X=$BASE/X100k_1k_dense C=$BASE/C100k_1k_dense Y=$BASE/y100k_1k_dense YbyC=$BASE/YbyC100k_1k_dense fmt=$FORMAT
+#${CMD} -f ../datagen/genRandData4Kmeans.dml $DASH-nvargs nr=100000 nf=1000 nc=5 dc=10.0 dr=1.0 fbf=100.0 cbf=100.0 X=$BASE/X100k_1k_dense C=$BASE/C100k_1k_dense Y=$BASE/y100k_1k_dense YbyC=$BASE/YbyC100k_1k_dense fmt=$FORMAT
 #${CMD} -f extractTestData.dml $DASH-args $BASE/X100k_1k_dense $BASE/y100k_1k_dense $BASE/X100k_1k_dense_test $BASE/y100k_1k_dense_test $FORMAT
 
 #generate M scenarios (8GB)
-#${CMD} -f ../datagen/genRandData4Kmeans.dml $DASH-nvargs nr=1000000 nf=1000 nc=50 dc=10.0 dr=1.0 fbf=100.0 cbf=100.0 X=$BASE/X1M_1k_dense C=$BASE/C1M_1k_dense Y=$BASE/y1M_1k_dense YbyC=$BASE/YbyC1M_1k_dense fmt=$FORMAT
+#${CMD} -f ../datagen/genRandData4Kmeans.dml $DASH-nvargs nr=1000000 nf=1000 nc=5 dc=10.0 dr=1.0 fbf=100.0 cbf=100.0 X=$BASE/X1M_1k_dense C=$BASE/C1M_1k_dense Y=$BASE/y1M_1k_dense YbyC=$BASE/YbyC1M_1k_dense fmt=$FORMAT
 #${CMD} -f extractTestData.dml $DASH-args $BASE/X1M_1k_dense $BASE/y1M_1k_dense $BASE/X1M_1k_dense_test $BASE/y1M_1k_dense_test $FORMAT
 
 #generate L scenarios (80GB)
-#${CMD} -f ../datagen/genRandData4Kmeans.dml $DASH-nvargs nr=10000000 nf=1000 nc=50 dc=10.0 dr=1.0 fbf=100.0 cbf=100.0 X=$BASE/X10M_1k_dense C=$BASE/C10M_1k_dense Y=$BASE/y10M_1k_dense YbyC=$BASE/YbyC10M_1k_dense fmt=$FORMAT
+#${CMD} -f ../datagen/genRandData4Kmeans.dml $DASH-nvargs nr=10000000 nf=1000 nc=5 dc=10.0 dr=1.0 fbf=100.0 cbf=100.0 X=$BASE/X10M_1k_dense C=$BASE/C10M_1k_dense Y=$BASE/y10M_1k_dense YbyC=$BASE/YbyC10M_1k_dense fmt=$FORMAT
 #${CMD} -f extractTestData.dml $DASH-args $BASE/X10M_1k_dense $BASE/y10M_1k_dense $BASE/X10M_1k_dense_test $BASE/y10M_1k_dense_test $FORMAT
 
 #generate LARGE scenarios (800GB)
-#${CMD} -f ../datagen/genRandData4Kmeans.dml $DASH-nvargs nr=100000000 nf=1000 nc=50 dc=10.0 dr=1.0 fbf=100.0 cbf=100.0 X=$BASE/X100M_1k_dense C=$BASE/C100M_1k_dense Y=$BASE/y100M_1k_dense YbyC=$BASE/YbyC100M_1k_dense fmt=$FORMAT
+#${CMD} -f ../datagen/genRandData4Kmeans.dml $DASH-nvargs nr=100000000 nf=1000 nc=5 dc=10.0 dr=1.0 fbf=100.0 cbf=100.0 X=$BASE/X100M_1k_dense C=$BASE/C100M_1k_dense Y=$BASE/y100M_1k_dense YbyC=$BASE/YbyC100M_1k_dense fmt=$FORMAT
 #${CMD} -f extractTestData.dml $DASH-args $BASE/X100M_1k_dense $BASE/y100M_1k_dense $BASE/X100M_1k_dense_test $BASE/y100M_1k_dense_test $FORMAT
  

http://git-wip-us.apache.org/repos/asf/systemml/blob/687e19c5/scripts/perftest/runKmeans.sh
----------------------------------------------------------------------
diff --git a/scripts/perftest/runKmeans.sh b/scripts/perftest/runKmeans.sh
index 33b487c..f94f8bb 100755
--- a/scripts/perftest/runKmeans.sh
+++ b/scripts/perftest/runKmeans.sh
@@ -29,7 +29,7 @@ export HADOOP_CLIENT_OPTS="-Xmx2048m -Xms2048m -Xmn256m"
 
 #training
 tstart=$SECONDS
-${CMD} -f ../algorithms/Kmeans.dml $DASH-explain $DASH-stats $DASH-nvargs X=$1 k=50 C=${BASE}/centroids.mtx maxi=$2 tol=0.0001
+${CMD} -f ../algorithms/Kmeans.dml $DASH-explain $DASH-stats $DASH-nvargs X=$1 k=5 C=${BASE}/centroids.mtx maxi=$2 tol=0.0001
 ttrain=$(($SECONDS - $tstart - 3))
 echo "Kmeans train on "$1": "$ttrain >> times.txt