[SPARK-6262][MLLIB]Implement missing methods for MultivariateStatisticalSummary

Add below methods in pyspark for MultivariateStatisticalSummary
- normL1
- normL2

Author: lewuathe <[email protected]>

Closes apache#5359 from Lewuathe/SPARK-6262 and squashes the following commits:

cbe439e [lewuathe] Implement missing methods for MultivariateStatisticalSummary

Author: Lewuathe

python/pyspark/mllib/stat/_statistics.py

@@ -49,6 +49,12 @@ def max(self):
     def min(self):
         return self.call("min").toArray()
 
+    def normL1(self):
+        return self.call("normL1").toArray()
+
+    def normL2(self):
+        return self.call("normL2").toArray()
+
 
 class Statistics(object):
 

python/pyspark/mllib/tests.py

@@ -357,6 +357,12 @@ def test_col_with_different_rdds(self):
         summary = Statistics.colStats(data)
         self.assertEqual(10, summary.count())
 
+    def test_col_norms(self):
+        data = RandomRDDs.normalVectorRDD(self.sc, 1000, 10, 10)
+        summary = Statistics.colStats(data)
+        self.assertEqual(10, len(summary.normL1()))
+        self.assertEqual(10, len(summary.normL2()))
+
 
 class VectorUDTTests(PySparkTestCase):