Fix appendBias return type

Author: Lewuathe

mllib/src/main/scala/org/apache/spark/mllib/api/python/PythonMLLibAPI.scala

@@ -77,8 +77,7 @@ private[python] class PythonMLLibAPI extends Serializable {
    * @param path file or directory path in any Hadoop-supported file system URI
    * @return serialized vectors in a RDD
    */
-  def loadVectors(jsc: JavaSparkContext,
-                  path: String): RDD[Vector] =
+  def loadVectors(jsc: JavaSparkContext, path: String): RDD[Vector] =
     MLUtils.loadVectors(jsc.sc, path)
 
   private def trainRegressionModel(

python/pyspark/mllib/tests.py

@@ -824,6 +824,13 @@ def test_append_bias(self):
         data = [2.0, 2.0, 2.0]
         ret = MLUtils.appendBias(data)
         self.assertEqual(ret[3], 1.0)
+        self.assertEqual(type(ret), list)
+
+    def test_append_bias_with_vector(self):
+        data = Vectors.dense([2.0, 2.0, 2.0])
+        ret = MLUtils.appendBias(data)
+        self.assertEqual(ret[3], 1.0)
+        self.assertEqual(type(ret), list)
 
     def test_load_vectors(self):
         import shutil

python/pyspark/mllib/util.py

@@ -23,7 +23,7 @@
     xrange = range
 
 from pyspark.mllib.common import callMLlibFunc, inherit_doc
-from pyspark.mllib.linalg import Vectors, SparseVector, _convert_to_vector
+from pyspark.mllib.linalg import Vector, Vectors, SparseVector, _convert_to_vector
 
 
 class MLUtils(object):
@@ -172,10 +172,13 @@ def loadLabeledPoints(sc, path, minPartitions=None):
     @staticmethod
     def appendBias(data):
         """
-        Returns a new vector with `1.0` (bias) appended to the input vector.
+        Returns a new vector with `1.0` (bias) appended to
+        the end of the input vector.
         """
         vec = _convert_to_vector(data)
-        return np.append(vec, 1.0)
+        if isinstance(vec, Vector):
+            vec = vec.toArray()
+        return np.append(vec, 1.0).tolist()
 
     @staticmethod
     def loadVectors(sc, path):