[SPARK-6263][MLLIB] Python MLlib API missing items: Utils

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

@@ -74,6 +74,15 @@ private[python] class PythonMLLibAPI extends Serializable {
       minPartitions: Int): JavaRDD[LabeledPoint] =
     MLUtils.loadLabeledPoints(jsc.sc, path, minPartitions)
 
+  /**
+   * Loads and serializes vectors saved with `RDD#saveAsTextFile`.
+   * @param jsc Java SparkContext
+   * @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] =
+    MLUtils.loadVectors(jsc.sc, path)
+
   private def trainRegressionModel(
       learner: GeneralizedLinearAlgorithm[_ <: GeneralizedLinearModel],
       data: JavaRDD[LabeledPoint],

python/pyspark/mllib/tests.py

@@ -24,7 +24,6 @@
 import tempfile
 import array as pyarray
 from time import time, sleep
-
 from numpy import array, array_equal, zeros, inf, all, random
 from numpy import sum as array_sum
 from py4j.protocol import Py4JJavaError
@@ -50,6 +49,7 @@
 from pyspark.mllib.feature import IDF
 from pyspark.mllib.feature import StandardScaler
 from pyspark.mllib.feature import ElementwiseProduct
+from pyspark.mllib.util import MLUtils
 from pyspark.serializers import PickleSerializer
 from pyspark.streaming import StreamingContext
 from pyspark.sql import SQLContext
@@ -1011,6 +1011,48 @@ def collect(rdd):
         self.assertEqual(predict_results, [[0, 1, 1], [1, 0, 1]])
 
 
+class MLUtilsTests(MLlibTestCase):
+    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), DenseVector)
+
+    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), DenseVector)
+
+    def test_append_bias_with_sp_vector(self):
+        data = Vectors.sparse(3, {0: 2.0, 2: 2.0})
+        expected = Vectors.sparse(4, {0: 2.0, 2: 2.0, 3: 1.0})
+        # Returned value must be SparseVector
+        ret = MLUtils.appendBias(data)
+        self.assertEqual(ret, expected)
+        self.assertEqual(type(ret), SparseVector)
+
+    def test_load_vectors(self):
+        import shutil
+        data = [
+            [1.0, 2.0, 3.0],
+            [1.0, 2.0, 3.0]
+        ]
+        temp_dir = tempfile.mkdtemp()
+        load_vectors_path = os.path.join(temp_dir, "test_load_vectors")
+        try:
+            self.sc.parallelize(data).saveAsTextFile(load_vectors_path)
+            ret_rdd = MLUtils.loadVectors(self.sc, load_vectors_path)
+            ret = ret_rdd.collect()
+            self.assertEqual(len(ret), 2)
+            self.assertEqual(ret[0], DenseVector([1.0, 2.0, 3.0]))
+            self.assertEqual(ret[1], DenseVector([1.0, 2.0, 3.0]))
+        except:
+            self.fail()
+        finally:
+            shutil.rmtree(load_vectors_path)
+
+
 if __name__ == "__main__":
     if not _have_scipy:
         print("NOTE: Skipping SciPy tests as it does not seem to be installed")

python/pyspark/mllib/util.py

@@ -18,12 +18,18 @@
 import sys
 import numpy as np
 import warnings
+try:
+    import scipy.sparse
+    _have_scipy = True
+except:
+    # No SciPy in environment, but that's okay
+    _have_scipy = False
 
 if sys.version > '3':
     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):
@@ -169,6 +175,28 @@ def loadLabeledPoints(sc, path, minPartitions=None):
         minPartitions = minPartitions or min(sc.defaultParallelism, 2)
         return callMLlibFunc("loadLabeledPoints", sc, path, minPartitions)
 
+    @staticmethod
+    def appendBias(data):
+        """
+        Returns a new vector with `1.0` (bias) appended to
+        the end of the input vector.
+        """
+        vec = _convert_to_vector(data)
+        if isinstance(vec, SparseVector):
+            l = scipy.sparse.csc_matrix(np.append(vec.toArray(), 1.0))
+            return _convert_to_vector(l.T)
+        elif isinstance(vec, Vector):
+            vec = vec.toArray()
+        return _convert_to_vector(np.append(vec, 1.0).tolist())
+
+    @staticmethod
+    def loadVectors(sc, path):
+        """
+        Loads vectors saved using `RDD[Vector].saveAsTextFile`
+        with the default number of partitions.
+        """
+        return callMLlibFunc("loadVectors", sc, path)
+
 
 class Saveable(object):
     """