[SPARK-10028][MLLIB][PYTHON] Add Python API for PrefixSpan

Author: Yu ISHIKAWA <yuu.ishikawa@gmail.com>

Closes #9469 from yu-iskw/SPARK-10028.

1 files changed, 68 insertions, 1 deletions

diff --git a/python/pyspark/mllib/fpm.py b/python/pyspark/mllib/fpm.py
index bdabba9602..2039decc0c 100644
--- a/python/pyspark/mllib/fpm.py
+++ b/python/pyspark/mllib/fpm.py
@@ -23,7 +23,7 @@ from pyspark import SparkContext, since
 from pyspark.rdd import ignore_unicode_prefix
 from pyspark.mllib.common import JavaModelWrapper, callMLlibFunc, inherit_doc
 
-__all__ = ['FPGrowth', 'FPGrowthModel']
+__all__ = ['FPGrowth', 'FPGrowthModel', 'PrefixSpan', 'PrefixSpanModel']
 
 
 @inherit_doc
@@ -85,6 +85,73 @@ class FPGrowth(object):
         """
 
 
+@inherit_doc
+@ignore_unicode_prefix
+class PrefixSpanModel(JavaModelWrapper):
+    """
+    .. note:: Experimental
+
+    Model fitted by PrefixSpan
+
+    >>> data = [
+    ...    [["a", "b"], ["c"]],
+    ...    [["a"], ["c", "b"], ["a", "b"]],
+    ...    [["a", "b"], ["e"]],
+    ...    [["f"]]]
+    >>> rdd = sc.parallelize(data, 2)
+    >>> model = PrefixSpan.train(rdd)
+    >>> sorted(model.freqSequences().collect())
+    [FreqSequence(sequence=[[u'a']], freq=3), FreqSequence(sequence=[[u'a'], [u'a']], freq=1), ...
+
+    .. versionadded:: 1.6.0
+    """
+
+    @since("1.6.0")
+    def freqSequences(self):
+        """Gets frequence sequences"""
+        return self.call("getFreqSequences").map(lambda x: PrefixSpan.FreqSequence(x[0], x[1]))
+
+
+class PrefixSpan(object):
+    """
+    .. note:: Experimental
+
+    A parallel PrefixSpan algorithm to mine frequent sequential patterns.
+    The PrefixSpan algorithm is described in J. Pei, et al., PrefixSpan:
+    Mining Sequential Patterns Efficiently by Prefix-Projected Pattern Growth
+    ([[http://doi.org/10.1109/ICDE.2001.914830]]).
+
+    .. versionadded:: 1.6.0
+    """
+
+    @classmethod
+    @since("1.6.0")
+    def train(cls, data, minSupport=0.1, maxPatternLength=10, maxLocalProjDBSize=32000000):
+        """
+        Finds the complete set of frequent sequential patterns in the input sequences of itemsets.
+
+        :param data: The input data set, each element contains a sequnce of itemsets.
+        :param minSupport: the minimal support level of the sequential pattern, any pattern appears
+            more than  (minSupport * size-of-the-dataset) times will be output (default: `0.1`)
+        :param maxPatternLength: the maximal length of the sequential pattern, any pattern appears
+            less than maxPatternLength will be output. (default: `10`)
+        :param maxLocalProjDBSize: The maximum number of items (including delimiters used in
+            the internal storage format) allowed in a projected database before local
+            processing. If a projected database exceeds this size, another
+            iteration of distributed prefix growth is run. (default: `32000000`)
+        """
+        model = callMLlibFunc("trainPrefixSpanModel",
+                              data, minSupport, maxPatternLength, maxLocalProjDBSize)
+        return PrefixSpanModel(model)
+
+    class FreqSequence(namedtuple("FreqSequence", ["sequence", "freq"])):
+        """
+        Represents a (sequence, freq) tuple.
+
+        .. versionadded:: 1.6.0
+        """
+
+
 def _test():
     import doctest
     import pyspark.mllib.fpm