[SPARK-1743][MLLIB] add loadLibSVMFile and saveAsLibSVMFile to pyspark

Make loading/saving labeled data easier for pyspark users.

Also changed type check in `SparseVector` to allow numpy integers.

Author: Xiangrui Meng <meng@databricks.com>

Closes #672 from mengxr/pyspark-mllib-util and squashes the following commits:

2943fa7 [Xiangrui Meng] format docs
d61668d [Xiangrui Meng] add loadLibSVMFile and saveAsLibSVMFile to pyspark

1 files changed, 1 insertions, 2 deletions

diff --git a/python/pyspark/mllib/linalg.py b/python/pyspark/mllib/linalg.py
index 0aa3a51de7..7511ca7573 100644
--- a/python/pyspark/mllib/linalg.py
+++ b/python/pyspark/mllib/linalg.py
@@ -49,8 +49,7 @@ class SparseVector(object):
         >>> print SparseVector(4, [1, 3], [1.0, 5.5])
         [1: 1.0, 3: 5.5]
         """
-        assert type(size) == int, "first argument must be an int"
-        self.size = size
+        self.size = int(size)
         assert 1 <= len(args) <= 2, "must pass either 2 or 3 arguments"
         if len(args) == 1:
             pairs = args[0]