SPARK-1426: Make MLlib work with NumPy versions older than 1.7

Currently it requires NumPy 1.7 due to using the copyto method (http://docs.scipy.org/doc/numpy/reference/generated/numpy.copyto.html) for extracting data out of an array.
Replace it with a fallback

Author: Sandeep <sandeep@techaddict.me>

Closes #391 from techaddict/1426 and squashes the following commits:

d365962 [Sandeep] SPARK-1426: Make MLlib work with NumPy versions older than 1.7 Currently it requires NumPy 1.7 due to using the copyto method (http://docs.scipy.org/doc/numpy/reference/generated/numpy.copyto.html) for extracting data out of an array. Replace it with a fallback

4 files changed, 16 insertions, 16 deletions

diff --git a/docs/mllib-guide.md b/docs/mllib-guide.md
index a5e0cc5080..eff856104c 100644
--- a/docs/mllib-guide.md
+++ b/docs/mllib-guide.md
@@ -11,7 +11,7 @@ namely, binary classification, regression, clustering and collaborative
 filtering, as well as an underlying gradient descent optimization primitive.
 
 # Available Methods
-The following links provide a detailed explanation of the methods and usage examples for each of them: 
+The following links provide a detailed explanation of the methods and usage examples for each of them:
 
 * <a href="mllib-classification-regression.html">Classification and Regression</a>
   * Binary Classification
@@ -33,10 +33,9 @@ The following links provide a detailed explanation of the methods and usage exam
 
 # Dependencies
 MLlib uses the [jblas](https://github.com/mikiobraun/jblas) linear algebra library, which itself
-depends on native Fortran routines. You may need to install the 
+depends on native Fortran routines. You may need to install the
 [gfortran runtime library](https://github.com/mikiobraun/jblas/wiki/Missing-Libraries)
-if it is not already present on your nodes. MLlib will throw a linking error if it cannot 
+if it is not already present on your nodes. MLlib will throw a linking error if it cannot
 detect these libraries automatically.
 
-To use MLlib in Python, you will need [NumPy](http://www.numpy.org) version 1.7 or newer.
-
+To use MLlib in Python, you will need [NumPy](http://www.numpy.org) version 1.4 or newer.
diff --git a/docs/python-programming-guide.md b/docs/python-programming-guide.md
index 888631e702..39de603b29 100644
--- a/docs/python-programming-guide.md
+++ b/docs/python-programming-guide.md
@@ -100,8 +100,8 @@ $ MASTER=local[4] ./bin/pyspark
 
 ## IPython
 
-It is also possible to launch PySpark in [IPython](http://ipython.org), the 
-enhanced Python interpreter. PySpark works with IPython 1.0.0 and later. To 
+It is also possible to launch PySpark in [IPython](http://ipython.org), the
+enhanced Python interpreter. PySpark works with IPython 1.0.0 and later. To
 use IPython, set the `IPYTHON` variable to `1` when running `bin/pyspark`:
 
 {% highlight bash %}
@@ -153,7 +153,7 @@ Many of the methods also contain [doctests](http://docs.python.org/2/library/doc
 # Libraries
 
 [MLlib](mllib-guide.html) is also available in PySpark. To use it, you'll need
-[NumPy](http://www.numpy.org) version 1.7 or newer. The [MLlib guide](mllib-guide.html) contains
+[NumPy](http://www.numpy.org) version 1.4 or newer. The [MLlib guide](mllib-guide.html) contains
 some example applications.
 
 # Where to Go from Here
diff --git a/python/pyspark/mllib/__init__.py b/python/pyspark/mllib/__init__.py
index 538ff26ce7..4149f54931 100644
--- a/python/pyspark/mllib/__init__.py
+++ b/python/pyspark/mllib/__init__.py
@@ -19,8 +19,8 @@
 Python bindings for MLlib.
 """
 
-# MLlib currently needs and NumPy 1.7+, so complain if lower
+# MLlib currently needs and NumPy 1.4+, so complain if lower
 
 import numpy
-if numpy.version.version < '1.7':
-    raise Exception("MLlib requires NumPy 1.7+")
+if numpy.version.version < '1.4':
+    raise Exception("MLlib requires NumPy 1.4+")
diff --git a/python/pyspark/mllib/_common.py b/python/pyspark/mllib/_common.py
index 7ef251d24c..e19f5d2aaa 100644
--- a/python/pyspark/mllib/_common.py
+++ b/python/pyspark/mllib/_common.py
@@ -15,7 +15,7 @@
 # limitations under the License.
 #
 
-from numpy import ndarray, copyto, float64, int64, int32, ones, array_equal, array, dot, shape, complex, issubdtype
+from numpy import ndarray, float64, int64, int32, ones, array_equal, array, dot, shape, complex, issubdtype
 from pyspark import SparkContext, RDD
 import numpy as np
 
@@ -72,8 +72,8 @@ def _serialize_double_vector(v):
     header = ndarray(shape=[2], buffer=ba, dtype="int64")
     header[0] = 1
     header[1] = length
-    copyto(ndarray(shape=[length], buffer=ba, offset=16,
-            dtype="float64"), v)
+    arr_mid = ndarray(shape=[length], buffer=ba, offset=16, dtype="float64")
+    arr_mid[...] = v
     return ba
 
 def _deserialize_double_vector(ba):
@@ -112,8 +112,9 @@ def _serialize_double_matrix(m):
         header[0] = 2
         header[1] = rows
         header[2] = cols
-        copyto(ndarray(shape=[rows, cols], buffer=ba, offset=24,
-                       dtype="float64", order='C'), m)
+        arr_mid = ndarray(shape=[rows, cols], buffer=ba, offset=24,
+                      dtype="float64", order='C')
+        arr_mid[...] = m
         return ba
     else:
         raise TypeError("_serialize_double_matrix called on a "