diff options
Diffstat (limited to 'python/pyspark/mllib/linalg.py')
| -rw-r--r-- | python/pyspark/mllib/linalg.py | 48 |
1 files changed, 28 insertions, 20 deletions
diff --git a/python/pyspark/mllib/linalg.py b/python/pyspark/mllib/linalg.py index a80320c52d..38b3aa3ad4 100644 --- a/python/pyspark/mllib/linalg.py +++ b/python/pyspark/mllib/linalg.py @@ -25,7 +25,13 @@ SciPy is available in their environment. import sys import array -import copy_reg + +if sys.version >= '3': + basestring = str + xrange = range + import copyreg as copy_reg +else: + import copy_reg import numpy as np @@ -57,7 +63,7 @@ except: def _convert_to_vector(l): if isinstance(l, Vector): return l - elif type(l) in (array.array, np.array, np.ndarray, list, tuple): + elif type(l) in (array.array, np.array, np.ndarray, list, tuple, xrange): return DenseVector(l) elif _have_scipy and scipy.sparse.issparse(l): assert l.shape[1] == 1, "Expected column vector" @@ -88,7 +94,7 @@ def _vector_size(v): """ if isinstance(v, Vector): return len(v) - elif type(v) in (array.array, list, tuple): + elif type(v) in (array.array, list, tuple, xrange): return len(v) elif type(v) == np.ndarray: if v.ndim == 1 or (v.ndim == 2 and v.shape[1] == 1): @@ -193,7 +199,7 @@ class DenseVector(Vector): DenseVector([1.0, 0.0]) """ def __init__(self, ar): - if isinstance(ar, basestring): + if isinstance(ar, bytes): ar = np.frombuffer(ar, dtype=np.float64) elif not isinstance(ar, np.ndarray): ar = np.array(ar, dtype=np.float64) @@ -321,11 +327,13 @@ class DenseVector(Vector): __sub__ = _delegate("__sub__") __mul__ = _delegate("__mul__") __div__ = _delegate("__div__") + __truediv__ = _delegate("__truediv__") __mod__ = _delegate("__mod__") __radd__ = _delegate("__radd__") __rsub__ = _delegate("__rsub__") __rmul__ = _delegate("__rmul__") __rdiv__ = _delegate("__rdiv__") + __rtruediv__ = _delegate("__rtruediv__") __rmod__ = _delegate("__rmod__") @@ -344,12 +352,12 @@ class SparseVector(Vector): :param args: Non-zero entries, as a dictionary, list of tupes, or two sorted lists containing indices and values. - >>> print SparseVector(4, {1: 1.0, 3: 5.5}) - (4,[1,3],[1.0,5.5]) - >>> print SparseVector(4, [(1, 1.0), (3, 5.5)]) - (4,[1,3],[1.0,5.5]) - >>> print SparseVector(4, [1, 3], [1.0, 5.5]) - (4,[1,3],[1.0,5.5]) + >>> SparseVector(4, {1: 1.0, 3: 5.5}) + SparseVector(4, {1: 1.0, 3: 5.5}) + >>> SparseVector(4, [(1, 1.0), (3, 5.5)]) + SparseVector(4, {1: 1.0, 3: 5.5}) + >>> SparseVector(4, [1, 3], [1.0, 5.5]) + SparseVector(4, {1: 1.0, 3: 5.5}) """ self.size = int(size) assert 1 <= len(args) <= 2, "must pass either 2 or 3 arguments" @@ -361,8 +369,8 @@ class SparseVector(Vector): self.indices = np.array([p[0] for p in pairs], dtype=np.int32) self.values = np.array([p[1] for p in pairs], dtype=np.float64) else: - if isinstance(args[0], basestring): - assert isinstance(args[1], str), "values should be string too" + if isinstance(args[0], bytes): + assert isinstance(args[1], bytes), "values should be string too" if args[0]: self.indices = np.frombuffer(args[0], np.int32) self.values = np.frombuffer(args[1], np.float64) @@ -591,12 +599,12 @@ class Vectors(object): :param args: Non-zero entries, as a dictionary, list of tupes, or two sorted lists containing indices and values. - >>> print Vectors.sparse(4, {1: 1.0, 3: 5.5}) - (4,[1,3],[1.0,5.5]) - >>> print Vectors.sparse(4, [(1, 1.0), (3, 5.5)]) - (4,[1,3],[1.0,5.5]) - >>> print Vectors.sparse(4, [1, 3], [1.0, 5.5]) - (4,[1,3],[1.0,5.5]) + >>> Vectors.sparse(4, {1: 1.0, 3: 5.5}) + SparseVector(4, {1: 1.0, 3: 5.5}) + >>> Vectors.sparse(4, [(1, 1.0), (3, 5.5)]) + SparseVector(4, {1: 1.0, 3: 5.5}) + >>> Vectors.sparse(4, [1, 3], [1.0, 5.5]) + SparseVector(4, {1: 1.0, 3: 5.5}) """ return SparseVector(size, *args) @@ -645,7 +653,7 @@ class Matrix(object): """ Convert Matrix attributes which are array-like or buffer to array. """ - if isinstance(array_like, basestring): + if isinstance(array_like, bytes): return np.frombuffer(array_like, dtype=dtype) return np.asarray(array_like, dtype=dtype) @@ -677,7 +685,7 @@ class DenseMatrix(Matrix): def toSparse(self): """Convert to SparseMatrix""" indices = np.nonzero(self.values)[0] - colCounts = np.bincount(indices / self.numRows) + colCounts = np.bincount(indices // self.numRows) colPtrs = np.cumsum(np.hstack( (0, colCounts, np.zeros(self.numCols - colCounts.size)))) values = self.values[indices] |