[pypy-issue] [issue1674] broken mask indexing / selection
remi
tracker at bugs.pypy.org
Mon Jan 13 18:49:57 CET 2014
New submission from remi <remi2402 at free.fr>:
>>> import numpy as np
>>> data = np.arange(15).reshape(3, 5) * 1.0
>>> data
array([[ 0., 1., 2., 3., 4.],
[ 5., 6., 7., 8., 9.],
[ 10., 11., 12., 13., 14.]])
>>> m = data[:, 0] % 2 == 0
>>> m
array([ True, False, True], dtype=bool)
on CPython 2.7 with numpy 1.6.2 (debian stable/wheezy)
>>> data[m, 0]
array([ 0., 10.])
on PyPy / Python 2.7.3 (25070f2267c4, Jan 13 2014, 14:17:42)
>>>> data[m, 0]
array([ 5., 0., 5.])
----------
assignedto: r.lobb
messages: 6485
nosy: pypy-issue, r.lobb, remi
priority: bug
status: unread
title: broken mask indexing / selection
________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1674>
________________________________________
More information about the pypy-issue
mailing list