[Numpy-discussion] concatenation changes dtype

[Martin W. <mar...@gm...>]

Hi list,

Apparently r_[x,x] does not necessarily have the same dtype as x:

>>> from numpy import *
>>> dt = dtype ([('a', 'b'), ('f4', 'i4')])
>>> x = zeros ((1,), dt)
>>> x.dtype
dtype([('a', '|i1'), ('f4', '<i4')])
>>> r_[x,x].dtype
dtype('object')
>>>
>>> import numpy
>>> numpy.version.version
'1.0rc1.dev3171'

Does anybody know how to avoid this change of dtype?

Thanks, Martin.