[Numpy-discussion] int and long issues
Hi, I find this to be a little strange: x = numpy.arange(10) isinstance(x[0],int) gives True y = numpy.where(x 5)[0] isinstance(y[0],int) gives False isinstance(y[0],long) gives True Specs: Python 2.7.2, numpy-1.6.1, Win7, 64 bit Best regards, Mads -- +-+ | Mads Ipsen | +--+--+ | Gåsebæksvej 7, 4. tv | | | DK-2500 Valby| phone: +45-29716388 | | Denmark | email: mads.ip...@gmail.com | +--+--+ ___ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion
Re: [Numpy-discussion] int and long issues
On Thu, 2013-01-10 at 11:32 +0100, Mads Ipsen wrote: Hi, I find this to be a little strange: x = numpy.arange(10) isinstance(x[0],int) gives True y = numpy.where(x 5)[0] isinstance(y[0],int) gives False isinstance(y[0],long) Check what type(x[0])/type(y[0]) prints, I expect these are very different, because the default integer type and the integer type used for indexing (addressing memory in general) are not necessarily the same. And because of that, `y[0]` probably simply isn't compatible to the datatype of a python integer for your hardware and OS (for example for me, your code works). So on python 2 (python 3 abolishes int and makes long the only integer, so this should work as expected there) you have to just check both even in the python context, because you can never really know (there may be some nice trick for that, but not sure). And if you want to allow for rare 0d arrays as well (well they are very rare admittingly)... it gets even a bit hairier. gives True Specs: Python 2.7.2, numpy-1.6.1, Win7, 64 bit Best regards, Mads -- +-+ | Mads Ipsen | +--+--+ | Gåsebæksvej 7, 4. tv | | | DK-2500 Valby| phone: +45-29716388 | | Denmark | email: mads.ip...@gmail.com | +--+--+ ___ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion ___ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion
Re: [Numpy-discussion] int and long issues
On Thu, 2013-01-10 at 11:32 +0100, Mads Ipsen wrote: Hi, I find this to be a little strange: x = numpy.arange(10) isinstance(x[0],int) gives True y = numpy.where(x 5)[0] isinstance(y[0],int) gives False isinstance(y[0],long) Check what type(x[0])/type(y[0]) prints, I expect these are very different, because the default integer type and the integer type used for indexing (addressing memory in general) are not necessarily the same. And because of that, `y[0]` probably simply isn't compatible to the datatype of a python integer for your hardware and OS (for example for me, your code works). So on python 2 (python 3 abolishes int and makes long the only integer, so this should work as expected there) you have to just check both even in the python context, because you can never really know (there may be some nice trick for that, but not sure). And if you want to allow for rare 0d arrays as well (well they are very rare admittingly)... it gets even a bit hairier. Regards, Sebastian gives True Specs: Python 2.7.2, numpy-1.6.1, Win7, 64 bit Best regards, Mads -- +-+ | Mads Ipsen | +--+--+ | Gåsebæksvej 7, 4. tv | | | DK-2500 Valby| phone: +45-29716388 | | Denmark | email: mads.ip...@gmail.com | +--+--+ ___ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion ___ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion
Re: [Numpy-discussion] int and long issues
Sebastian - thanks - very helpful. Best regards, Mads On 10/01/2013 12:06, Sebastian Berg wrote: On Thu, 2013-01-10 at 11:32 +0100, Mads Ipsen wrote: Hi, I find this to be a little strange: x = numpy.arange(10) isinstance(x[0],int) gives True y = numpy.where(x 5)[0] isinstance(y[0],int) gives False isinstance(y[0],long) Check what type(x[0])/type(y[0]) prints, I expect these are very different, because the default integer type and the integer type used for indexing (addressing memory in general) are not necessarily the same. And because of that, `y[0]` probably simply isn't compatible to the datatype of a python integer for your hardware and OS (for example for me, your code works). So on python 2 (python 3 abolishes int and makes long the only integer, so this should work as expected there) you have to just check both even in the python context, because you can never really know (there may be some nice trick for that, but not sure). And if you want to allow for rare 0d arrays as well (well they are very rare admittingly)... it gets even a bit hairier. gives True Specs: Python 2.7.2, numpy-1.6.1, Win7, 64 bit Best regards, Mads -- +-+ | Mads Ipsen | +--+--+ | Gåsebæksvej 7, 4. tv | | | DK-2500 Valby| phone: +45-29716388 | | Denmark | email: mads.ip...@gmail.com | +--+--+ ___ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion ___ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion -- +-+ | Mads Ipsen | +--+--+ | Gåsebæksvej 7, 4. tv | | | DK-2500 Valby| phone: +45-29716388 | | Denmark | email: mads.ip...@gmail.com | +--+--+ ___ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion
