[Numpy-discussion] Antwort: Re: why does binary_repr don't support arrays

markus.proeller@if... markus.proeller@if...
Thu Oct 22 06:45:46 CDT 2009


numpy-discussion-bounces@scipy.org schrieb am 22.10.2009 12:36:46:

> >
> >
> > On Tue, Oct 20, 2009 at 11:17 AM, <markus.proeller@ifm.com> wrote:
> > 
> > Hello, 
> > 
> > I'm always wondering why binary_repr doesn't allow arrays as input 
> > values. I always have to use a work around like: 
> > 
> > import numpy as np 
> > 
> > def binary_repr(arr, width=None): 
> >     binary_list = map((lambda foo: np.binary_repr(foo, width)), 
arr.flatten())
> >     str_len_max = len(np.binary_repr(arr.max(), width=width)) 
> >     str_len_min = len(np.binary_repr(arr.min(), width=width)) 
> >     if str_len_max > str_len_min: 
> >         str_len = str_len_max 
> >     else: 
> >         str_len = str_len_min 
> >     binary_array = np.fromiter(binary_list, dtype='|S'+str(str_len)) 
> >     return binary_array.reshape(arr.shape) 
> > 
> > Is there a reason why arrays are not supported or is there another 
> > function that does support arrays? 
> 
> Not sure if there was/is a reason, but imho it would be nice to have
> support for arrays. Also in base_repr. Could you file a ticket in trac?
> 
> Cheers,
> Ralf
>  

Okay, I opened a new ticket:

http://projects.scipy.org/numpy/ticket/1270

Markus
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.scipy.org/pipermail/numpy-discussion/attachments/20091022/995b6839/attachment.html 


More information about the NumPy-Discussion mailing list