numpy-discussion

[Numpy-discussion] Dimension reduction

[David H. <dav...@gm...>]

Hi,

Is there an elegant way to reduce an array but conserve the reduced
dimension ?

Currently,
>>> a = random.random((10,10,10))
>>> a.sum(1).shape
(10,10)

but i'd like to keep (10,1,10) so I can do a/a.sum(1) directly.

Thanks,

David

Re: [Numpy-discussion] Dimension reduction

[Robert K. <rob...@gm...>]

David Huard wrote:
> Hi,
> 
> Is there an elegant way to reduce an array but conserve the reduced 
> dimension ?
> 
> Currently,
>  >>> a = random.random((10,10,10))
>  >>> a.sum(1).shape
> (10,10)
> 
> but i'd like to keep (10,1,10) so I can do a/a.sum(1) directly.

def nonreducing_reducer(reducing_func, arr, axis):
     reduced = reducing_func(arr, axis=axis)
     shape = list(reduced.shape)
     axis = axis % len(arr.shape)
     shape.insert(axis, 1)
     reduced.shape = tuple(shape)
     return reduced


I think.

-- 
Robert Kern

"I have come to believe that the whole world is an enigma, a harmless enigma
  that is made terrible by our own mad attempt to interpret it as though it had
  an underlying truth."
   -- Umberto Eco

Re: [Numpy-discussion] Dimension reduction

[Tim H. <tim...@ie...>]

Robert Kern wrote:
> David Huard wrote:
>   
>> Hi,
>>
>> Is there an elegant way to reduce an array but conserve the reduced 
>> dimension ?
>>
>> Currently,
>>  >>> a = random.random((10,10,10))
>>  >>> a.sum(1).shape
>> (10,10)
>>
>> but i'd like to keep (10,1,10) so I can do a/a.sum(1) directly.
>>     
>
> def nonreducing_reducer(reducing_func, arr, axis):
>      reduced = reducing_func(arr, axis=axis)
>      shape = list(reduced.shape)
>      axis = axis % len(arr.shape)
>      shape.insert(axis, 1)
>      reduced.shape = tuple(shape)
>      return reduced
>
>
> I think.
>
>   
Alternatively (untested):

def nonreducing_reducer(reducing_func, arr, axis):
     return reducing_func(arr.swapaxis(0, axis), axis=0)[newaxis].swapaxis(0, axis)


Adding some vertical whitespace probably wouldn't hurt for readability I suppose. 

-tim

Re: [Numpy-discussion] Dimension reduction

[David H. <dav...@gm...>]

2006/10/19, Robert Kern <rob...@gm...>:

> def nonreducing_reducer(reducing_func, arr, axis):
>      reduced = reducing_func(arr, axis=axis)
>      shape = list(reduced.shape)
>      axis = axis % len(arr.shape)
>      shape.insert(axis, 1)
>      reduced.shape = tuple(shape)
>      return reduced


Cute !

Here is another one:
1dfunc = lambda x: atleast_1d(reducing_function(x))
apply_along_axis(1dfunc, axis, arr)

Is it something people often need ? Would it be worth, (for numpy 1.1 ?), to
consider complex axis arguments as a wish to preserve the rank of the array?
>>> a.shape
(n,m,o,p)
>>> a.sum(axis=1j)
(n,1,o,p)
>>> a.sum(axis=1)
(n,o,p)

Thanks,

David

Re: [Numpy-discussion] Dimension reduction

[Charles R H. <cha...@gm...>]

On 10/19/06, David Huard <dav...@gm...> wrote:
>
> Hi,
>
> Is there an elegant way to reduce an array but conserve the reduced
> dimension ?
>
> Currently,
> >>> a = random.random((10,10,10))
> >>> a.sum(1).shape
> (10,10)
>
> but i'd like to keep (10,1,10) so I can do a/a.sum(1) directly.


In [8]: a.sum(1)[:,newaxis,:].shape
Out[8]: (10, 1, 10)

Don't know if this is as universal as you want, but it works for this.
Robert's answer is more general.

Chuck