A slightly related question on this topic...
Is there a good loopless way to identify all of the unique rows in an
array? Something like numpy.unique() is ideal, but capable of
extracting unique subarrays along an axis.
Thanks,
-Mark
mark wrote:
> Maybe this is not the intended use of where,
On Tue, Aug 14, 2007 at 11:53:03AM +0100, Andy Cheesman wrote:
> Dear nice people
>
> I'm trying to match a row (b) within a large numpy array (a). My most
> successful attempt is below
>
> hit = equal(b, a)
> total_hits = add.reduce(hit, 1)
> max_hit = argmax(total_hits, 0)
> answer = a[max_hit]
On Tue, Aug 14, 2007 at 11:53:03AM +0100, Andy Cheesman wrote:
> Dear nice people
>
> I'm trying to match a row (b) within a large numpy array (a). My most
> successful attempt is below
>
> hit = equal(b, a)
> total_hits = add.reduce(hit, 1)
> max_hit = argmax(total_hits, 0)
> answer = a[max_hit]
Dear nice people
I'm trying to match a row (b) within a large numpy array (a). My most
successful attempt is below
hit = equal(b, a)
total_hits = add.reduce(hit, 1)
max_hit = argmax(total_hits, 0)
answer = a[max_hit]
where ...
a = array([[ 0, 1, 2, 3],
[ 4, 5, 6, 7],
Maybe this is not the intended use of where, but it seems to work:
>>> from numpy import * # No complaining now
>>> a = arange(12)
>>> a.shape = (4,3)
>>> a
array([[ 0, 1, 2],
[ 3, 4, 5],
[ 6, 7, 8],
[ 9, 10, 11]])
>>> b = array([6,7,8])
>>> row = all( equal(a,b), 1 )
>>>
The where function ?
Matthieu
2007/8/15, mark <[EMAIL PROTECTED]>:
>
> Oops, 'find' is in pylab (matplotlib).
> I guess in numpy you have to use 'where', which does almost the same,
> but it returns a Tuple.
> Is there a function that is more like the find in matplotlib?
> Mark
>
>
> On Aug 15, 1
Oops, 'find' is in pylab (matplotlib).
I guess in numpy you have to use 'where', which does almost the same,
but it returns a Tuple.
Is there a function that is more like the find in matplotlib?
Mark
On Aug 15, 12:26 pm, Andy Cheesman <[EMAIL PROTECTED]>
wrote:
> Thanks for the speedy response bu
Thanks for the speedy response but where can I locate the find function
as it isn't in numpy.
Andy
mark wrote:
> I think you can create an array with a true value in the right spot as
> folows:
>
> row = all( equal(a,b), 1 )
>
> Then you can either find the row (but you already knew that one, a
I think you can create an array with a true value in the right spot as
folows:
row = all( equal(a,b), 1 )
Then you can either find the row (but you already knew that one, as it
is b)
a[row]
or the row index
find(row==True)
Mark
On Aug 15, 11:53 am, Andy Cheesman <[EMAIL PROTECTED]>
wrote:
>
Dear nice people
I'm trying to match a row (b) within a large numpy array (a). My most
successful attempt is below
hit = equal(b, a)
total_hits = add.reduce(hit, 1)
max_hit = argmax(total_hits, 0)
answer = a[max_hit]
where ...
a = array([[ 0, 1, 2, 3],
[ 4, 5, 6, 7],
10 matches
Mail list logo
