how to assign to a "sub set" of an array
Hi, First here is some test I ran, where I think the last command shows a bug:
a = N.arange(4); a.shape=2,2; a [[0 1] [2 3]] aa = N.array((a,a,a)); aa [[[0 1] [2 3]]
[[0 1] [2 3]] [[0 1] [2 3]]]
N.nonzero(a==0) ([0], [0]) aa[N.nonzero(a==0)] = 5,5; aa [[[ 5 544434034] [ 2 3]]
[[ 0 1] [ 2 3]] [[ 0 1] [ 2 3]]] What happend here ? The value 544434034 is different every time..... Background: I have a gray-scale 2D image (in this examples it's a 2x2 pixel image named 'a') I would like to show this image using grey-values EXCEPT I would like to show 0-values as blue. For use with pyOpenGL textures, I convert the image into an RGB image, in the example called 'aa', with shape 3,2,2. This is where I'm getting lost, and I tried assigning "something" (5,5) to aa "only at certain pixels".... Maybe someone can teach what the proper indexing-constuct, that I would need to do what I want. Did I find a bug ? Thanks, Sebastian Haase
On Nov 14, 2007 9:08 AM, Sebastian Haase <haase@msg.ucsf.edu> wrote:
Hi, First here is some test I ran, where I think the last command shows a bug:
a = N.arange(4); a.shape=2,2; a [[0 1] [2 3]] aa = N.array((a,a,a)); aa [[[0 1] [2 3]]
[[0 1] [2 3]]
[[0 1] [2 3]]]
N.nonzero(a==0) ([0], [0]) aa[N.nonzero(a==0)] = 5,5; aa [[[ 5 544434034] [ 2 3]]
[[ 0 1] [ 2 3]]
[[ 0 1] [ 2 3]]]
What happend here ? The value 544434034 is different every time.....
I'm not seeing that here on WinXP with numpy 1.0.4.dev3462 ; can you tell me what platform and version you are using?
Background: I have a gray-scale 2D image (in this examples it's a 2x2 pixel image named 'a') I would like to show this image using grey-values EXCEPT I would like to show 0-values as blue. For use with pyOpenGL textures, I convert the image into an RGB image, in the example called 'aa', with shape 3,2,2.
This is where I'm getting lost, and I tried assigning "something" (5,5) to aa "only at certain pixels".... Maybe someone can teach what the proper indexing-constuct, that I would need to do what I want.
Did I find a bug ?
It looks like a bug, although it's possible it's been fixed. Or maybe it's a new one and I need to update my version of numpy to see it...
Thanks, Sebastian Haase _______________________________________________ Numpy-discussion mailing list Numpy-discussion@scipy.org http://projects.scipy.org/mailman/listinfo/numpy-discussion
-- . __ . |-\ . . tim.hochberg@ieee.org
On Nov 14, 2007 5:59 PM, Timothy Hochberg <tim.hochberg@ieee.org> wrote:
<snip>
On Nov 14, 2007 9:08 AM, Sebastian Haase <haase@msg.ucsf.edu> wrote:
Hi, First here is some test I ran, where I think the last command shows a bug:
a = N.arange(4); a.shape=2,2; a [[0 1] [2 3]] aa = N.array((a,a,a)); aa [[[0 1] [2 3]]
[[0 1] [2 3]]
[[0 1] [2 3]]]
N.nonzero(a==0) ([0], [0]) aa[N.nonzero(a==0)] = 5,5; aa [[[ 5 544434034] [ 2 3]]
[[ 0 1] [ 2 3]]
[[ 0 1] [ 2 3]]]
What happend here ? The value 544434034 is different every time.....
I'm not seeing that here on WinXP with numpy 1.0.4.dev3462 ; can you tell me what platform and version you are using?
Background: I have a gray-scale 2D image (in this examples it's a 2x2 pixel image
named 'a')
I would like to show this image using grey-values EXCEPT I would like to show 0-values as blue. For use with pyOpenGL textures, I convert the image into an RGB image, in the example called 'aa', with shape 3,2,2.
This is where I'm getting lost, and I tried assigning "something" (5,5) to aa "only at certain pixels".... Maybe someone can teach what the proper indexing-constuct, that I would need to do what I want.
Did I find a bug ?
It looks like a bug, although it's possible it's been fixed. Or maybe it's a new one and I need to update my version of numpy to see it...
I'm still on version 1.0,1 -- winXP. looks like it's fixed and I need to upgrade ;-) Thanks for checking, -Sebastian
participants (2)
-
Sebastian Haase -
Timothy Hochberg