matplotlib-users Mailing List for matplotlib

On 08/20/2011 02:03 AM, Alexa Villaume wrote:
> Hello,
>
> I'm trying to make a contour plot and I'm running into problems when
> specifying the coordinates for the plot. I do,
>
> #Specify coordinates of the surface
> delta0=1.0
> delta1=0.1
> x=np.arange(0.0, 6.0, delta0)
> y=np.arange(-1.0,3.0, delta1)
>
> #Contour the arrays.
> plt.contour(x, y, O3, O3Level)
>
> And I get an error that says,
>
> "TypeError: Length of x must be number of columns in z,
> and length of y must be number of rows."
>
> I don't understand how I'm supposed to get the coordinates for the plot
> that I want and make x and y the same shape as z. Any suggestions?

Contouring requires data on a grid, so your O3 needs to be a 2-D array 
of z values, and your x and y need to specify the x and y locations of 
*those* z values.  The orientation is such that the first dimension of 
the z array corresponds to y, and the second corresponds to x.

Eric

>
> Thanks,
> Alexa
>
>
>
> ------------------------------------------------------------------------------
> Get a FREE DOWNLOAD! and learn more about uberSVN rich system,
> user administration capabilities and model configuration. Take
> the hassle out of deploying and managing Subversion and the
> tools developers use with it. http://p.sf.net/sfu/wandisco-d2d-2
>
>
>
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Hello,

I'm trying to make a contour plot and I'm running into problems when
specifying the coordinates for the plot. I do,

#Specify coordinates of the surface
delta0=1.0
delta1=0.1
x=np.arange(0.0, 6.0, delta0)
y=np.arange(-1.0,3.0, delta1)

#Contour the arrays.
plt.contour(x, y, O3, O3Level)

And I get an error that says,

"TypeError: Length of x must be number of columns in z,
and length of y must be number of rows."

I don't understand how I'm supposed to get the coordinates for the plot that
I want and make x and y the same shape as z. Any suggestions?

Thanks,
Alexa

Thanks Eric and JJ,

Both of your answers are solutions to my problem actually.
I spent a while trying to figure this out and didn't get anywhere.
This was an exercise in frustration with matplotlib's documentation.
Thankfully this list and its members are here to save us. I assumed it
was just a simple flag or option I had missed and this turned out to
be the case. I had even tried setting aspect=False when creating my
AxesGrid object and setting aspect="auto", but because I was poking
around in the dark, I must not have set both at the same time. I also
thought I'd seen an example of this somewhere, which is what Eric
pointed out, but even thinking I'd seen it, I couldn't find it again.
I had looked in the gallery but missed the example - looking back at
the gallery now, I think it might be because every other related
example uses the jet colour scheme and it simply didn't register.

regards,
Gary

On Sat, Aug 20, 2011 at 6:49 PM, Jae-Joon Lee <lee...@gm...> wrote:
> If you want aspect="auto", this must also be set when you create ImageGrid.
> A simple example is attached.
> If you want a fixed aspect other than 1, it is doable but gets a bit
> tricky. Let me know if this is what you want.
>
> Regards,
>
> -JJ
>
>
>
> from mpl_toolkits.axes_grid1 import ImageGrid
>
> fig = plt.figure(1)
>
> grid = ImageGrid(fig, 111, (2, 1),
>                 aspect=False,
>                 label_mode='L', cbar_mode="single",
>                 )
>
> arr = np.arange(100).reshape((10, 10))
> im1 = grid[0].imshow(arr, aspect="auto")
> im2 = grid[1].imshow(arr, aspect="auto")
>
> grid[0].cax.colorbar(im1)
>
>
>
>
> On Sat, Aug 20, 2011 at 2:43 PM, gr...@bi...
> <gr...@bi...> wrote:
>> Usually imshow(arr, aspect='auto') or imshow(arr, aspect=2.0) will
>> display the image with pixels having some aspect ratio other than 1:1
>> However, I cannot get this to work when using imshow within an AxesGrid axis.
>> Is there a way to get an array shown with imshow() within an AxesGrid
>> axis to have a pixel aspect other than 1:1 ?
>> If not, is there a simple way to add a shared colorbar when using subplots() ?
>>
>> Gary
>>
>> ------------------------------------------------------------------------------
>> Get a FREE DOWNLOAD! and learn more about uberSVN rich system,
>> user administration capabilities and model configuration. Take
>> the hassle out of deploying and managing Subversion and the
>> tools developers use with it. http://p.sf.net/sfu/wandisco-d2d-2
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>
>

If you want aspect="auto", this must also be set when you create ImageGrid.
A simple example is attached.
If you want a fixed aspect other than 1, it is doable but gets a bit
tricky. Let me know if this is what you want.

Regards,

-JJ



from mpl_toolkits.axes_grid1 import ImageGrid

fig = plt.figure(1)

grid = ImageGrid(fig, 111, (2, 1),
                 aspect=False,
                 label_mode='L', cbar_mode="single",
                 )

arr = np.arange(100).reshape((10, 10))
im1 = grid[0].imshow(arr, aspect="auto")
im2 = grid[1].imshow(arr, aspect="auto")

grid[0].cax.colorbar(im1)




On Sat, Aug 20, 2011 at 2:43 PM, gr...@bi...
<gr...@bi...> wrote:
> Usually imshow(arr, aspect='auto') or imshow(arr, aspect=2.0) will
> display the image with pixels having some aspect ratio other than 1:1
> However, I cannot get this to work when using imshow within an AxesGrid axis.
> Is there a way to get an array shown with imshow() within an AxesGrid
> axis to have a pixel aspect other than 1:1 ?
> If not, is there a simple way to add a shared colorbar when using subplots() ?
>
> Gary
>
> ------------------------------------------------------------------------------
> Get a FREE DOWNLOAD! and learn more about uberSVN rich system,
> user administration capabilities and model configuration. Take
> the hassle out of deploying and managing Subversion and the
> tools developers use with it. http://p.sf.net/sfu/wandisco-d2d-2
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

On 08/19/2011 07:43 PM, gr...@bi... wrote:
> Usually imshow(arr, aspect='auto') or imshow(arr, aspect=2.0) will
> display the image with pixels having some aspect ratio other than 1:1
> However, I cannot get this to work when using imshow within an AxesGrid axis.
> Is there a way to get an array shown with imshow() within an AxesGrid
> axis to have a pixel aspect other than 1:1 ?
> If not, is there a simple way to add a shared colorbar when using subplots() ?

By shared colorbar, do you mean something like this?
http://matplotlib.sourceforge.net/examples/pylab_examples/multi_image.html

It is using ordinary axes, and is a bit more complex than needed for 
many purposes.

Eric

>
> Gary
>
> ------------------------------------------------------------------------------
> Get a FREE DOWNLOAD! and learn more about uberSVN rich system,
> user administration capabilities and model configuration. Take
> the hassle out of deploying and managing Subversion and the
> tools developers use with it. http://p.sf.net/sfu/wandisco-d2d-2
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Usually imshow(arr, aspect='auto') or imshow(arr, aspect=2.0) will
display the image with pixels having some aspect ratio other than 1:1
However, I cannot get this to work when using imshow within an AxesGrid axis.
Is there a way to get an array shown with imshow() within an AxesGrid
axis to have a pixel aspect other than 1:1 ?
If not, is there a simple way to add a shared colorbar when using subplots() ?

Gary

On 08/15/2011 11:10 PM, Till Stensitzki wrote:
> Hello,
> for a interactive gui program i want to change the shown data without changing
> the rest of the figure. For normal plots, there is a set_xdata/set_ydata
> methods. For QuadMesh, there is a set_array method, which i don't get to work.

You have to use ravel().  Illustrated with ipython -pylab:

xx = pcolormesh(rand(3,3))
xx.set_array(rand(3,3).ravel())
draw()
xx.set_array(rand(3,3).ravel())
draw()


Eric

>
> Thanks,
> Till