[Matplotlib-users] streamplots in axesgrid

[Yoshi Rokuko]

Hey,

I'm trying to plot streamplots into an axesgrid object with something
like:

fig = pl.figure(1, (13, 20))
grid = AxesGrid(fig, 111,
nrows_ncols = (3, 2),
axes_pad = 0.6,
cbar_location = 'top',
cbar_mode = 'each',
cbar_size = '2%',
cbar_pad = '1%',
   )
[...]
norm = mpl.colors.LogNorm(vmin=1, vmax=5000)
im = grid[i].streamplot(XPTS, YPTS, zx, zy,
color=zr,
arrowsize=.001,
norm=norm)
grid.cbar_axes[i].colorbar(im)
[...]

and then it failes with:

Traceback (most recent call last):
  File "./results.py", line 96, in 
grid.cbar_axes[i].colorbar(im)
  File
"/usr/lib/python2.7/site-packages/mpl_toolkits/axes_grid1/axes_grid.py",
line 85, in colorbar cb = Colorbar(self, mappable,
orientation=orientation, **kwargs) File
"/usr/lib/python2.7/site-packages/mpl_toolkits/axes_grid1/colorbar.py",
line 706, in __init__ mappable.autoscale_None() # Ensure
mappable.norm.vmin, vmax AttributeError: 'StreamplotSet' object has no
attribute 'autoscale_None'

can't I use streamplots in axesgrid? If I comment out the colorbar for
the streamplot colors it works ...
Is this a bug? Is there a right way to do it? Is there a work around?

Many thanks in advance!

Regards, Yoshi

--
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60133471&iu=/4140/ostg.clktrk
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users

[Matplotlib-users] specifying path to libpng

[Nat Echols]

Is there a way to force matplotlib to build (on Mac, 10.6, Xcode 3) against
a specific, non-standard location of libpng?  Right now it appears to be
linking to /usr/X11/lib/libpng12.0.dylib, which is making it very difficult
to install the resulting binaries on other Macs.  I couldn't figure out a
way to point it somewhere else.

thanks,
Nat
--
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60133471&iu=/4140/ostg.clktrk___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Re: [Matplotlib-users] curtain plots

[Jody Klymak]

Hi All,

To follow up on my own post - because my curtains and contours were 
well-ordered, I simply set the "zorder" on each call and got the right effect.  

Thanks,   Jody

On Sep 25, 2013, at  15:15 PM, Jody Klymak  wrote:

>   
> Hi all,
> 
> I am trying to make 3-D "curtain" plots.  Basically, x,y are N-vectors, z is 
> an M-vector, and C is MxN data set collected on the path with z.  Application 
> is a ship's track through the ocean. I also want to be able to contour a 
> second variable C2 also MxN. I know how to do that, but the example below 
> just uses plot3D, because thats how I do the curtain contouring.  
> 
> If I plot three such "curtains" they look OK, including the magenta line in 
> each.  
> 
> If I plot a fourth, the magenta line is obscured by the curtain, and so on 
> for more curtains.
> 
> 
> 
> Any clue what the problem is? The code for this example is below, and I think 
> is self contained, plus or minus running in pylab.
> 
> Thanks,  Jody
> 
> from mpl_toolkits.mplot3d import Axes3D
> from matplotlib import cm
> import matplotlib.pyplot as plt
> import numpy as np
> 
> n=0
> fig = figure()
> for Nn in array([3,4]):
> n=n+1
> ax = fig.add_subplot(2,1,n,projection='3d')
> for off in arange(0,Nn*2,2)*50.:
> 
> x = np.arange(-5, 5, .5)
> y = np.arange(-5, 5, .5)
> Z = np.arange(0,200,1)
> Z=np.tile(np.reshape(Z,(200,1)),(1,size(y)))
> X = np.tile(y,(200,1))
> Y = np.tile(y,(200,1))
>
> N = X*Y*Z
> N = N/N.max()  # normalize 0..1
> surf = ax.plot_surface(
> X+off, Y, Z, rstride=20, cstride=4,
> facecolors=cm.jet(N),
> linewidth=0, antialiased=False, shade=False,alpha=0.9)
> ax.plot(x+off+0.001,y,(y+5)*25.,'m')
> ax.set_xlim([-50,350])
> ax.set_ylim([-8.,8.])
> fig.savefig('doc/bad3dslices.png',res=72)
> 
> --
> Jody Klymak
> http://web.uvic.ca/~jklymak/
> 
> 
> 
> 
> --
> October Webinars: Code for Performance
> Free Intel webinars can help you accelerate application performance.
> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
> the latest Intel processors and coprocessors. See abstracts and register >
> http://pubads.g.doubleclick.net/gampad/clk?id=60133471&iu=/4140/ostg.clktrk___
> Matplotlib-users mailing list
> Matplotlib-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

--
Jody Klymak
http://web.uvic.ca/~jklymak/




--
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60133471&iu=/4140/ostg.clktrk___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users