Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

Animation with contours matplotlib

I am trying to animate contours in matplotlib 1.1.0

Based on: http://www.mail-archive.com/[email protected]/msg17614.html, http://matplotlib.1069221.n5.nabble.com/Matplotlib-1-1-0-animation-vs-contour-plots-td18703.html and http://www.scipy.org/Cookbook/Matplotlib/Gridding_irregularly_spaced_data

from numpy import linspace,exp,vstack
from scipy.interpolate import griddata
import matplotlib.pyplot as plt
import matplotlib.animation as animation
from numpy.random import uniform, seed

def main():
    seed(1234)
    x = uniform(-2,2,100)
    y = uniform(-2,2,100)
    data = vstack((x*exp(-x**2-y**2),0.5*x*exp(-x**2-y**2),0.2*x*exp(-x**2-y**2)))
    xi = linspace(min(x), max(x))
    yi = linspace(min(y), max(y))
    zi = []
    numframes = data.shape[0]
    for ii in range(numframes):
        zi.append(griddata((x, y), data[ii], (xi[None,:], yi[:,None]), method='cubic'))

    fig = plt.figure()
    im = plt.contour(xi, yi, zi[0], 15, linewidths=0.5, colors='k')
    ani = animation.FuncAnimation(fig, update_contour_plot, frames=xrange(numframes), fargs=(zi, im, fig, xi, yi), interval=100)
    plt.colorbar()
    plt.show()

def update_contour_plot(i, data, im, fig, xi, yi):
    for coll in im.collections:
        try:
            plt.gca().collections.remove(coll)
        except ValueError: #Everything is not removed for some reason!                                                                                      
            pass
    im = plt.contour(xi, yi, data[i], 15, linewidths=0.5, colors='k')
    plt.title(str(i))
    return im,

main()

1) Is this the best way to do this? Mention otherwise.

2) In the final output, contour lines from previous frame are still visible. How do I remove them? See except ValueError:

like image 855
chaitu Avatar asked Jan 02 '13 15:01

chaitu


1 Answers

You have a scoping problem, the new im you create in your updater is not being passed back out, so the second time through the loop you are trying to remove the first layer of lines again, which it rightly complains it can't because you already removed them.

Tweaking your code a bit:

def main():
    seed(1234)
    x = uniform(-2,2,100)
    y = uniform(-2,2,100)
    data = vstack((x*exp(-x**2-y**2),0.5*x*exp(-x**2-y**2),0.2*x*exp(-x**2-y**2)))
    xi = linspace(min(x), max(x))
    yi = linspace(min(y), max(y))
    zi = []
    numframes = data.shape[0]
    for ii in range(numframes):
        zi.append(griddata((x, y), data[ii], (xi[None,:], yi[:,None]), method='cubic'))

    fig = plt.figure()
    im = plt.contour(xi, yi, zi[0], 15, linewidths=0.5, colors='k')
    ax = fig.gca()
    ani = animation.FuncAnimation(fig, update_contour_plot, frames=xrange(numframes), fargs=(zi, ax, fig, xi, yi), interval=100)
    plt.colorbar(im)
    plt.show()
    return ani


def update_contour_plot(i, data,  ax, fig, xi, yi):
    ax.cla()
    im = ax.contour(xi, yi, data[i], 15, linewidths=0.5, colors='k')
    plt.title(str(i))
    return im,

Another option is to make im a mutable of some sort (like wrap it in a dictionary) so that your changes will propagate from frame to frame.

like image 93
tacaswell Avatar answered Oct 21 '22 00:10

tacaswell