update figure inside loop

Question

I'm new to Python matplotlib and want to make a figure updated in a loop. This is some simple test code. I want to draw 3 graphs, each for 2 seconds, with different slopes each time. Of course, I know I should read some more about matplotlib but I need to do something quick. What is wrong below? Thanks!

#!/usr/bin/env python

import matplotlib.pyplot as plt
import numpy as np
import time

def demo(a):
    y = [xt*a+1 for xt in x]
    ax.plot(x,y)

if __name__ == '__main__':
    plt.ion()
    fig, ax = plt.subplots()
    ax.set_ylim([0,15])
    x = range(0,5)
    for a in range(1,4):
        demo(a)
        time.sleep(2)

I thought plt.ion() makes it interactive and ax.plot(x,y) take effect instantly (but guess not). I tried adding ax.draw() after ax.plot(x,y) but it requires some arguments like artist or something which I don't know yet :). Also, I have this error message coming (Ubuntu 14.04.LTS).

Xlib:  extension "XInputExtension" missing on display ":1.0".
X Error: BadDrawable (invalid Pixmap or Window parameter) 9
  Major opcode: 62 (X_CopyArea)
  Resource id:  0x0

EDIT : I add the correct code below. (according to Estilus's solution to which I modified a little for correct graph display.

#!/usr/bin/env python

import matplotlib.pyplot as plt
import numpy as np
import time

def demo(a):
    plt.cla()
    y = [xt*a+1 for xt in x]
    ax.set_ylim([0,15])
    ax.plot(x,y)

if __name__ == '__main__':
    plt.ion()
    fig, ax = plt.subplots()
    x = range(5)
    for a in range(1,4):
        demo(a)
        plt.pause(3)
        plt.draw()

Answer 1

So plt.ion() needs to be paused for a short period of time for it to be interactive. Otherwise, you'll just run into a frozen white screen. Secondly you want to use draw() to update the figure. Therefore, your code would look something like this:

import matplotlib.pyplot as plt
import numpy as np
import time

def demo(a):
    y = [xt*a+1 for xt in x]
    ax.plot(x,y)

if __name__ == '__main__':
    plt.ion()
    fig, ax = plt.subplots()
    ax.set_ylim([0,15])
    x = range(0,5)
    for a in range(1,4):
        demo(a)
        plt.pause(3)
        plt.draw()

I set the pause time to be 3 seconds, but it could be a pretty small number if you'd like (like 0.000001). Hope this helped!

Answer 2

It seems that you're trying to understand how to update a plot in matplotlib.

You should use this code to understand how to use fig1.canvas.draw() to update your charts data:

fig1 = figure()

for n in range(30):

    ## Clear the figure
    clf()

    # Do a plot that changes for each iteration
    plot(range(n))

    # Now set the xlimit so that it doesn’t change the plot
    xlim((0,30))
    ylim((0,30))

    ## Now the magic, we must update the canvas
    fig1.canvas.draw()

Answer 3

Here is a code which works. The key lines are:

• plt.cla(): to clear the axis between each draw. Remove this line if you want to keep all graphs.
• fig.canvas.draw(): to actually draw the plot into the canvas of the figure

---

#!/usr/bin/env python

import matplotlib.pyplot as plt
import numpy as np
import time

def demo(a):
    plt.cla()
    y = [xt*a+1 for xt in x]
    ax.set_ylim([0,200])
    ax.plot(x,y)
    fig.canvas.draw()

if __name__ == '__main__':
    plt.ion()
    fig, ax = plt.subplots()
    x = range(5)
    for a in range(40):
        demo(a)
        time.sleep(0.1)