Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

Updating arrow position in matplotlib

I would like to update the arrow position while in a loop of plots. I found this post that has an analogous question for the situation in which the patch is a rectangle. Below, the solution proposed in the mentioned post with the addition of the Arrow patch.

from matplotlib import pyplot as plt
from matplotlib.patches import Rectangle, Arrow
import numpy as np

nmax = 10
xdata = range(nmax)
ydata = np.random.random(nmax)


fig, ax = plt.subplots()
ax.plot(xdata, ydata, 'o-')
ax.xaxis.set_ticks(xdata)
plt.ion()

rect = plt.Rectangle((0, 0), nmax, 1, zorder=10)
ax.add_patch(rect)

arrow = Arrow(0,0,1,1)
ax.add_patch(arrow)

for i in range(nmax):
    rect.set_x(i)
    rect.set_width(nmax - i)

    #arrow.what --> which method?

    fig.canvas.draw()
    plt.pause(0.1)

The problem with the Arrow patch is that apparently it does not have a set method related with its position as the Rectangle patch has. Any tip is welcome.

like image 862
Randerson Avatar asked Mar 17 '17 23:03

Randerson


1 Answers

The matplotlib.patches.Arrow indeed does not have a method to update its position. While it would be possible to change its transform dynamically, I guess the easiest solution is to simply remove it and add a new Arrow in each step of the animation.

from matplotlib import pyplot as plt
from matplotlib.patches import Rectangle, Arrow
import numpy as np

nmax = 9
xdata = range(nmax)
ydata = np.random.random(nmax)

plt.ion()
fig, ax = plt.subplots()
ax.set_aspect("equal")
ax.plot(xdata, ydata, 'o-')
ax.set_xlim(-1,10)
ax.set_ylim(-1,4)


rect = Rectangle((0, 0), nmax, 1, zorder=10)
ax.add_patch(rect)

x0, y0 = 5, 3
arrow = Arrow(1,1,x0-1,y0-1, color="#aa0088")

a = ax.add_patch(arrow)

plt.draw()

for i in range(nmax):
    rect.set_x(i)
    rect.set_width(nmax - i)

    a.remove()
    arrow = Arrow(1+i,1,x0-i+1,y0-1, color="#aa0088")
    a = ax.add_patch(arrow)

    fig.canvas.draw_idle()
    plt.pause(0.4)

plt.waitforbuttonpress()    
plt.show()
like image 107
ImportanceOfBeingErnest Avatar answered Nov 04 '22 04:11

ImportanceOfBeingErnest