How to update plot title with matplotlib using animation?

Question

Below my code. Why doesn't title updates every tick? I read this: Matplotlib animation with blit -- how to update plot title? but it wasn't useful.

#! /usr/bin/python3
import matplotlib
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.animation as animation
import time
import random as rr

plt.rc('grid', color='#397939', linewidth=1, linestyle='-')
plt.rc('xtick', labelsize=10)
plt.rc('ytick', labelsize=5)

width, height = matplotlib.rcParams['figure.figsize']
size = min(width, height)
fig = plt.figure(figsize=(size, size))
ax = fig.add_axes([0.1, 0.1, 0.8, 0.8], polar=True, facecolor='#cfd98c')

ax.set_rmax(20.0)
plt.grid(True)
plt.title("")
def data_gen(t=0):
    tw = 0
    phase = 0
    ctn = 0
    while True:
        ctn += 1
        if ctn == 1000:
            phase=round(rr.uniform(0,180),0)
            tw = round(rr.uniform(0,20),0)
            ctn = 0
            yield tw, phase
def update(data):
    tw, phase = data
    print(data)
    ax.set_title("|TW| = {}, Angle: {}°".format(tw, phase)) 
    arr1 = plt.arrow(phase, 0, 0, tw, alpha = 0.5, width = 0.080,
             edgecolor = 'red', facecolor = 'red', lw = 2, zorder = 5)
    return arr1,

ani = animation.FuncAnimation(fig, update, data_gen, interval=100, blit=True, repeat=False)
plt.show()

EDIT n. 1 After @eyllanesc answer, I edit this piece of code:

def data_gen(t=0):
    tw = 10
    phase = 0
    ctn = 0
    while True:
        if phase < 2*180:
            phase += 1
        else:
            phase=0
        yield tw, phase

def update(data):
    tw, phase = data
    angolo = phase /180 * np.pi
    print("|TW| = {}, Angle = {}°".format(tw,phase))
    ax.set_title("|TW| = {}, Angle: {}°".format(tw, phase)) 
    arr1 = plt.arrow(angolo, 0, 0, tw, alpha = 0.5, width = 0.080, edgecolor = 'red', facecolor = 'red', lw = 2, zorder = 5)
    plt.draw()
    return arr1,

Now text works correctly, but the arrow updating is not "fluid" (it appeares and disappeares every update).

Answer 1

The problem arises when using blit=True in the FuncAnimation. This will store the background and only update the artists returned by the update function. However, the restored background will overwrite the title, since it is outside the axes.

Possible solutions are:

1. Put the title inside the axes.
2. Don't use blitting
3. Possibly using an ArtistAnimation instead of a FuncAnimation would work as well. But I haven't tested it.

Note that using plt.draw or similar inside the updating function (as proposed in other answers) does not make sense, since it destroys all the advatages of using blitting and makes the animation even slower than in the case of not using blitting.

1. Put the title inside the axes (blit=True)

Instead of a title outside the axes, you may use a title = ax.text(..) positionned inside the axes. This text can be updated for each iteration title.set_text("..") and must then be returned by the updating function (return arr1, title,).

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

plt.rc('grid', color='#397939', linewidth=1, linestyle='-')
plt.rc('xtick', labelsize=10)
plt.rc('ytick', labelsize=5)

width, height = matplotlib.rcParams['figure.figsize']
size = min(width, height)
fig = plt.figure(figsize=(size, size))
ax = fig.add_axes([0.1, 0.1, 0.8, 0.8], polar=True, facecolor='#cfd98c')

ax.set_rmax(20.0)
plt.grid(True)

title = ax.text(0.5,0.85, "", bbox={'facecolor':'w', 'alpha':0.5, 'pad':5},
                transform=ax.transAxes, ha="center")

def data_gen(t=0):
    tw = 10
    phase = 0
    while True:
        if phase < 2*180:
            phase += 2
        else:
            phase=0
        yield tw, phase

def update(data):
    tw, phase = data
    title.set_text(u"|TW| = {}, Angle: {}°".format(tw, phase))
    arr1 = ax.arrow(np.deg2rad(phase), 0, 0, tw, alpha = 0.5, width = 0.080,
             edgecolor = 'red', facecolor = 'red', lw = 2, zorder = 5)
    return arr1,title,

ani = animation.FuncAnimation(fig, update, data_gen, interval=100, blit=True, repeat=False)

plt.show()

Note that I slightly changed the angle setting, since it the angle for arrow must be in radiants instead of degrees.

2. Not use blitting (blit=False)

You may decide not to use blitting, which makes sense when the requirements on the speed of the animation are not so high. Not using blitting allows to use a normal title outside the axes. However, in this case you would need to remove the artist for each iteration (otherwise one would end up with a lot of arrows in the plot).

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

plt.rc('grid', color='#397939', linewidth=1, linestyle='-')
plt.rc('xtick', labelsize=10)
plt.rc('ytick', labelsize=5)

width, height = matplotlib.rcParams['figure.figsize']
size = min(width, height)
fig = plt.figure(figsize=(size, size))
ax = fig.add_axes([0.1, 0.1, 0.8, 0.8], polar=True, facecolor='#cfd98c')

ax.set_rmax(20.0)
plt.grid(True)

def data_gen(t=0):
    tw = 10
    phase = 0
    while True:
        if phase < 2*180:
            phase += 2
        else:
            phase=0
        yield tw, phase

arr1 = [None]
def update(data):
    tw, phase = data
    ax.set_title(u"|TW| = {}, Angle: {}°".format(tw, phase))
    if arr1[0]: arr1[0].remove()
    arr1[0] = ax.arrow(np.deg2rad(phase), 0, 0, tw, alpha = 0.5, width = 0.080,
             edgecolor = 'red', facecolor = 'red', lw = 2, zorder = 5)

ani = animation.FuncAnimation(fig, update, data_gen, interval=100, blit=False, repeat=False)

plt.show()