Animate labels using FuncAnimation in Matplotlib

Question

I am not able to make (animated) labels using FuncAnimation from matplotlib. Please find below a minimal code that I made. ax.annotate has no effect at all - the animation itself works though. What can I change to get animated labels/titles, which are different for each frame?

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

fig, ax = plt.subplots()
fig.clear()
steps = 10
data = np.random.rand(20,20,10)
imagelist = [data[:,:,i] for i in range(steps) ]
im = plt.imshow(imagelist[0], cmap='Greys',  origin='lower', animated=True)
plt.colorbar(shrink=1, aspect=30, label='Counts')

# does not work
ax.annotate("Frame: %d " % steps,(0.09,0.92),xycoords ='figure fraction')  

def updatefig(j):
    im.set_array(imagelist[j])
    return [im]

ani = animation.FuncAnimation(fig, updatefig, frames=range(steps), interval=200, blit=True)
plt.show()

Answer 1

Two problems overall:

• The annotation text never gets updated in updatefig()
• The canvas gets cleared+blitted, which wipes out annotations

Five steps to resolve:

1. Remove fig.clear() to preserve annotations
2. Save the initial annotation's handle
3. Update the annotation's text in updatefig()
4. Include the annotation in the return of updatefig()
5. Set blit=False to preserve annotations

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

fig, ax = plt.subplots()
#1 do NOT call fig.clear()

steps = 10
data = np.random.rand(20, 20, steps)
im = plt.imshow(data[:, :, 0], cmap='Greys', origin='lower', animated=True)
plt.colorbar(shrink=1, aspect=30, label='Counts')

#2 annotate frame 0 and save handle
annot = ax.annotate('Frame: 0', (0.09, 0.92), xycoords='figure fraction')

def updatefig(j):
    im.set_array(data[:, :, j])

    #3 update annotation text
    annot.set_text(f'Frame: {j}')

    #4 include annotation when returning
    return im, annot

#5 set blit=False
anim = animation.FuncAnimation(fig, updatefig, frames=steps, blit=False)