Animation is not working on a matplotlib gridspec - python-3.x

Question

I have reviewed the answer to the question:
how to use GridSpec() with FuncAnimation in matplotlib?
however I believe I am doing something wrong but do not know what, my mission here is to create multiple animations for each of the grids and I also used a code mode for a gauge to animate the count of used data.
x1 = np.random.normal(-2.5, 1, 10000)
x2 = np.random.gamma(2, 1.5, 10000)
x3 = np.random.exponential(2, 10000)+7
x4 = np.random.uniform(14,20, 10000)
xrand = np.random.random(size = 10000)
x = [x1, x2, x3, x4]
plt.figure()
gspec = gridspec.GridSpec(3,3, wspace = 0.25)
GS1 = plt.subplot(gspec[0, 0])
GS2 = plt.subplot(gspec[0, 1])
GS3 = plt.subplot(gspec[1, 0])
GS4 = plt.subplot(gspec[1, 1])
GS5 = plt.subplot(gspec[0:,2:])
GS6 = plt.subplot(gspec[2,:2])
GS = [GS1, GS2, GS3, GS4, GS5, GS6]
bins1 = np.arange(-7.5, 2.5, 0.2)
bins2 = np.arange(0, 10, 0.2)
bins3 = np.arange(7, 17, 0.2)
bins4 = np.arange(12, 22, 0.2)
bins = [bins1, bins2, bins3, bins4]
axis1 = [-7.5, 2.5, 0, 0.6]
axis2 = [0, 10, 0, 0.6]
axis3 = [7, 17, 0, 0.6]
axis4 = [12, 22, 0, 0.6]
axis = [axis1, axis2, axis3, axis4]
GS1.hist(x1, bins = 1000)
GS2.hist(x2, bins = 1000)
GS3.hist(x3, bins = 1000)
GS4.hist(x4, bins = 1000)
GS5.scatter(x1, xrand, norm = True, c = 'r', s= 0.7)
GS5.scatter(x2, xrand, norm = True, c = 'g', s= 0.7)
GS5.scatter(x3, xrand, norm = True, c = 'b', s= 0.7)
GS5.scatter(x4, xrand, norm = True, c = 'y', s= 0.7)
for s in GS:
s.spines['right'].set_visible(False)
s.spines['top'].set_visible(False)
gspec.update(wspace = .6, hspace = .6)
fig, ((GS1,GS2),(GS3, GS4)) = plt.subplots(2, 2, sharey = True)
GS = [GS1, GS2, GS3, GS4]
def update(curr):
if curr == 500:
a.event_source.stop()
for i in range(0, len(GS)):
GS[i].hist(x[i][:curr], bins = bins[i], normed = True)
GS[i].axis(axis[i])
GS[i].gca().set_title('Sampling random distribution')
GS[i].gca().set_ylabel('Frequency')
GS[i].gca().set_xlabel('Value')
GS[i].annotate('n = {}'.format(curr), [3,27])
plt.tight_layout()
plt.show()
fig = plt.gcf()
a = animation.FuncAnimation(fig, update, interval = 10) #, blit = True, repeat = True)
I do not understand what is the problem with the code.

Answer 1

Problems:
You cannot use plt.show() inside the updating function.
GS[i].gca().set_title does not make sense, because GS[i] already is an axes, which does not have any gca() method. Same for set_ylabel and set_xlabel.
You need to call tight_layout() only once outside the loop.
You will need to remove the old histogram before plotting a new one on the axes.
Possibly the following is what you're after:
import matplotlib.pyplot as plt
import numpy as np
from matplotlib import animation
x1 = np.random.normal(-2.5, 1, 10000)
x2 = np.random.gamma(2, 1.5, 10000)
x3 = np.random.exponential(2, 10000)+7
x4 = np.random.uniform(14,20, 10000)
xrand = np.random.random(size = 10000)
x = [x1, x2, x3, x4]
bins1 = np.arange(-7.5, 2.5, 0.2)
bins2 = np.arange(0, 10, 0.2)
bins3 = np.arange(7, 17, 0.2)
bins4 = np.arange(12, 22, 0.2)
bins = [bins1, bins2, bins3, bins4]
axis1 = [-7.5, 2.5, 0, 0.6]
axis2 = [0, 10, 0, 0.6]
axis3 = [7, 17, 0, 0.6]
axis4 = [12, 22, 0, 0.6]
axis = [axis1, axis2, axis3, axis4]
fig, ((GS1,GS2),(GS3, GS4)) = plt.subplots(2, 2, sharey = True)
GS = [GS1, GS2, GS3, GS4]
def update(curr):
if curr == 500:
a.event_source.stop()
for i in range(0, len(GS)):
GS[i].clear()
GS[i].hist(x[i][:curr], bins = bins[i], normed = True)
GS[i].axis(axis[i])
GS[i].set_title('Sampling random distribution')
GS[i].set_ylabel('Frequency')
GS[i].set_xlabel('Value')
GS[i].annotate('n = {}'.format(curr), [3,27])
plt.tight_layout()
fig = plt.gcf()
a = animation.FuncAnimation(fig, update, interval = 10)
plt.show()