How to make an animation show each change in the dataset? With matplotlib

Question

I am using matplotlib to show my 3d dataset. But I would like to show each data set change in a 3D graph. And create an animation of that or maybe save the animation. Follow the code I’m using:

from mpl_toolkits.mplot3d import Axes3D
from matplotlib import cm
import matplotlib.pyplot as plt
from matplotlib.animation import FuncAnimation
import numpy as np
import random
import time

K = 3
ATTRIBUTE_COUNT = 4
_random = random.seed()
data = []
data_classes = {}


def manhattan_distance(point_x, point_y):
    distance, length = 0, len(point_x)
    for i in range(length):
        distance += abs( point_x[i] - point_y[i])
    return distance

def get_centroid(cluster):
    summ = []
    for j in range(ATTRIBUTE_COUNT):
        summ.append(0)
    for element in cluster:
        for index, attribute in enumerate(element):
            summ[index] += attribute
    if len(cluster) != 0:
        new_centroid = list(map(lambda x: x/len(cluster), summ))
        return new_centroid
    else:
        return None

#read data set from file
with open('iris.data') as f:
    for line in f:
        line_data = line.split(',I')
        classe_name = 'I' + line_data[1]
        point = list((map(float, line_data[0].split(','))))
        data.append(point)
        data_classes[str(point)] = classe_name.rstrip()

# I am just showing the original data set.
aux = [[],[],[],[]]
for point in data:
    aux[0].append(point[0])
    aux[1].append(point[1])
    aux[2].append(point[2])
    aux[3].append(point[3])
fig = plt.figure(1)
fig.suptitle('Data Set')
ax = fig.gca(projection='3d')
ax.scatter(aux[0],aux[1],aux[2], c='b', s=[25 * x for x in aux[3]],cmap=plt.hot())
plt.show()

centroids = []
centroids = random.sample(data, K)
changed_centroid = True

while(changed_centroid):
    clusters = []
    new_centroids = []
    for i in range(K):
        clusters.append([])

    for element in data:
        distances = []
        for centroid in centroids:
            distances.append(manhattan_distance(element, centroid))
        index = distances.index(min(distances))#index of minimuim distance
        clusters[index].append(element)

    #for each cluster I would like to show it in a different color.
    for cluster in clusters:
        new_centroid = get_centroid(cluster)
        new_centroids.append(new_centroid)

    if new_centroids == centroids:
        changed_centroid = False
    centroids = new_centroids

    #for centroid in centroids:
       #for each centroid I would like to show like a triangle.

#show result of classification
for index, cluster in enumerate(clusters):
    print('classe', index ,'------------------------------')
    for point in cluster:
        print(data_classes[str(point)], end=' ')
    print('\n')