Unsupervised-Learning.py

import pandas as pd  
from sklearn.cluster import KMeans  

dataset = pd.read_csv('https://storage.googleapis.com/dqlab-dataset/pythonTutorial/mall_customers.csv')   
X = dataset[['annual_income','spending_score']]  

cluster_model = KMeans(n_clusters = 5, random_state = 24)  
labels = cluster_model.fit_predict(X)

#import library
import matplotlib.pyplot as plt
#convert dataframe to array
X = X.values
#Separate X to xs and ys --> use for chart axis
xs = X[:,0]
ys = X[:,1]
# Make a scatter plot of xs and ys, using labels to define the colors
plt.scatter(xs, ys, c=labels,  alpha = 0.5)

# Assign the cluster centers: centroids
centroids = cluster_model.cluster_centers_
# Assign the columns of centroids: centroids_x, centroids_y
centroids_x = centroids[:,0]
centroids_y = centroids[:,1]
# Make a scatter plot of centroids_x and centroids_y
plt.scatter(centroids_x, centroids_y, marker='D', s=50)
plt.title('K Means Clustering', fontsize = 20)
plt.xlabel('Annual Income')
plt.ylabel('Spending Score')
plt.show()

#Elbow Method - Inertia plot
inertia = []
#looping the inertia calculation for each k
for k in range (1,10):
    #Assign KMeans as cluster_model
    cluster_model = KMeans(n_clusters = k, random_state=24)
    #Fit the model to X
    cluster_model.fit(X)
    #Get the inertia value
    inertia_value = cluster_model.inertia_
    #Append the inertia value to inertia list
    inertia.append(inertia_value)

#Inertia plot
plt.plot(range(1,10), inertia)
plt.title('The Elbow Method - Inertia plot', fontsize=20)
plt.xlabel('No. of Cluesters')
plt.ylabel('Inertia')
plt.show()