My Brain Cells

Easiest (and best) learning materials for anyone with a curiosity for machine learning and artificial intelligence, Deep learning, Programming, and other fun life hacks.

K Means Clustering

 

What is K-means Clustering?

K-means is a centroid-based algorithm, or a distance-based algorithm, where we calculate the distances to assign a point to a cluster. In K-Means, each cluster is associated with a centroid.

Applications of Clustering

Clustering has a large no. of applications spread across various domains. Some of the most popular applications of clustering are:

  • Recommendation engines
  • Market segmentation
  • Social network analysis
  • Search result grouping
  • Medical imaging
  • Image segmentation
  • Anomaly detection


K-Means using Numpy:
import io
import pandas as pd
import matplotlib.pyplot as plt
from scipy.spatial.distance import cdist
import numpy as np

df2 = pd.read_csv(io.BytesIO(uploaded[‘dataset_3.csv’]))
data = df2.values

print(data) 

def normalize(data):
     return (data – data.min(axis=0))/(data.max(axis=0)-data.min(axis=0))

data = normalize(data)

plt.plot(data[:,0],data[:,1],‘.’)
means = np.random.random((3,2))
plt.plot(data[:,0],data[:,1],‘.’)
plt.plot(means[0,0],means[0,1],‘ro’)
plt.plot(means[1,0],means[1,1],‘go’)
plt.show()

for i in range(5):
        labels = cdist(data,means).argmin(axis=1)
        for k in range (3):
            dataset = data[labels == k]
            plt.plot(dataset[:,0],dataset[:,1],‘.’)
            means[k] = dataset.mean(axis=0)
            plt.plot(means[k][0],means[k][1],‘X’)
        plt.show()
K-Means using Scikit Learn:

import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
from sklearn.datasets.samples_generator import make_blobs
from sklearn.cluster import KMeans
X, y = make_blobs(n_samples=300, centers=4, cluster_std=0.60, random_state=0)
plt.scatter(X[:,0], X[:,1])
wcss = []
for i in range(111):
    kmeans = KMeans(n_clusters=i, init=‘k-means++’, max_iter=300, n_init=10, random_state=0)
    kmeans.fit(X)
    wcss.append(kmeans.inertia_)
plt.plot(range(111), wcss)
plt.title(‘Elbow Method’)
plt.xlabel(‘Number of clusters’)
plt.ylabel(‘WCSS’)
plt.show()
kmeans = KMeans(n_clusters=4, init=‘k-means++’, max_iter=300, n_init=10, random_state=0)
pred_y = kmeans.fit_predict(X)
plt.scatter(X[:,0], X[:,1])
plt.scatter(kmeans.cluster_centers_[:, 0], kmeans.cluster_centers_[:, 1], s=300, c=‘red’)
plt.show()

Anthony

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top