Perform K-means Clustering

Tarefa

Swipe to start coding

Let's check the efficiency of the algorithm on different types of clusters. Now we will use the three built-in datasets of the sklearn library and try to use the K-means algorithm to cluster the corresponding points. We will provide visualizations and try to estimate the quality of clustering using these visualizations.

Your task is to use the K-means clustering algorithm and to solve 3 different clustering problems. Compare the results and make conclusions about clustering quality. You have to:

Use KMeans class from cluster module for import.
Use KMeans class to instantiate a class object
Use.fit()method to train model.
Use .labels_attribute to extract fitted clusters.

_{Once you've completed this task, click the button below the code to check your solution.}

Solução

from sklearn.datasets import make_blobs, make_moons, make_circles

import matplotlib.pyplot as plt

import numpy as np

from sklearn.cluster import KMeans

import warnings

warnings.filterwarnings('ignore')

def check_clustering_quality(X, y, n_clusters):

fig, axes = plt.subplots(1,2)

axes[0].scatter(X[:, 0], X[:, 1], c=y, cmap='tab20b')

axes[0].set_title('Real clusters')

kmeans=KMeans(n_clusters = n_clusters).fit(X)

axes[1].scatter(X[:, 0], X[:, 1], c=kmeans.labels_, cmap='tab20b')

axes[1].set_title('Clusters with K-means')

X, y = make_blobs(n_samples=500, cluster_std=1, centers=3)

check_clustering_quality(X, y, 3)

X, y = make_moons(n_samples=500)

check_clustering_quality(X, y, 2)

X, y = make_circles(n_samples=500)

check_clustering_quality(X, y, 2)

Note

In visualizations, it is necessary to look not at the color of clusters, but at the relative position of points in real and predicted clusters (Python can color the same clusters with different colors in different pictures due to implementation features)

Tudo estava claro?

Obrigado pelo seu feedback!