聚类常见算法K-means & DBSCAN

一.K-means Clustering

1.读取数据

import pandas as pd
beer = pd.read_csv('data.txt', sep=' ')
beer

2.只取有用的四列数据

X = beer[["calories","sodium","alcohol","cost"]]
X.head()

3.使用Kmeans对X进行分类

from sklearn.cluster import KMeans

km = KMeans(n_clusters=3).fit(X)
km2 = KMeans(n_clusters=2).fit(X)
km.labels_

4.根据分类来排序

beer['cluster'] = km.labels_
beer['cluster2'] = km2.labels_
beer.sort_values('cluster')

5.cluster=3的平均值

beer.groupby("cluster").mean()

6.cluster=2的平均值

beer.groupby("cluster2").mean()

7.reset_index

centers = beer