Perform k-means clustering over multiple columns and get the cluster center values?

Question

I read here how to show the number of clusters over $n$ columns.

I would like to know how to get in a table, the values of the clusters centers. Could someone help me with this?

The KMeans class stores the cluster centers in the cluster_centers_ attribute, which returns an array of shape (n_clusters, n_features) (see also the documentation). — Oxbowerce, Sep 10 '22 at 15:42

score 0 · Accepted Answer · answered Sep 10 '22 at 20:12

sklearn.clusters.KMeans has an attribute cluster_centers_, which stores the array of cluster centers.

You can add them to the dataframe as new columns this way:

clusters = KMeans(n_clusters = n)
predict = clusters.fit_predict(data)
centers = pd.DataFrame(clusters.cluster_centers_[predict, :])
centers.index = data.index
data = pd.concat([data, centers], axis=1)

Perform k-means clustering over multiple columns and get the cluster center values?

1 Answers1