Clustering using Scipy k-means using simple Euclidean distance - Cacher Snippet

7/4/2016 - 1:10 AM

Clustering using Scipy k-means using simple Euclidean distance

Clustering using Scipy k-means using simple Euclidean distance

clustering_kmeans_euclidean.py

from scipy.cluster.vq import kmeans,vq

# input
data = DATA[["real"]]
ncores = 10

# computing K-Means with K = ncores (ncores clusters)
centroids,_ = kmeans(data,ncores,, max_iter = 300, n_jobs = 3) # values of centroids
# assign each value of dataset a respective cluster index (between 1 and ncores)
idx,_ = vq(data,centroids)