How to get N numbers of data points which are nearest from a cluster's center?

Question

I want to get N nearest data points from center (based on Euclidean Distance) in each cluster after deploying K-means algorithm. I am able to get the indices of data points using

np.where(km.labels_ == 0)

Answer 1

You can use the transform method of the kmeans class which calculates the distance of each data point to each of the cluster.

Then assuming you want the top N points from the 0th index cluster then you can just do:

cluster = 0
N = 2
np.sort(kmeans.transform(X)[:,cluster])[:N]