JuniorTheoryCommon

What does this program print?

Python

import numpy as np

X = np.array([[0], [1], [3], [4], [11]])
centroids = np.array([[0.0], [5.0]])
dist = np.array([((X - c) ** 2).sum(axis=1) for c in centroids])
labels = dist.argmin(axis=0)
new = np.array([X[labels == j].mean(axis=0) for j in range(2)])
print(new[0, 0], new[1, 0])

0.5 6.0

Expecting 0.5 4.0: the centroid moves to the mean of its points 3, 4 and 11, which is 6.0; 4.0 is their median.
Expecting the centroids to move only part of the way: one k-means step puts each centroid exactly at the mean of its points, there is no step size.

Which centroid would the point 3 join if the second centroid started at 7.0 instead of 5.0?
What should the update step do if no point is assigned to a centroid?

Back to questions Open topic in roadmap →Answer it yourself — Quiz

Back to questions

JuniorTheoryCommon

What does this program print?

Python

import numpy as np

X = np.array([[0], [1], [3], [4], [11]])
centroids = np.array([[0.0], [5.0]])
dist = np.array([((X - c) ** 2).sum(axis=1) for c in centroids])
labels = dist.argmin(axis=0)
new = np.array([X[labels == j].mean(axis=0) for j in range(2)])
print(new[0, 0], new[1, 0])

0.5 6.0

Expecting 0.5 4.0: the centroid moves to the mean of its points 3, 4 and 11, which is 6.0; 4.0 is their median.
Expecting the centroids to move only part of the way: one k-means step puts each centroid exactly at the mean of its points, there is no step size.

Which centroid would the point 3 join if the second centroid started at 7.0 instead of 5.0?
What should the update step do if no point is assigned to a centroid?

Back to questions Open topic in roadmap →Answer it yourself — Quiz