JuniorTheoryCommon

What does this program print?

Python

import numpy as np

X_train = np.array([[0, 0], [1, 3], [3, 0], [3, 3], [5, 5]])
y_train = np.array([0, 0, 0, 1, 1])
x = np.array([3, 2])
dist = np.sqrt(((X_train - x) ** 2).sum(axis=1))
order = np.argsort(dist)
print(y_train[order[0]], np.bincount(y_train[order[:3]]).argmax())

1 0

Expecting 1 1: the nearest point [3, 3] is class 1, but the next two, [3, 0] and [1, 3], are class 0, so three neighbors vote 0.
Expecting 0 for one neighbor because class 0 has more training points: with one neighbor only the closest point counts, and [3, 3] at distance 1.0 is class 1.

Which label would this code return with all five points as neighbors, and why?
How would weighting each vote by inverse distance change the answer for three neighbors?

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_train = np.array([[0, 0], [1, 3], [3, 0], [3, 3], [5, 5]])
y_train = np.array([0, 0, 0, 1, 1])
x = np.array([3, 2])
dist = np.sqrt(((X_train - x) ** 2).sum(axis=1))
order = np.argsort(dist)
print(y_train[order[0]], np.bincount(y_train[order[:3]]).argmax())

1 0

Expecting 1 1: the nearest point [3, 3] is class 1, but the next two, [3, 0] and [1, 3], are class 0, so three neighbors vote 0.
Expecting 0 for one neighbor because class 0 has more training points: with one neighbor only the closest point counts, and [3, 3] at distance 1.0 is class 1.

Which label would this code return with all five points as neighbors, and why?
How would weighting each vote by inverse distance change the answer for three neighbors?

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