Find the neighbor
It is clear that the local outlier factor algorithm depends a lot on the idea of a nearest neighbor, which in turn depends on the choice of distance metric. So you decide to experiment some more with the hepatitis dataset introduced in the previous lesson. You are given three examples stored in features
, whose classes are stored in labels
. You will identify the nearest neighbor to the first example (row with index 0
) using three different distance metrics, Euclidean, Hamming and Chebyshev, and on the basis of that choose which distance metric to use. You will import the necessary module as part of the exercise, but pandas
and numpy
already available, as are features
and their labels labels
.
This exercise is part of the course
Designing Machine Learning Workflows in Python
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Import DistanceMetric as dm
from sklearn.____ import ____ as dm