LOF for the first time
LOF
differs from KNN
only in the internal algorithm and the lack of the method
parameter. Practice detecting outliers with it using contamination filtering on the scaled version of females
dataset from previous exercises.
The dataset has been loaded as females_transformed
.
This exercise is part of the course
Anomaly Detection in Python
Exercise instructions
- Import the
LOF
estimator from the relevantpyod
module. - Instantiate an
LOF()
with 0.3% contamination, 20 neighbors andn_jobs
set to -1. - Create a boolean index that returns
True
values when thelabels_
returned fromlof
are equal to 1. - Isolate the outliers from
females_transformed
usingis_outlier
.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Import LOF from its relevant module
from pyod.____ import ____
# Instantiate LOF and fit to females_transformed
lof = ____
lof.____
# Create a boolean index that checks for outliers
is_outlier = ____
# Isolate the outliers
outliers = ____
print(len(outliers))