Isolation Forest on time series
If you want to use all the information available, you can fit a multivariate outlier detector to the entire dataset. The multivariate approach also enables you to extract more features from time series to enhance model performance.
Practice creating new features from a DatetimeIndex and fitting an outlier detector on them using the apple dataset, which has already been loaded with a DatetimeIndex.
Also, recall the random_state parameter, which can be used to generate reproducible results.
This exercise is part of the course
Anomaly Detection in Python
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Create three new features from the DatetimeIndex
apple['day_of_week'] = ____
apple['month'] = ____
apple['day_of_month'] = _____