This is a simple package implementation for the isolation forest method described (among other places) in this paper for detecting anomalies and outliers from a data point distribution.
For an extended version of this algorithm that produces more precise scoring maps please visit this repository
https://github.com/sahandha/eif/
pip install iso_forest
or directly from the Github repository
pip install git+https://github.com/mgckind/iso_forest.git
It supports python2 and python3
- numpy
No extra requirements are needed for the algorithm.
In addition, it also contains means to draw the trees created using the igraph library.
See these 2 notebooks examples on how to use it
- Initial Release