lcensemble 0.3.4

Local Cascade Ensemble package

Local Cascade Ensemble (LCE) is a high-performing, scalable and user-friendly machine learning method for the general tasks of Classification and Regression.

In particular, LCE:

• Enhances the prediction performance of Random Forest and XGBoost by combining their strengths and adopting a complementary diversification approach

• Supports parallel processing to ensure scalability

• Handles missing data by design

• Adopts scikit-learn API for the ease of use

• Adheres to scikit-learn conventions to allow interaction with scikit-learn pipelines and model selection tools

• Is released in open source and commercially usable - Apache 2.0 license

An article introducing LCE and illustrative code examples has been published in Towards Data Science.

Installation

---

LCE package can be installed using pip:

pip install lcensemble

Documentation

---

LCE documentation, including API documentation and general examples, can be found here.

Reference

---

The full information about LCE can be found in the associated journal paper:

[1]

Fauvel, K., E. Fromont, V. Masson, P. Faverdin and A. Termier. XEM: An Explainable-by-Design Ensemble Method for Multivariate Time Series Classification. Data Mining and Knowledge Discovery, 36(3):917–957, 2022