Python implementation of COBRA algorithm with regression analysis

pycobra

pycobra is a python library for ensemble learning. It serves as a toolkit for regression and classification using these ensembled machines, and also for visualisation of the performance of the new machine and constituent machines. Here, when we say machine, we mean any predictor or machine learning object - it could be a LASSO regressor, or even a Neural Network. It is scikit-learn compatible and fits into the existing scikit-learn ecosystem.

pycobra offers a python implementation of the COBRA algorithm introduced by Biau et al. (2016) for regression.

Another algorithm implemented is the EWA (Exponentially Weighted Aggregate) aggregation technique (among several other references, you can check the paper by Dalalyan and Tsybakov (2007).

Apart from these two regression aggregation algorithms, pycobra implements a version of COBRA for classification. This procedure has been introduced by Mojirsheibani (1999).

pycobra also offers various visualisation and diagnostic methods built on top of matplotlib which lets the user analyse and compare different regression machines with COBRA. The Visualisation class also lets you use some of the tools (such as Voronoi Tesselations) on other visualisation problems, such as clustering.

Documentation and Examples

The notebooks directory showcases the usage of pycobra, with examples and basic usage. The documentation page further covers how to use pycobra.

Installation

Run python setup.py install for default installation.

Run python setup.py test to run all tests.

Run pip install . to install from source.

Dependencies

• Python 2.7+, 3.4+

• numpy, scipy, scikit-learn, matplotlib

References

• G. Biau, A. Fischer, B. Guedj and J. D. Malley (2016), COBRA: A combined regression strategy, Journal of Multivariate Analysis.

• M. Mojirsheibani (1999), Combining Classifiers via Discretization, Journal of the American Statistical Association.

• A. S. Dalalyan and A. B. Tsybakov (2007) Aggregation by exponential weighting and sharp oracle inequalities, Conference on Learning Theory.

Project details

Uploaded source