pycobra: A Python toolkit for ensemble regression analysis and visualisation

Benjamin Guedj (1) , Bhargav Srinivasa Desikan (1)

1 MODAL - MOdel for Data Analysis and Learning

We introduce pycobra, a Python library devoted to ensemble regression and visualisation. Its main assets are the implementation of the COBRA algorithm (an ensemble method introduced by Biau et al., 2016), a flexible and generic interface to compare most regression methods available in Python libraries (including, but not limited to, scikit-learn), and visualisation tools based on Voronoi tessellations. pycobra is released under the MIT open-source license and can be downloaded from the Python Package Index (PyPi) and Machine Learning Open Source Software. The current development (along with Jupyter notebooks) is available at https://github.com/bhargavvader/pycobra.

Mots clés: Regression and Ensemble methods; Machine learning; Voronoi tesselation; Python; Open Source Software

Domaines: Machine Learning [stat.ML]; Méthodologie [stat.ME]

Benjamin Guedj, Bhargav Srinivasa Desikan. pycobra: A Python toolkit for ensemble regression analysis and visualisation. 2017. ⟨hal-01514059v1⟩