Researchers at UC Berkeley have developed a Python library called ‘imodels’, a python package with cutting-edge techniques for concise, transparent, and accurate predictive modelling.
There are a myriad of challenges involved in implementing interpretable models. Implementations are frequently challenging to locate, utilise and compare. Despite the development of many methods for fitting interpretable models, implementations for such models are often difficult to find, use, and compare. ‘imodels’ aims to fill this gap by providing a simple unified interface and implementation for many state-of-the-art interpretable modelling techniques. ‘imodels’ has implementations of various such methods, including RuleFit, Bayesian Rule Lists, FIGS, Optimal Rule Lists etc.
imodels is extremely simple and can be easily installed. Once installed, one just has to import a classifier or regressor and use the fit and predict methods.
Interpretable modelling offers an alternative to common black-box modelling, and in many cases can offer massive improvements in terms of efficiency and transparency without suffering from a loss in performance.