pyMARS (MARS): Pure Python Earth (Multivariate Adaptive Regression Splines) #271
Description
Submitting Author: @edithatogo)
Package Name: pyMARS
One-Line Description of Package: Pure Python Implementation of Multivariate Adaptive Regression Splines (MARS)
Repository Link (if existing): https://github.com/edithatogo/mars
Code of Conduct & Commitment to Maintain Package
- I agree to abide by pyOpenSci's Code of Conduct during the review process and in maintaining my package after should it be accepted.
- I have read and will commit to package maintenance after the review as per the pyOpenSci Policies Guidelines.
Description
mars is a pure Python implementation of Multivariate Adaptive Regression Splines (MARS), inspired by the popular py-earth library by Jason Friedman and an R package earth by Stephen Milborrow. The goal of pymars is to provide an easy-to-install, scikit-learn compatible version of the MARS algorithm without C/Cython dependencies.
Scope
- [x] Data processing/munging
- [x] Data visualization
-
Explain how and why the package falls under these categories (briefly, 1-2 sentences):
This is a Pure Python implementation of py-earth, a scikit-learn project that was deprecated in 2023. This is a statistical/machine-learning method that is really useful, py-earth is the only Python implementation and is pinned to Python 2. This project creates a contemporary version, leveraging some additional features. -
Who is the target audience and what are the scientific applications of this package?
It's broad. I'm a physician and health economist, however economists, policy analysts, statisticians and anyone doing data science may find this useful. -
Are there other Python packages that accomplish similar things? If so, how does yours differ?
Yes, py-earth, which is archived. Others have started but not completed. This is feature complete. There are some additional features but it was designed so that existing py-earth code could be switched over without needing to change anything. The benefit is that with contemporary python and dependency support, it can be integrated into machine learning pipelines such as AutoML (not easy to do with py-earth). -
Any other questions or issues we should be aware of:
This wasn't intended to do anything other than provide a contemporary version of py-earth, however now that it's able to be integrated with modern pipelines, additional extensions and features have started to become apparent.
P.S. Have feedback/comments about our review process? Leave a comment:
I'm new to this. I essentially am keen for peer review to identify improvements, gain collaborators and ultimately submit it for publication, probably in Journal of Statistical Software.