bevel
bevel copied to clipboard
Shopify
Ordinal regression in Python
It should be straightforward to implement an L2 regularization of linear ordinal regression. Doing so for L1 and/or then writing tests will be more challenging.
Overall, it's a deep and thorough article on why to use ordinal regression when you have ordinal data. https://osf.io/9h3et/download > Arguing that it is okay to use a metric model...
Remove burden/surprise from user to convert the type. Scikitlearn automatically converts it - I think bevel should too
👍 https://www.jstor.org/stable/2984952?seq=1#page_scan_tab_contents
In setup.py, there is ```python try: import pypandoc readme_rst = pypandoc.convert_file(readme_md, 'rst') except(ImportError): readme_rst = open(readme_md).read() ``` that will convert the markdown readme to an rst file. This was because...
- [ ] One or two pedagogical examples on toy datasets would be very useful. - [ ] Upload notes that establish math conventions
ex: ```python from bevel import OrderedLogit w = np.array([1, 2., 3.4, 1.1, 0.5]) orf = OrderedLogit() orf.fit(X, y, weights=w) ``` And defaults to an array of 1s if not provided.
Post solution here: https://stats.stackexchange.com/questions/168262/ordinal-logistic-regression-in-python
https://stats.stackexchange.com/questions/89474/interpretation-of-ordinal-logistic-regression