Ross Diener
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.
👍 https://www.jstor.org/stable/2984952?seq=1#page_scan_tab_contents
- [ ] One or two pedagogical examples on toy datasets would be very useful. - [ ] Upload notes that establish math conventions
Post solution here: https://stats.stackexchange.com/questions/168262/ordinal-logistic-regression-in-python
Variables like `X_scale` have a different meaning in sklearn. It would be nice to be consistent with the industry standard where possible (or at least not inconsistent).
This paper talks about the benefit of using binary classification for ordinal regression: https://papers.nips.cc/paper/3125-ordinal-regression-by-extended-binary-classification.pdf The major benefits are: 1) no need to invent new loss function and optimization; 2) whatever...