Ridge regression with fraction regularization
Project description
fracridge
Is an implementation of fractional ridge regression (FRR).
Installation:
Matlab
Download and copy the files from the [https://github.com/arokem/fracridge/tree/master/matlab](Matlab directory) into your Matlab path.
Python
To install the release version:
pip install fracridge
Or to install the development version:
pip install -r requirements.txt
pip install .
Usage
Matlab
[coef,alphas] = fracridge(X,fracs,y,tol,mode)
Python
There's a functional API:
from fracridge import fracridge
coefs, alphas = fracridge(X, y, fracs)
Or a sklearn-compatible OO API:
from fracridge import FracRidge
fr = FracRridge(fracs=fracs)
fr.fit(X, y)
coefs = fr.coef_
alphas = fr.alpha_
How to cite
"Improved hyperparameter selection through fractional ridge regression", Rokem & Kay (in preparation)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
fracridge-0.2.1.tar.gz
(88.4 kB
view hashes)
Built Distribution
Close
Hashes for fracridge-0.2.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | b28a8b7d463fe8dcfd6fb7d82f0a8ad6328318d259971c104785c9039aa67251 |
|
MD5 | a3ce1d894325af949bf0e245503ea066 |
|
BLAKE2b-256 | 02544b5e3c1bce19a45b20bffb5af1a0333ab446000d1ef9f36a86b7b8115278 |