Major (breaking) changes:

Full compatibility with scikit-learn estimators!
Renamed distributions and estimators for more concise and consistent naming scheme

What's Changed

Make estimators compatible with sklearn by @simon-hirsch in https://github.com/simon-hirsch/ondil/pull/128
Cleanup exports by @BerriJ in https://github.com/simon-hirsch/ondil/pull/139
Fix docs by @BerriJ in https://github.com/simon-hirsch/ondil/pull/140
Add Zero Inflated Beta Distribution by @murthy-econometrics-5819 in https://github.com/simon-hirsch/ondil/pull/132
corrected IC formulas and documetation by @flziel in https://github.com/simon-hirsch/ondil/pull/143
rename T() to StudentT() by @simon-hirsch in https://github.com/simon-hirsch/ondil/pull/142
Fixes by @simon-hirsch in https://github.com/simon-hirsch/ondil/pull/146
Add Zero Adjusted Gamma Distribution by @murthy-econometrics-5819 in https://github.com/simon-hirsch/ondil/pull/130

New Contributors

@flziel made their first contribution in https://github.com/simon-hirsch/ondil/pull/143

Full Changelog: https://github.com/simon-hirsch/ondil/compare/v0.3.2...v0.4.0

- Python
Published by simon-hirsch 5 months ago

New distributions

Add distribution: InverseGaussian by @simon-hirsch in https://github.com/simon-hirsch/ondil/pull/118
Add distribution: BetaInflated by @murthy-econometrics-5819 in https://github.com/simon-hirsch/ondil/pull/99
Add distribution: Gumbel by @simon-hirsch in https://github.com/simon-hirsch/ondil/pull/120
Add distribution: InverseGamma by @Jaiminoza229984 in https://github.com/simon-hirsch/ondil/pull/125
Add distribution: ReverseGumbel by @Jaiminoza229984 in https://github.com/simon-hirsch/ondil/pull/126

Fixes and maintenance

Parameterize tests for distributions by @BerriJ in https://github.com/simon-hirsch/ondil/pull/123
Improve OnlineScaler by @BerriJ in https://github.com/simon-hirsch/ondil/pull/127
Pass sample weights to scaler by @simon-hirsch in https://github.com/simon-hirsch/ondil/pull/131

Full Changelog: https://github.com/simon-hirsch/ondil/compare/v0.3.1...v0.3.2

- Python
Published by simon-hirsch 7 months ago

New distributions

Beta continuous implementation by @murthy-econometrics-5819 in https://github.com/simon-hirsch/ondil/pull/84
Logistic Distribution Implementation by @Jaiminoza229984 in https://github.com/simon-hirsch/ondil/pull/86
Exponential Distribution Implementation by @Jaiminoza229984 in https://github.com/simon-hirsch/ondil/pull/101

Maintenance

Improvements for docstrings by @BerriJ in https://github.com/simon-hirsch/ondil/pull/95
Improvements for testing using R and r2py by @BerriJ in https://github.com/simon-hirsch/ondil/pull/90
Various small refactorings @simon-hirsch
Create template for MR for dists by @simon-hirsch in https://github.com/simon-hirsch/ondil/pull/97

New Contributors

@murthy-econometrics-5819 made their first contribution in https://github.com/simon-hirsch/ondil/pull/84

Full Changelog: https://github.com/simon-hirsch/ondil/compare/v0.3.0...v0.3.1

- Python
Published by simon-hirsch 7 months ago

What's Changed

This is the first release of ondil - formerly known as rolch.

Please update your installation, forks and paths and mind that we have changed the licence MIT -> GPL v3 as in the R gamlss package.

Full Changelog: https://github.com/simon-hirsch/ondil/compare/v0.2.5...v0.3.0

- Python
Published by simon-hirsch 7 months ago

ondil - v0.2.5 - rename package

This is the last release of rolch. We will continue the development under the new name ondil.

Changelog: added depreciation warning on import @simon-hirsch

- Python
Published by simon-hirsch 7 months ago

New distribution

LogNormal and LogNormalMedian distribution implementation by @Jaiminoza229984

Fixes & Maintenance

Allow multi step online mean updates in the utils function by @simon-hirsch
Restructure distribution tests by @BerriJ
Delete unused files by @simon-hirsch

New Contributors

@Jaiminoza229984 made their first contribution

- Python
Published by simon-hirsch 7 months ago

ondil - v0.2.3

New Methods: Ridge and ElasticNet regression @BerriJ
New Distribution: Normal with Mean Variance Parameterization @simon-hirsch
New Link Functions: InverseSoftPlus and friends @simon-hirsch
Line Tracking and adaptive stepsize for Newton-Raphson step @simon-hirsch
More logging and explicit warnings @simon-hirsch
More docs for many things @BerriJ @simon-hirsch

- Python
Published by simon-hirsch 8 months ago

This release includes - a nice overhaul of the distributions backend by @BerriJ - sanity checks for distribution and link function compatibility - sanity checks for the distribution support - some new docs for building new links and distributions - more functionality for the debug mode by @simon-hirsch

- Python
Published by simon-hirsch 10 months ago

ondil - v0.2.1

Improving the scaler: - Allow for scaling of certain variables only. See https://simon-hirsch.github.io/rolch/scaler/ for details. You can now pass either a bool into scale_inputs or a numpy array with indices of the columns of $X$ you'd like to scale.

Some fixes: - Add MLE estimates as start values to improve convergence in the beginning. This is more robust, since the previous start values might have been overly optimistic. - Allow for boolean arrays in the equation and the scale_inputs.

- Python
Published by simon-hirsch 11 months ago

ondil - v0.2.0

This release introduces the distinction between the Estimator() classes, which provides the .fit(), .update(), .predict() methods and the EstimationMethod() classes, which do the actual fitting of $X$, $y$ to coefficients / weights / betas. This gives

easier integration of new estimation methods
easier handling of default parameters for methods
easier handling of non-default parameters for methods, especially for non-standard parameters like bounds, regularization strengths, etc..
much cleaner code in OnlineGamlss()

Breaking change: This change gets rid of the estimation_kwargs parameter in OnlineGamlss() which was cumbersome and poorly documented anyways.

Furthermore, we - align the API of OnlineGamlss() and OnlineLinearModel() and derive the OnlineLasso() from OnlineLinearModel() to show the flexibility of the new approach. - Introduce a verbosity parameter for OnlineGamlss() to print information to the user. - Add some properties to (slowly) align more to the sklearn API - Some minor fixes like fixing #36 - Add a lot of documentation

- Python
Published by simon-hirsch 12 months ago

ondil - v0.1.11

Add OnlineLasso() Estimator to docs and ensure basic functionality
Fix bugs in OnlineLasso()
Add some link derivatives and second derivatives of link functions

- Python
Published by simon-hirsch about 1 year ago

ondil - v0.1.10

This release implements an important fix and does some maintenance and improvements in the backend.

Fixes - Don't allow inversion of rank-deficit Gramian matrices - thanks to @katche1010 for spotting.

Improvements - Allow batch updates of Gramian and Inverse Gramians - Proper naming of links and their derivatives

- Python
Published by simon-hirsch about 1 year ago

ondil - v0.1.9 - Major API Improvements

This release implements a more structured API for model estimation. We introduce the equation dictionary, which specifies the model for each distribution parameter. Additionally, we fit intercepts for each distribution parameter per defaul now.

Fixes issues #23, #22 by @BerriJ, @simon-hirsch.

Much appreciated feedback on the API design by Franz Kiraly.

- Python
Published by simon-hirsch about 1 year ago