Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Set parameter of ElasticNetRegression to set the ratio between L1 (lasso) and L2 (ridge) regularization #166

Closed
lars-reimann opened this issue Apr 4, 2023 · 1 comment · Fixed by #237
Closed

Set parameter of ElasticNetRegression to set the ratio between L1 (lasso) and L2 (ridge) regularization #166

lars-reimann opened this issue Apr 4, 2023 · 1 comment · Fixed by #237
Assignees
@zzril @robmeth
Labels
enhancement 💡 New feature or request good first issue Good for newcomers released Included in a release

Comments

@lars-reimann
Copy link
Member

lars-reimann commented Apr 4, 2023 •
edited
Loading

Is your feature request related to a problem?

It's not possible to configure in the elastic net regression model whether it should apply more L1 or L2 regularization.

Desired solution

  • Add a new parameter lasso_ratio: float to the initializer of safeds.ml.regression.ElasticNetRegression
  • Raise a value error if lasso_ratio < 0 or lasso_ratio > 1
  • Warn if lasso_ratio == 0 -> Users should use RidgeRegression instead
  • Warn if lasso_ratio == 1 -> Users should use LassoRegression instead
  • Pass it along to l1_ratio of the wrapped scikit-learn model in the fit method

Possible alternatives (optional)

No response

Screenshots (optional)

No response

Additional Context (optional)

  • Use warnings.warn to warn.
  • Use with pytest.warns to test that the warning is created.
@github-project-automation github-project-automation bot moved this to Backlog in Library Apr 4, 2023
@lars-reimann lars-reimann added the enhancement 💡 New feature or request label Apr 4, 2023
@lars-reimann lars-reimann added the good first issue Good for newcomers label Apr 13, 2023
@lars-reimann lars-reimann changed the title Set parameter of ElasticNet to set the ratio between L1 (lasso) and L2 (ridge) regularization Set parameter of ElasticNetRegression to set the ratio between L1 (lasso) and L2 (ridge) regularization Apr 21, 2023
@alex-senger alex-senger moved this from Backlog to Todo in Library Apr 21, 2023
@robmeth robmeth moved this from Todo to In Progress in Library Apr 21, 2023
@robmeth robmeth mentioned this issue Apr 21, 2023
Merged
@robmeth robmeth linked a pull request Apr 21, 2023 that will close this issue
feat: add parameter lasso_ratio to ElasticNetRegression #237
Merged
@robmeth robmeth moved this from In Progress to Ready for Review in Library Apr 21, 2023
lars-reimann pushed a commit that referenced this issue Apr 21, 2023
@robmeth @megalinter-bot
feat: add parameter lasso_ratio to ElasticNetRegression (#237)
4a1a736 
Closes #166.

### Summary of Changes

Added parameter `lasso_ratio` to `ElasticNetRegression` and tests for
edge cases 0, 1, invalid and default.

---------

Co-authored-by: zzril <>
Co-authored-by: megalinter-bot <[email protected]>
@github-project-automation github-project-automation bot moved this from Ready for Review to ✔️ Done in Library Apr 21, 2023
lars-reimann pushed a commit that referenced this issue May 11, 2023
@semantic-release-bot
chore(release): 0.12.0 [skip ci]
9189356 
## [0.12.0](v0.11.0...v0.12.0) (2023-05-11)

### Features

* add `learning_rate` to AdaBoost classifier and regressor. ([#251](#251)) ([7f74440](7f74440)), closes [#167](#167)
* add alpha parameter to `lasso_regression` ([#232](#232)) ([b5050b9](b5050b9)), closes [#163](#163)
* add parameter `lasso_ratio` to `ElasticNetRegression` ([#237](#237)) ([4a1a736](4a1a736)), closes [#166](#166)
* Add parameter `number_of_tree` to `RandomForest` classifier and regressor ([#230](#230)) ([414336a](414336a)), closes [#161](#161)
* Added `Table.plot_boxplots` to plot a boxplot for each numerical column in the table ([#254](#254)) ([0203a0c](0203a0c)), closes [#156](#156) [#239](#239)
* Added `Table.plot_histograms` to plot a histogram for each column in the table ([#252](#252)) ([e27d410](e27d410)), closes [#157](#157)
* Added `Table.transform_table` method which returns the transformed Table ([#229](#229)) ([0a9ce72](0a9ce72)), closes [#110](#110)
* Added alpha parameter to `RidgeRegression` ([#231](#231)) ([1ddc948](1ddc948)), closes [#164](#164)
* Added Column#transform ([#270](#270)) ([40fb756](40fb756)), closes [#255](#255)
* Added method `Table.inverse_transform_table` which returns the original table ([#227](#227)) ([846bf23](846bf23)), closes [#111](#111)
* Added parameter `c` to `SupportVectorMachines` ([#267](#267)) ([a88eb8b](a88eb8b)), closes [#169](#169)
* Added parameter `maximum_number_of_learner` and `learner` to `AdaBoost` ([#269](#269)) ([bb5a07e](bb5a07e)), closes [#171](#171) [#173](#173)
* Added parameter `number_of_trees` to `GradientBoosting` ([#268](#268)) ([766f2ff](766f2ff)), closes [#170](#170)
* Allow arguments of type pathlib.Path for file I/O methods ([#228](#228)) ([2b58c82](2b58c82)), closes [#146](#146)
* convert `Schema` to `dict` and format it nicely in a notebook ([#244](#244)) ([ad1cac5](ad1cac5)), closes [#151](#151)
* Convert between Excel file and `Table` ([#233](#233)) ([0d7a998](0d7a998)), closes [#138](#138) [#139](#139)
* convert containers for tabular data to HTML ([#243](#243)) ([683c279](683c279)), closes [#140](#140)
* make `Column` a subclass of `Sequence` ([#245](#245)) ([a35b943](a35b943))
* mark optional hyperparameters as keyword only ([#296](#296)) ([44a41eb](44a41eb)), closes [#278](#278)
* move exceptions back to common package ([#295](#295)) ([a91172c](a91172c)), closes [#177](#177) [#262](#262)
* precision metric for classification ([#272](#272)) ([5adadad](5adadad)), closes [#185](#185)
* Raise error if an untagged table is used instead of a `TaggedTable` ([#234](#234)) ([8eea3dd](8eea3dd)), closes [#192](#192)
* recall and F1-score metrics for classification ([#277](#277)) ([2cf93cc](2cf93cc)), closes [#187](#187) [#186](#186)
* replace prefix `n` with `number_of` ([#250](#250)) ([f4f44a6](f4f44a6)), closes [#171](#171)
* set `alpha` parameter for regularization of `ElasticNetRegression` ([#238](#238)) ([e642d1d](e642d1d)), closes [#165](#165)
* Set `column_names` in `fit` methods of table transformers to be required ([#225](#225)) ([2856296](2856296)), closes [#179](#179)
* set learning rate of Gradient Boosting models ([#253](#253)) ([9ffaf55](9ffaf55)), closes [#168](#168)
* Support vector machine for regression and for classification ([#236](#236)) ([7f6c3bd](7f6c3bd)), closes [#154](#154)
* usable constructor for `Table` ([#294](#294)) ([56a1fc4](56a1fc4)), closes [#266](#266)
* usable constructor for `TaggedTable` ([#299](#299)) ([01c3ad9](01c3ad9)), closes [#293](#293)

### Bug Fixes

* OneHotEncoder no longer creates duplicate column names ([#271](#271)) ([f604666](f604666)), closes [#201](#201)
* selectively ignore one warning instead of all warnings ([#235](#235)) ([3aad07d](3aad07d))
@lars-reimann
Copy link
Member Author

🎉 This issue has been resolved in version 0.12.0 🎉

The release is available on:

Your semantic-release bot 📦🚀

@lars-reimann lars-reimann added the released Included in a release label May 11, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@zzril zzril

@robmeth robmeth

Labels
enhancement 💡 New feature or request good first issue Good for newcomers released Included in a release
Projects
Library
Archived in project
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

feat: add parameter lasso_ratio to ElasticNetRegression Safe-DS/Library
3 participants
@lars-reimann @zzril @robmeth