2.1.2: Minor update

Add n_features_in_

Inclusion of logistic regression for classification

We're thrilled to announce the release of version 2.1.0 of the asgl package, bringing powerful enhancements to logistic regression models. Below are the key highlights of this release:
Enhanced Logistic Regression Capabilities: Users can now seamlessly address binary classification problems by setting model='logit'. For those requiring more granular control, the options model='logit_raw' and model='logit_proba' are available, providing outputs before logistic transformation and probability outputs, respectively.
Advanced Penalization Options: This update also introduces the implementation of ridge and adaptive ridge penalizations, accessible via penalization='ridge' or penalization='aridge'. These features allow for more flexible model tuning, enhancing adaptability and precision.

Major update

We're excited to announce the release of version 2.0.0 of the asgl package, which introduces significant enhancements and new features to improve usability and performance. Below are the key highlights of this release:

Scikit-learn Compatibility: The new Regressor class now offers full compatibility with scikit-learn, allowing users to leverage scikit-learn's powerful tools for hyperparameter optimization, model evaluation, and performance metrics. This integration enables seamless use of functions like sklearn.model_selection.GridSearchCV for tuning and evaluating models.

Deprecation of ASGL Class: The old ASGL class is still available but will now raise a DeprecationWarning. It remains functional but is no longer supported and will be removed in future versions. Users are encouraged to transition to the new Regressor class for continued support and access to the latest features.

This release marks a significant step forward in making the asgl package more versatile and user-friendly, particularly for those already familiar with the scikit-learn ecosystem. We highly recommend users to update to this latest version to take advantage of these improvements.

For detailed information on changes and how to migrate to the new Regressor class, please refer to the updated README and documentation.

Update

1.0.6

Update setup.py

Release 8

v1.0.5

Update solver logic

Release 7

v1.0.4

Added GroupKFold in the cross validation process

Release 6

v1.0.3

Update alasso

Release 5

v1.0.2

add logging

Release 4

v1.0.1

Update setup.py

Third release: change to 1.0 version

Major update

Include alasso, agl, and lasso weight estimation alternative. Update long_description and user_guide