Release notes#
1.0.7#
Released on 2024-04-15 - GitHub - PyPI
Enhancements
- Add compatibility with latest changes in scikit-learn 1.4+ (#54)
1.0.6#
Released on 2023-11-22 - GitHub - PyPI
Enhancements
- Added support for auto detection from byte file-like objects (#53).
Bugfixes
- Ensure to reseek before loading segmentation models if needed. This could cause unexpected crashes when auto detecting random forest pmml files (a4bed07).
1.0.5#
Released on 2023-11-05 - GitHub - PyPI
Enhancements
- Add support for non-seekable file-like objects (d0c0034)
Bugfixes
- Fix for loading file-like objects directly instead of providing a file path; caused a file already closed error (55de16a)
1.0.4#
Released on 2023-11-01 - GitHub - PyPI
Enhancements
- 🌟Added method to automatically detect the model type from the PMML file, and return the corresponding scikit-learn model (#51).
1.0.3#
Released on 2023-07-16 - GitHub - PyPI
Bugfixes
- Fix typo that caused memory leak using tree based models (#50)
1.0.2#
Released on 2023-04-25 - GitHub - PyPI
Bugfixes
- Fix for building on Apple Silicon (83519d9)
- Several changes to make
sklearn-pmml-model
work with the latest versions ofscikit-learn
.
1.0.1#
Released on 2022-08-15 - GitHub - PyPI
Bugfixes
- Correctly retrieve votes distribution for float target trees (#45)
1.0.0#
Released on 2022-02-03 - GitHub - PyPI
All major estimator types which are supported in both PMML and scikit-learn are now supported!
0.0.22#
Released on 2022-01-29 - GitHub - PyPI
Enhancements
- 🌟Added support for multi-layer perceptron estimators
PMMLMLPClassifier
andPMMLMLPRegressor
(#40).
0.0.21#
Released on 2022-01-24 - GitHub - PyPI
Bugfixes
- Improved compatibility with scikit-learn major release 1.0 (#37).
Enhancements
- 🌟Added support for k-nearest neighbor estimators
PMMLKNeighborsClassifier
andPMMLKNeighborsRegressor
(#38).
0.0.20#
Released on 2021-09-07 - GitHub - PyPI
Enhancements
- Support one-vs-rest logistic regression models (#36)
0.0.19#
Released on 2021-07-27 - GitHub - PyPI
Enhancements
- Updated automatic wheel building on CI to use GitHub Actions (#33)
0.0.18#
Released on 2021-07-27 - GitHub - PyPI
Bugfixes
- Fixed a problem that could block logistic regression from working in new versions of scikit-learn (4b6e11c).
- The input array shape now gets validated, and pandas dataframes gets subscripted and reordered to match the PMML file (2555898).
Enhancements
- 🌟Added support for support vector machine classification and regression (#31).
- Several improvements to documentation and code style.
0.0.17#
Released on 2021-06-16 - GitHub - PyPI
Bugfixes
- Fixed an issue with categorical features that occurred when categories contained spaces.
Enhancements
- 🌟Added support for tree-based regression, including decision trees (
PMMLTreeRegressor
), random forests (PMMLForestRegressor
) and gradient boosting (PMMLGradientBoostingRegressor
) (#25). - Added support for classification with linear models, with
PMMLLogisticRegression
for regression models, andPMMLRidgeClassifier
for general regression models (#24).
0.0.16#
Released on 2021-05-26 - GitHub - PyPI
Bugfixes
- Fix compatibility issues with the latest version of scikit-learn
- prevent inheritance problem related to
_more_tags
property
Enhancements
- 🌟Support gradient boosting classifiers, including categorical features (#20)
- Add an example that highlights the usage of
tree.decision_path
0.0.15#
Released on 2020-08-18 - GitHub - PyPI
Bugfixes
- Correctly parse ensemble trees with only a subset of target features. This happened to be the case for certain PMML models created with MatLab (#21)
0.0.14.1#
Released on 2020-08-18 - GitHub - PyPI
Bugfixes
- Prevent Buffer dtype mismatch on 64 bit windows (reported in #19)
0.0.14#
Released on 2020-07-22 - GitHub - PyPI
Bugfixes
- LocalTransformations are now parsed rather than only the global TransformationDictionary
Enhancements
- Added support for multiSplit decision trees
- Support average multipleModelStrategy for Random Forests
0.0.13#
Released on 2020-03-24 - GitHub - PyPI
Bugfixes
- Ensures compatibility with the latest scikit-learn versions.
0.0.12#
Released on 2020-03-24 - GitHub - PyPI
Bugfixes
- Define a minimum numpy version to ensure the library is compatible. Fixes issue #18.
0.0.11#
Released on 2020-03-24 - GitHub - PyPI
Enhancements
- Added support for Naive Bayes models
- Automatically build wheels on CI to make the installation from pip easier: no c compiler is required, which is especially useful on windows.
0.0.10#
Released on 2019-10-14 - GitHub - PyPI
Bugfixes
- Fixes output for tree_.threshold for categorical trees. This is useful for inspecting decision trees, as done in this sklearn example. Thanks to @kodonnell for reporting!