Category-encoders

Latest version: v2.6.4

Safety actively analyzes 681866 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 3 of 6

2.1.0

======

* Added experimental support for multithreading in hashing encoder
* Support for pandas >=0.24
* Removed support for missing values represented by None due to changes in Pandas 0.24. Use numpy.NaN
* Changed the default setting of Helmert encoder for handle_missing and handle_unknown
* Fixed wrong calculation in m-estimate encoder
* Fixed missing value handling in CatBoost encoder

2.0.0

======

* Added James-Stein, CatBoost and m-estimate encoders
* Added get_feature_names method
* Refactored treatment of missing and new values
* Speed up the encoders with vectorization
* Improved compatibility with Pandas Series and Numpy Arrays

1.3.0

======

* Added Weight of Evidence encoder

1.2.8

======

* Critical bugfix in hashing encoder

1.2.7

======

* Bugfixes related to missing value imputation
* Category names optionally added to encoded column names for some encoders
* Documentation updates
* Stats models pinned to avoid errors
* Performance enhancements

1.2.6

======

* Release for zenodo DOI
* Inverse transform implemented for some encoders

Page 3 of 6

© 2024 Safety CLI Cybersecurity Inc. All Rights Reserved.