1.1.0

[1.1.0] - 2025-05-26

Added

• Included get_dataset_regresion for separation of labeled and unlabeled data. For the regression dataset, the unlabeled instances are marked with NaN; by @aliciaolivaresgil.
• Includes check_regressor in the sslearn.utils module to check if the estimator is a regressor or not; by @aliciaolivaresgil.
• The regressor CoReg is now included, described in the paper "Semi-supervised regression with co-training" by Zhi-Hua Zhou and Ming Li. Implemented in the sslearn.wrapper module; by @aliciaolivaresgil.
• The regression TriTrainingRegressor is now included, made by adapting the TriTraining method to regression tasks created by Alicia Olivares Gil. Implemented in the sslearn.wrapper module; by @aliciaolivaresgil.

Changed

• The functions read_keel and read_csv from the sslearn.datasets module now has a new parameter is_regression to indicate if the dataset is for regression or classification. This will change the way the dataset is processed and returned; by @aliciaolivaresgil.

Fixed

• Repaired restricted.feature_fusion and restricted.probability_fusion when the instances that cannot link are not consecutive.

Testing

• Added tests for the new regression methods and the changes in the dataset reading functions; by @aliciaolivaresgil.

1.0.5.3

[1.0.5.3] - 2024-11-29

HotFix

• Remove debug logs in DeTriTraining.

v1.0.5.2

[1.0.5.2] - 2024-05-27

HotFix

• Remove some files that are not necessary in the package.

v1.0.5.1

[1.0.5.1] - 2024-05-20

Fixed

• Fixed bugs in artificial_ssl_dataset, now support again pandas DataFrame and y_unlabeled returns the right values

v1.0.5

[1.0.5] - 2024-05-08

Added

• feature_fusion and probability_fusion methods for restricted in sslearn.restricted module.

Fixed

• CoForest random integer is now compatible with Windows.

v1.0.4.1

[1.0.4.1] - 2024-02-06

Fix a problem with pypi

Added

• Add a parameter to artificial_ssl_dataset to force a minimum of instances. Issue #11
• Add a parameter to artificial_ssl_dataset to return indexes. Issue #13

Changed

• The artificial_ssl_dataset changed the process to generate the dataset, based in indexes. Issue #13

Fixed

• DeTriTraining now is vectorized and is faster than before.

v1.0.4

[1.0.4] - 2024-01-31

Added

• Add a parameter to artificial_ssl_dataset to force a minimum of instances. Issue #11
• Add a parameter to artificial_ssl_dataset to return indexes. Issue #13

Changed

• The artificial_ssl_dataset changed the process to generate the dataset, based in indexes. Issue #13

Fixed

• DeTriTraining now is vectorized and is faster than before.

v1.0.3.1

[1.0.3.1] - 2023-03-29

Added

• Methods now support no unlabeled data. In this case, the method will return the same as the base estimator.

Changed

• In OneHotEncoder, the sparse parameter is now sparse_output to avoid a FutureWarning.

Fixed

• CoForest now is most similar to the original paper.
• TriTraining can use at least 3 n_jobs. Fixed the bug that allows using as many n_jobs as cpus in the machine.

v1.0.3

[1.0.3] - 2023-03-29

Added

• Methods now support no unlabeled data. In this case, the method will return the same as the base estimator.

Changed

• In OneHotEncoder, the sparse parameter is now sparse_output to avoid a FutureWarning.

Fixed

• CoForest now is most similar to the original paper.
• TriTraining can use at least 3 n_jobs. Fixed the bug that allows using as many n_jobs as cpus in the machine.

v1.0.2

Change Log

[1.0.2] - 2023-02-17

Fixed

• Fixed a bug in TriTraining when one of the base estimators has not a random_state parameter.
• Fixed OneVsRestSSL with the random_state parameter.
• Fixed WiWTriTraining when no instance_group parameter is not provided.
• Fixed a FutureWarning for sparse parameter in OneHotEncoder. Changed to sparse_output.