This patch fixes the following issues:

• Applying fixes to both open issues #51 and #52
• Adding support for python 3.10. This includes:
  • Upgrading the sdv package dependency to version 0.18.0.
  • Making appropriate bug fixes to accommodate this newer sdv version.

• New cohort module, which contains the following new classes:
  • CohortManager: class that manages multiple cohort pipelines that are applied over each cohort separately using the fit(), transform(), fit_resample(), predict(), and predict_proba() interfaces;
  • DecoupledClass: class that trains different models over different subsets of data (cohorts). Based on the work presented in the following paper: Decoupled classifiers for group-fair and efficient machine learning.
• New imputation approaches. The following classes were added into the dataprocessing.imputer module:
  • KNNDataImputer: class that imputes missing data of a feature using K-nearest neighbors;
  • IterativeDataImputer: class that imputes missing data of a feature using the other features. It uses a round-robin method
    of modeling each feature with missing values to be imputed as a function of the other features.
• New notebooks:
  • new tutorial notebooks showcasing the new classes
  • new case studies that showcases the benefits of using the CohortManager and DecoupledClass classes in real datasets.
• Bug fixes and new tests

First release. This release includes two sub-modules:

• databalanceanalysis: provides metrics for diagnosing errors that originate from data imbalance either on class labels or feature values
• dataprocessing: offers mitigation techniques for improving model performance for specific cohorts.

Educational notebooks located in the notebooks folder.