experiments.experiment_utils.prep_data_for_fairlearn

prep_data_for_fairlearn(spec, results_dir, trial_i, data_frac, datagen_method, fairlearn_sensitive_feature_names, verbose)

Utility function for preparing features and labels for a given fairlearn trial.

Parameters:
  • spec – A seldonian.spec.Spec object.

  • results_dir (str) – The directory in which results are saved for this trial

  • trial_i (int) – Trial index

  • data_frac (float) – data fraction

  • datagen_method – Method for generating the trial datasets.

  • fairlearn_sensitive_feature_names – List of names of the sensitive attributes that fairlearn will use.