v1.0.5

Version 1.0.5

• Separated exported functions into separate .R files.
• Removed Helper.R as the functions contained are not used for function operations.
• Refactored get_fairness_metrics to only preform the bootstrap once (as opposed to repeating multiple times for each metric) improving speed.
• Emphasizing in DESCRIPTION and README.md file (on Github) that the package scope is specifically for evaluating fairness of statistical/ML models accross binary protected attributes.
• Edited documentation to use the term "binary protected attribute" over "sensitive attribute" (personal preference)
• Updating functions to ensure that group arguments are strictly binary.
• Handling NA inputs with helper functions in ModelPreformance.R
• Confidence interval lengths are now reflected dynamically in the column names.
• Removed eval_cond_stats_parity and conditional statistical parity as a metric in get_fairness_metrics
• get_fairness_metrics returns a single data frame.
• Added note to get_fairness_metrics and to vigbette discussing limitations of statistical inferece with bootstraped CIs.