spotify/confidence

[Question] Docstring for Experiment class

Closed this issue · 1 comments

Isn't there a docstring for the Experiment class?

conf

Can you explain the parameters of the class to me? It would be very helpful.

Hopefully, below explanation will suffice for a docstring. I'll create a PR.

Docstring for Experiment class


"""
    This class represents an A/B experiment. It provides several methods to 
    analyze and visualize the results of the experiment. The experiment is 
    based on data provided as a DataFrame, with information about the numerator 
    and denominator of the tested metrics and their grouping columns.

    Attributes:
        data_frame (DataFrame): DataFrame that contains the experimental data.
        numerator_column (str): Name of the DataFrame column that contains the numerator of the tested metric.
        numerator_sum_squares_column (Union[str, None]): Name of the DataFrame column that contains the sum of squares of the numerator.
        denominator_column (str): Name of the DataFrame column that contains the denominator of the tested metric.
        categorical_group_columns (Union[str, Iterable]): Column(s) that categorically group the data.
        ordinal_group_column (Union[str, None]): Column that ordinally group the data.
        interval_size (float): Size of the confidence interval. Defaults to 0.95.
        correction_method (str): Method for correction. Defaults to BONFERRONI.
        confidence_computer (ConfidenceComputerABC): ConfidenceComputerABC object to compute confidence intervals. 
        confidence_grapher (ConfidenceGrapherABC): ConfidenceGrapherABC object to plot confidence intervals.
        method_column (str): Column that contains the experimental method.
        bootstrap_samples_column (str): Column that contains the bootstrap samples.
        metric_column (str): Column that contains the metric of interest.
        treatment_column (str): Column that contains the treatment groups.
        power (float): Desired statistical power. Defaults to 0.8.
        feature_column (str): Column that contains the features.
        feature_sum_squares_column (str): Column that contains the sum of squares of the features.
        feature_cross_sum_column (str): Column that contains the cross product sum of the features.
    """