def convert_df_to_readable_format()

in parse.py [0:0]

• 45 lines of code
• 11 McCabe index (conditional complexity)

def convert_df_to_readable_format(reduced, bold=None, latex=None):
    # Formatting table contents with mean (std)
    summary = pd.DataFrame()
    pm_sign = "$\\pm$" if latex else "+/-"

    for c in reduced.columns.get_level_values(0):
        if "mean" in reduced[c] and "std" in reduced[c]:
            if "acc" in c.lower():
                summary[c] = (
                    (100 * reduced[c]["mean"]).map("{:.1f}".format)
                    + pm_sign
                    + (100 * reduced[c]["std"]).map("{:.1f}".format)
                )
            else:
                summary[c] = (
                    reduced[c]["mean"].map("{:.1f}".format)
                    + pm_sign
                    + reduced[c]["std"].map("{:.1f}".format)
                )
        elif "min" in reduced[c]:
            summary[c + " range"] = (
                "["
                + reduced[c]["min"].map("{:.1f}".format)
                + ", "
                + reduced[c]["max"].map("{:.1f}".format)
                + "]"
            )
        else:
            if is_numeric_dtype(reduced[c]) and reduced[c].dtype == "float":
                summary[c] = reduced[c].map("{:.1f}".format)
            else:
                summary[c] = reduced[c]
    if bold:
        if latex:
            bold_l, bold_r = r"\textbf{", "}"
        else:
            bold_l, bold_r = "*", ""

        best_algos = (
            reduced.sort_values((bold["best_metric"], "mean"), ascending=bold["order"])
            .groupby(bold["best_metric_group"])
            .head(1)
            .index
        )
        summary.loc[best_algos, bold["best_metric"]] = summary.loc[
            best_algos, bold["best_metric"]
        ].map(lambda x: bold_l + x + bold_r)
    return summary