def compute_prob_stats()

in src/utils.py [0:0]

• 14 lines of code
• 3 McCabe index (conditional complexity)

def compute_prob_stats(df_col):
    stats = {
        'nan': df_col.isna().sum()
    }

    if df_col.dtype == 'int64' or df_col.dtype == 'float64':
        stats['type'] = ColType.NUMERICAL
        stats['mean'] = df_col.mean()
        stats['std'] = df_col.std()
        stats['dtype'] = df_col.dtype
    else:
        stats['type'] = ColType.CATEGORICAL

        prob = df_col.value_counts(normalize=True, dropna=False).to_dict()
        stats['prob'] = prob

    return stats