58ec59639a00d0bf885968dd885aaba610ece6d5,autosklearn/metrics/classification_metrics.py,,bac_metric,#Any#Any#Any#,62

Before Change

    label_num = solution.shape[1]
    score = np.zeros(label_num)
    bin_prediction = binarize_predictions(prediction, task)
    [tn, fp, tp, fn] = acc_stat(solution, bin_prediction)
    // Bounding to avoid division by 0
    eps = 1e-15
    tp = sp.maximum(eps, tp)
    pos_num = sp.maximum(eps, tp + fn)

After Change

    :return:
    
    label_num = solution.shape[1]
    bin_prediction = binarize_predictions(prediction, task)
    fn = np.sum(np.multiply(solution, (1 - bin_prediction)), axis=0,
                dtype=float)
    tp = np.sum(np.multiply(solution, bin_prediction), axis=0, dtype=float)
    // Bounding to avoid division by 0
    eps = 1e-15
    tp = sp.maximum(eps, tp)
    pos_num = sp.maximum(eps, tp + fn)
    tpr = tp / pos_num  // true positive rate (sensitivity)

    if (task != MULTICLASS_CLASSIFICATION) or (label_num == 1):
        tn = np.sum(np.multiply((1 - solution), (1 - bin_prediction)),
                    axis=0, dtype=float)
        fp = np.sum(np.multiply((1 - solution), bin_prediction), axis=0,
                    dtype=float)
        tn = sp.maximum(eps, tn)
        neg_num = sp.maximum(eps, tn + fp)
        tnr = tn / neg_num  // true negative rate (specificity)
        bac = 0.5 * (tpr + tnr)

Instances