def plot_roc_curve_multiclass()

in distributed_training/util/inference_utils.py [0:0]

• 41 lines of code
• 9 McCabe index (conditional complexity)

def plot_roc_curve_multiclass(y_true_ohe, y_score, num_classes, color_table, skip_legend=5, is_single_fig=False):
    """
    Plot ROC curve to multi-class
    """     
    fpr = dict()
    tpr = dict()
    roc_auc = dict()
    for i in range(num_classes):
        fpr[i], tpr[i], _ = roc_curve(y_true_ohe[:, i], y_score[:, i])
        roc_auc[i] = auc(fpr[i], tpr[i])

    # Compute micro-average ROC curve and ROC area
    fpr["micro"], tpr["micro"], _ = roc_curve(y_true_ohe.ravel(), y_score.ravel())
    roc_auc["micro"] = auc(fpr["micro"], tpr["micro"])
    
    # First aggregate all false positive rates
    all_fpr = np.unique(np.concatenate([fpr[i] for i in range(num_classes)]))

    # Then interpolate all ROC curves at this points
    mean_tpr = np.zeros_like(all_fpr)
    for i in range(num_classes):
        mean_tpr += np.interp(all_fpr, fpr[i], tpr[i])

    # Finally average it and compute AUC
    mean_tpr /= num_classes

    fpr["macro"] = all_fpr
    tpr["macro"] = mean_tpr
    roc_auc["macro"] = auc(fpr["macro"], tpr["macro"])  
    
    colors = cycle(color_table)
    
    fig, ax = plt.subplots(figsize=(8,8))
    ax.plot(fpr["micro"], tpr["micro"],
            label='micro-average ROC curve (area = {0:0.5f})'.format(roc_auc["micro"]),
            color='deeppink', linewidth=3)

    ax.plot(fpr["macro"], tpr["macro"],
            label='macro-average ROC curve (area = {0:0.5f})'.format(roc_auc["macro"]),
            color='navy', linewidth=3)

    for i, color in zip(range(num_classes), colors):
        if i % skip_legend == 0:
            label='ROC curve of class {0} (area = {1:0.4f})'.format(i, roc_auc[i])
        else:
            label=None
        ax.plot(fpr[i], tpr[i], color=color, label=label, lw=2, alpha=0.3, linestyle=':')
        ax.grid(alpha=.4)

    ax.plot([0, 1], [0, 1], 'k--', lw=2)
    ax.set_xlim([0.0, 1.0])
    ax.set_ylim([0.0, 1.05])
    ax.set_xlabel('False Positive Rate')
    ax.set_ylabel('True Positive Rate')
    ax.set_title('Receiver operating characteristic to multi-class')
    ax.legend(loc="lower right", prop={'size':10})
    if is_single_fig:
        plt.show()