如何绘制多类分类中每个类的平均折叠的 ROC 曲线

Question

我在多类分类上使用 5 个交叉验证来评估随机森林的性能。我得到的曲线就像图片 我使用的代码如下

cv=StratifiedKFold(n_splits=5)

classifier = RandomForestClassifier(n_estimators=50,random_state=0)
y_tests = []
y_probabilities = []


for train_index, test_index in cv.split(X, y):
    X_train, X_test = X.iloc[train_index,:], X.iloc[test_index,:]
    y_train, y_test = y.iloc[train_index], y.iloc[test_index]
    probas_ = classifier.fit(X_train, y_train.values.ravel()).predict_proba(X_test)
    y_probabilities.append(probas_);
    y_tests.append(y_test)
i = 5
for i in range(5):
   skplt.metrics.plot_roc(y_tests[i], y_probabilities[i], plot_micro=False, plot_macro=False,title = 'Fold {} ROC Curve sebelum balancing'.format(i+1))

如何能够显示包含每个交叉验证类的折叠平均值的曲线