我试图在多标签分类中使用scikit计算宏观F1值

from sklearn.metrics import f1_scorey_true = [[1,2,3]]y_pred = [[1,2,3]]print f1_score(y_true, y_pred, average='macro')

然而，它失败并显示错误信息

ValueError: multiclass-multioutput is not supported

我如何在多标签分类中计算宏观F1值？

回答：

在当前的scikit-learn版本中，您的代码会产生以下警告：

DeprecationWarning: Direct support for sequence of sequences multilabel    representation will be unavailable from version 0.17. Use    sklearn.preprocessing.MultiLabelBinarizer to convert to a label    indicator representation.

根据这个建议，您可以使用sklearn.preprocessing.MultiLabelBinarizer将这个多标签类转换为f1_score可以接受的形式。例如：

from sklearn.preprocessing import MultiLabelBinarizerfrom sklearn.metrics import f1_scorey_true = [[1,2,3]]y_pred = [[1,2,3]]m = MultiLabelBinarizer().fit(y_true)f1_score(m.transform(y_true),         m.transform(y_pred),         average='macro')# 1.0