Scikit学习支持向量机的多类分类 [英] Scikit learn multi-class classification for support vector machines

问题描述

我想知道LinearSVC默认是否支持多类分类，还是我们必须将其包装在OneVsRestClassifier中?

I want to know whether LinearSVC supports multi-class classification by default or do we have to wrap it in OneVsRestClassifier like:

 OneVsRestClassifier(LinearSVC())

推荐答案

根据文档的这一部分:

SVC，NuSVC和 LinearSVC 是能够对数据集执行多类分类的类.

SVC, NuSVC and LinearSVC are classes capable of performing multi-class classification on a dataset.

[...]

另一方面， LinearSVC 实现了一个对另一个"的多类策略，从而训练了n_class个模型.如果只有两节课，则只会训练一种模型:

On the other hand, LinearSVC implements "one-vs-the-rest" multi-class strategy, thus training n_class models. If there are only two classes, only one model is trained:

因此，默认情况下它支持多类分类.

So it supports multiclass classification by default.

这篇关于Scikit学习支持向量机的多类分类的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！