known as arandom forest, often outperforms using a single tree. During the bootstrap process, random resamples of variables and records are often taken. The prediction error on each of the bootstrap samples is known as the OOB score. It is used to fine-tune ...
Bagging方法中Bootstrap每次约有1313的样本不会出现在Bootstrap所采集的样本集合中,当然也就没有参加决策树的建立,把这1313的数据称为袋外数据OOB(out of bag),它可以用于取代测试集误差估计方法。 袋外数据(OOB)误差的计算方法如下: 对于已经生成的随机森林,用袋外数据测试其性能,假设袋外数据总数为O,用这O个...