machine-learning - Matlab 2010 中训练测试验证集中的数据拆分

Question

我的问题是:Matlab 2010 提供了神经网络过程中的测试、验证期选项。这是数据拆分还是我必须使用“crossvalind”进行数据拆分？

Answer 1

这是 documentation 的摘录:

When training multilayer networks, the general practice is to first divide the data into three subsets. The first subset is the training set, which is used for computing the gradient and updating the network weights and biases. The second subset is the validation set. The error on the validation set is monitored during the training process. [...] The test set error is not used during training, but it is used to compare different models. [...]

提供了四个函数用于将数据划分为训练集、验证集和测试集:dividerand、divideblock、divideint 和红利。 (实际上还有第五个 dividetrain 将所有实例分配给训练)

对于更复杂的方法(交叉验证、分层等)，请查看 cvpartition 或 crossvalind 函数。