should I separate the data as training and testing when I use cross validation in ensemble methods? and is there any fully useful code? Thanks much.

Question

yusra asad 2021-8-21

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1438124-should-i-separate-the-data-as-training-and-testing-when-i-use-cross-validation-in-ensemble-methods

回答： Prantik Chatterjee 2024-3-21

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Prantik Chatterjee 2024-3-21

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1438124-should-i-separate-the-data-as-training-and-testing-when-i-use-cross-validation-in-ensemble-methods#answer_1428861

I am assuming that you have access to the latest MATLAB release.

You do not need to separate the data as training and test if you use the matlab library function 'crossval' for cross-validation. 'crossval' will, by default, split the given dataset into 10 folds internally. Alternatively, you can specify the number of folds using the argument 'Kfold' with 'crossval'. Please refer to the following documentation for more details on 'crossval'

https://in.mathworks.com/help/stats/classificationsvm.crossval.html

For a working example on using cross validation in ensemble methods, please refer to the following page

https://in.mathworks.com/help/stats/classreg.learning.partition.classificationpartitionedensemble.html