Classifier app K-fold cross validation: Is it possible to manually assign specific data sets within each fold?

1 次查看（过去 30 天）
Luckshman Bavan
Luckshman Bavan 2018-5-17
I am carrying out an activity classification study with 10 patients. Each patient has performed 10 repetitions of 5 exercises and each repetition has been segmented.
The matlab classifier app allows you to train and test algorithms using a K-fold cross validation technique which is fairly easy to implement. The division of the data set into k folds is performed automatically and I appreciate this ensures equal distribution of each labelled activity per fold.
I would like to assign data for each patient into each specific fold. This is to ensure each testing phase is performed on a new patient and that the algorithm being tested in that round hasn't been trained on any data from that specific patient.
Can anyone tell me if and how this can be done?
Thank you!

请先登录，再回答此问题。

回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Discriminant Analysis 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by