How create training and testing data with k-fold validation using neural network ?

11 次查看（过去 30 天）
Oman Wisni
Oman Wisni 2018-11-27
编辑： Oman Wisni 2018-11-27
Hi, I have finished training and testing data with the neural network formula that I calculated manually. Where is my data x = input (275x25) and t = target (275x1). Now I want to partition my data using K-fold validation where k = 5.
If I make (train or test) it manually, I have to train the input.mat data for the training, which consists of five files with dimension 220x25 every file.mat and five input.mat data for test with dimension 55x25 . I do this by inputting or loading the file repeatedly.
How can I implement the k-fold in the neural network code that I created? Is that possible, do the training and testing partitions then each data partition results in the accuracy of each partition both training and test?
please help me, I confused how where I should put code for k-fold. May anyone help some clear steps to explain it? Thanks

请先登录，再回答此问题。

采纳的回答

Yuvaraj Venkataswamy
编辑：madhan ravi 2018-11-27
  1 个评论
Oman Wisni
Oman Wisni 2018-11-27
编辑：Oman Wisni 2018-11-27
There are tutorial how create cross valitadion. should I partition first and then training or what?
input = inputs;
target =targets;
k=5;
cvFolds = crossvalind('Kfold');
How I create in cv ? can give me example ?

请先登录，再进行评论。

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Sequence and Numeric Feature Data Workflows 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