feature selection results vary every time.

1 次查看（过去 30 天）

Jacky Zhu 2018-11-21

Hi,

I used below code to do the feature selection. But every time I ran this code, the results were not the same.

I notice that the cvpartition function does do some random partition. so that why I got different results every time.

Are there any ways to improve this? So I could have more stable results.

Thanks,

Jacky

c = cvpartition(y,'k',5);

opts = statset('display','iter');

classf = @(train_data, train_labels, test_data, test_labels)...

sum(predict(fitcsvm(train_data, train_labels,'KernelFunction','rbf'), test_data) ~= test_labels);

[fs, history] = sequentialfs(classf, X_train, y_train, 'cv', c, 'options', opts);

在 Help Center 和 File Exchange 中查找有关 Classification Trees 的更多信息

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

Translated by