fitcecoc svm classifier error

4 次查看（过去 30 天）
Pradyumna Reddy
Pradyumna Reddy 2019-4-16
回答： Don Mathis 2019-5-13
I am using MATLAB R2018b. I have two matrices X_norm (2000*20 double) which has my normalized observations and matrix Y (2000 * 1 double) has the class labels. I want to get an svm classifier model using X_norm and Y.
I have executed the following line
Mdl = fitcecoc(X_norm,Y,'Learners','svm','BoxConstraint',0.1);
and I get an error showing "Error using classreg.learning.FitTemplate/fillIfNeeded (line 634), BoxConstraint is not a valid parameter name."
Could anyone help me out with this issue?
  1 个评论
Pradyumna Reddy
Pradyumna Reddy 2019-4-16
I have tried this:
params = hyperparameters('fitcecoc',X_norm,Y,'svm');
params(2).Name
It gives the output as 'BoxConstraint' but I am not able to edit the hyperparameter with the mentioned code line in the question.

请先登录，再进行评论。

请先登录，再回答此问题。

回答（1 个）

Don Mathis
Don Mathis 2019-5-13
You need to pass fitcecoc a templateSVM object to set an SVM hyperparameter,
Mdl = fitcecoc(X_norm, Y, 'Learners', templateSVM('BoxConstraint', 0.1))

类别

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

标签

产品


版本

R2018b

Community Treasure Hunt

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

Start Hunting!

Translated by