benmack / oneClass

One-class classification in the absence of test data.
GNU Affero General Public License v3.0
31 stars 12 forks source link

metric puF and trControl #2

Closed cafernandezlo closed 5 years ago

cafernandezlo commented 7 years ago

Following the vignette example, I found a issue adding the trainControl option of the Caret package.

tr.control <-trainControl(method="repeatedCV",number=10,repeats=5) ocsvm.fit <- trainOcc(x=tr.x, y=tr.y, method="ocsvm", tuneGrid=tuneGrid, index=tr.index,trControl = tr.control)

Warning message: Te metric puF was not in the result set. Accuracy will be used instead.

benmack commented 7 years ago

Please read the help of the trainOcc function carefully. Particularly, about the argument trControl and what you need to consider when you use it (see also the hints in the Details and the Examples of the function documentation).

In your code example, if you already created the resampling indices tr.index that you give to the argument index of trainOcc correctly (index <- createMultiFolds(y, k=10, times=5)) then you do not need to additionally do the same thing via trControl.

When you read the help but things remain unclear, please let me know an I will try to improve it.

benmack commented 5 years ago

I am closing the issue due to missing feedback. Feel free to reopen it...