You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Each algorithm that supports either Regression or Classification must be registered 2 times with different algorithm ids. This will give us the ability to modify it's parameters and in the 1st case allow only parameters that are valid for Regression, and on the 2nd case parameters valid for Classification.
For example, LibSVM algorithm's type parameter can take either of those values: NU_SVR, EPSILON_SVR NU_SVC, C_SVC, ONE_CLASS_SVM
The first 2 will try to do Regression, while the other 3 will do Classification.
So, weka-svm-regression will have allowedValues the first 2 on type parameter, while weka-svm-classification will have the other 3.
For now I will remove weka-pls from classification.
Eternally Queued
The text was updated successfully, but these errors were encountered: