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
{{ message }}
This repository was archived by the owner on Feb 17, 2026. It is now read-only.
Right now this seems to happen in BaseClassifier.cs
public PmlEvaluation EvaluateWithCrossValidation(int numfolds = 10, bool quiet = false, bool getPredictions = true)
{
Build(quiet);
return new ClassifierEvaluator(Runtime, (IBaseClassifier)this).
EvaluateWithCrossValidateion(numfolds, quiet, getPredictions);
}
However, the WEKA maintainers say that building the model before crossvalidation can lead to incorrect results:
http://weka.8497.n7.nabble.com/Model-serialization-and-prediction-td6574.html