[Clus-devel] Random forest of regression trees

[Nair b. N. Y. <nai...@ya...>]

Dear,
I have a multi-label regression problem and I want to use RF-PCT (random forest of clustering trees) for that. Actually, I only have tried RF-PCT for classification problems, using the following setting file and this command line: java -jar Clus.jar -forest  data.s . So what do I need to change or add to run random forest of regression trees instead ? surely, the voting type and what else ?
Thanks in advanceNoureddine Yassine NAIR BENREKIAPhD student 
-----------------------------------
[Data]File =  data_train.arffTestSet = data_test.arff
[Attributes]
Target = 1-20
[Tree]Heuristic = VarianceReduction
[Output]AllFoldModels = NoAllFoldErrors = YesWritePredictions = Test

[Ensemble]Iterations = 100EnsembleMethod = RForestVotingType = ProbabilityDistributionSelectRandomSubspaces = 52Optimize = Yes