[Clus-general] [CLUS] help for using CLUS RF-PCT

[寂. <944...@qq...>]

To Whom It May Concern,
Firstly, thank all of you very much in advance. I have a multi-label classification task and I'm trying using Clus to deal with it. But, first the problem is that the Clus manual doesn’t contain all features, and I don't know whether the output model is a  RF-PCT model or a PCT model with no ensemble. I indeed choose the RF model in the setting file but the output model looks like a singe PCT tree. The second problem is that can I get some details about the metrics to measure the output model? The only metric I see in the output file is ACCURACY. But if I want other metrics such as SubsetAccuracy,  Hamming Loss, or something like that, what should I do in the setting file?Thanks in advance!Best regards,Zhimeng Luo
Here are my settings and output model:-------------------------------------------------------------------------------------------------------------SETTINGS:[General] RandomSeed = 0 [Data] File = IMDB-F1.arff TestSet = 0.333 PruneSet = None [Attributes] Target = 1-28 Weights = 1 [Model] MinimalWeight = 5 [Tree] FTest = [0.001,0.01] Heuristic = VarianceReduction [Ensemble] EnsembleMethod = RForest VotingType = Majority [Output] WritePredictions = {Train,Test} TrainErrors = Yes AllFoldErrors = y WriteCurves = Yes [Hierarchical] Type = TREE WType = ExpAvgParentWeight HSeparator = / WParam = 0.75-------------------------------------------------------------------------------------------------------------MODEL:Pruned Model ************ vg > 0.0 +--yes: [0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0] [548.0,715.0,759.0,637.0,774.0,662.0,763.0,674.0,776.0,444.0,776.0,772.0,673.0,776.0,659.0,776.0,776.0,701.0,506.0,776.0,774.0,769.0,696.0,775.0,558.0,772.0,757.0,776.0]: 776 +--no:  [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0] [76489.0,73927.0,71623.0,75068.0,77397.0,59131.0,77680.0,74596.0,79610.0,74957.0,79557.0,77055.0,72174.0,78986.0,76476.0,58173.0,79676.0,50653.0,72698.0,68521.0,77784.0,78120.0,74474.0,79437.0,76679.0,79528.0,78518.0,78292.0]: 79877
-------------------------------------------------------------------------------------------------------------