Parameters (* = default)
───────────────────────────
Data directory = "Data/Multilabel"
Evaluate standard k-NN = false (*)
FASTA file = false (*)
Huge dataset = false (*)
Obtain AUC = true (*)
Only save selection = false (*)
Output 2 class measures = true (*)
Verbose standard error printing = false (*)
Save outputs = false (*)
Seed = 0 (*)
Bias = false (*)
Show progress = 2
Force sparse format = false (*)
Sparse format = true
Typify = true (*)
Use all data = false (*)
Verbose = -1 (*)
Remove features after selection = false (*)
Remove constant features = false (*)
Evaluate margin = true (*)
Time all = false (*)
Minimum information = false (*)
Precision for zero = 0.000000 (*)
Evaluate with SVM = false (*)
Evaluate with C4.5 = false (*)
Evaluate with k-NN = false (*)
Evaluate with MLP = false (*)
Evaluate with ELM = false (*)
Evaluate with NBC = false (*)
Evaluate with DT(Accuracy) = false (*)
Evaluate with DT(ID3) = false (*)
Evaluate with DT(CART) = false (*)
Evaluate with DT(C4.5) = false (*)
Evaluate with IBT = false (*)
Evaluate with RF = false (*)
Evaluate with 1-NN = false (*)
Evaluate with GaussianNBC = false (*)
Evaluate with LR = false (*)
Evaluate with RForest = false (*)
Evaluate with ExtraTree = false (*)
Evaluate with SVM-k = false (*)
Evaluate with OVO = false (*)
Evaluate with AdaBoost.M1 = false (*)
Evaluate with MadaBoost = false (*)
Evaluate with MultiBoost = false (*)
Evaluate with Bagging = false (*)
Evaluate with GentleBoost = false (*)
Evaluate with ReweightBoost = false (*)
Evaluate with FloatBoost = false (*)
Evaluate with FeatureBoost = false (*)
Evaluate with Arcing = false (*)
Evaluate with LogitBoost = false (*)
Evaluate with LearningFromBites = false (*)
Evaluate with RSM = false (*)
Evaluate with FilterBoost = false (*)
Evaluate with AdaC1 = false (*)
Evaluate with AdaC2 = false (*)
Evaluate with AdaC3 = false (*)
Evaluate with CSB0 = false (*)
Evaluate with CSB1 = false (*)
Evaluate with CSB2 = false (*)
Evaluate with AdaBoost = false (*)
Evaluate with OVA = false (*)
Evaluate with DiscreteAdaBoost = false (*)
Evaluate with RealAdaBoost = false (*)
Evaluate with AdaBoost.NC = false (*)
Evaluate with ModestAdaBoost = false (*)
Evaluate with MPBoost = false (*)
Evaluate with AdaBoost.OC = false (*)
Evaluate with AdaBoost.EEC = false (*)
Evaluate with AdaBoost.M2 = false (*)
Evaluate with ECOC = false (*)
Evaluate with ECOC(sparse) = false (*)
Evaluate with AdaBoost.MH = false (*)
Evaluate with EmphasisBoost = false (*)
Evaluate with AdaBoost.M1W = false (*)
Evaluate with BoostMA = false (*)
Evaluate with CascadeVoting = false (*)
Evaluate with AdaBoostEPIC = false (*)
Evaluate with Aveboost = false (*)
Evaluate with A-Boosting = false (*)
Evaluate with Voting-boost = false (*)
Evaluate with HBoost = false (*)
Evaluate with BBOB = false (*)
Evaluate with BCEI = false (*)
Evaluate with MarginBoost = false (*)
Evaluate with WeightBoost = false (*)
Evaluate with AdaBoost.SECC = false (*)
Evaluate with SmoothBoost = false (*)
Evaluate with EPIBoost = false (*)
Evaluate with FlexBoost = false (*)
Evaluate with LIUBoost = false (*)
Evaluate with DBUBoost = false (*)
Evaluate with Wagging = false (*)
Evaluate with JOUSBoostOS = false (*)
Evaluate with JOUSBoostUS = false (*)
Evaluate with ClusterBased = false (*)
Evaluate with GrPloss = false (*)
Evaluate with NNBoost = false (*)
Evaluate with AdaBoost-P = false (*)
Evaluate with DECORATE = false (*)
Evaluate with DELAK = false (*)
Evaluate with VadaBoost = false (*)
Evaluate with M-AdaBoost = false (*)
Evaluate with A-Bagging = false (*)
Evaluate with SSAdaBoost = false (*)
Evaluate with LocalBoost = false (*)
Evaluate with AdaBoost.ERP = false (*)
Evaluate with iBRF = false (*)
Evaluate with TLUSBoost = false (*)
Evaluate with SIBag = false (*)
Evaluate with MIBag = false (*)
Evaluate with SIBoost = false (*)
Evaluate with MIBoost = false (*)
Evaluate with DPF-EL = false (*)
Evaluate with sBoost-1 = false (*)
Evaluate with sBoost-2 = false (*)
Evaluate with MSmoothBoost = false (*)
Evaluate with MPSUBoost = false (*)
Evaluate with CUSBoost = false (*)
Evaluate with AsymmBC = false (*)
Evaluate with CSBBoostRF = false (*)
Evaluate with WHMBoost = false (*)
Evaluate with RBBoost = false (*)
Evaluate with ExtremeBoost = false (*)
Evaluate with ExtremeBoostC = false (*)
Evaluate with RandomForest = false (*)
Evaluate with XGBoost = false (*)
Evaluate with NaiveBayes = false (*)
Evaluate with DeepLearning = false (*)
Evaluate with DecisionTree = false (*)
Evaluate with PLS-DA = false (*)
Evaluate with SVC = false (*)
Evaluate with LDA = false (*)
Evaluate with QDA = false (*)
Evaluate with GPML = false (*)
Evaluate with RidgeRegression = false (*)
Evaluate with ExtremelyRandomizedTrees = false (*)
Evaluate with Kernel RidgeRegression = false (*)
Evaluate with AdaBoost.R = false (*)
Evaluate with AdaBoost.D = false (*)
Evaluate with TwinSVM = false (*)
Evaluate with HHCart = false (*)
Evaluate with CO2 = false (*)
Evaluate with BUTIF = false (*)
Evaluate with RandCART = false (*)
Evaluate with RidgeCART = false (*)
Evaluate with NLDT = false (*)
Evaluate with LinearTree = false (*)
Evaluate with OC1 = false (*)
Evaluate with LogitBoost(py) = false (*)
Evaluate with CatBoost = false (*)
Evaluate with RRULES = false (*)
Evaluate with RULES = false (*)
Evaluate with RLSC = false (*)
Evaluate with ExtremeLearningMachine = false (*)
Evaluate with RotationForest = false (*)
Evaluate with OneVsOne = false (*)
Evaluate with OneVsAll = false (*)
Evaluate with BrownBoost(ovo) = false (*)
Evaluate with BrownBoost(ova) = false (*)
Evaluate with EasyEnsemble(ovo) = false (*)
Evaluate with EasyEnsemble(ova) = false (*)
Evaluate with RUSBoost = false (*)
Evaluate with AdaCost = false (*)
Evaluate with AdaUBoost = false (*)
Evaluate with AsymBoost = false (*)
Evaluate with SMOTEBoost = false (*)
Evaluate with BalanceCascade = false (*)
Evaluate with EasyEnsemble = false (*)
Evaluate with LightGBM = false (*)
Evaluate with Lasso = false (*)
Evaluate with GradientBoostingTrees = false (*)
Evaluate with HistGradientBoostingTrees = false (*)
Evaluate with GPC = false (*)
Evaluate with MEBoostOVO = false (*)
Evaluate with MEBoostOVA = false (*)
Evaluate with Stacking = false (*)
Evaluate with HybridGB = false (*)
Evaluate with BoostTree = false (*)
Evaluate with BoostForest = false (*)
Evaluate with RobustLogit = false (*)
Evaluate with ABCRobustLogit = false (*)
Evaluate with Mart = false (*)
Evaluate with ABCMart = false (*)
Evaluate with SnapBoost = false (*)
Evaluate with SnapBoostOVO = false (*)
Evaluate with SnapBoostOVA = false (*)
Evaluate with WildWood = false (*)
Evaluate with BalancedRandomForest = false (*)
Evaluate with RolexBoost = false (*)
Evaluate with FlexBoost(py) = false (*)
Evaluate with RotationForest2 = false (*)
Evaluate with KTBoost = false (*)
Evaluate with NewtonBoost = false (*)
Evaluate with TSSE-BIM = false (*)
Evaluate with TSSE-BIM(ovo) = false (*)
Evaluate with TSSE-BIM(ova) = false (*)
Evaluate with EASE = false (*)
Evaluate with EASE(ovo) = false (*)
Evaluate with EASE(ova) = false (*)
Evaluate with SPE = false (*)
Evaluate with SPE(ovo) = false (*)
Evaluate with SPE(ova) = false (*)
Evaluate with HUE = false (*)
Evaluate with HUE(ovo) = false (*)
Evaluate with HUE(ova) = false (*)
Evaluate with EcuBoost = false (*)
Evaluate with EcuBoost(ovo) = false (*)
Evaluate with EcuBoost(ova) = false (*)
Evaluate with NGBoost = false (*)
Evaluate with GPBoost = false (*)
Evaluate with GPBoost(ovo) = false (*)
Evaluate with GPBoost(ova) = false (*)
Evaluate with LaGaBoost = false (*)
Evaluate with LaGaBoost(ovo) = false (*)
Evaluate with LaGaBoost(ova) = false (*)
Evaluate with GBDT-PL = false (*)
Evaluate with MILBoost(ovo) = false (*)
Evaluate with MILBoost(ova) = false (*)
Evaluate with LPBoost(ova) = false (*)
Evaluate with LPBoost(ovo) = false (*)
Evaluate with GraphSepBoost(ova) = false (*)
Evaluate with GraphSepBoost(ovo) = false (*)
Evaluate with MLPBoost(ova) = false (*)
Evaluate with MLPBoost(ovo) = false (*)
Evaluate with CERLPBoost(ova) = false (*)
Evaluate with CERLPBoost(ovo) = false (*)
Evaluate with ERLPBoost(ova) = false (*)
Evaluate with ERLPBoost(ovo) = false (*)
Evaluate with SoftBoost(ova) = false (*)
Evaluate with SoftBoost(ovo) = false (*)
Evaluate with TotalBoost(ova) = false (*)
Evaluate with TotalBoost(ovo) = false (*)
Evaluate with AdaBoostV(ova) = false (*)
Evaluate with AdaBoostV(ovo) = false (*)
Evaluate with SmoothBoost(ova) = false (*)
Evaluate with SmoothBoost(ovo) = false (*)
Evaluate with MadaBoost(ova) = false (*)
Evaluate with MadaBoost(ovo) = false (*)
Evaluate with GBM(ova) = false (*)
Evaluate with GBM(ovo) = false (*)
Evaluate with DGBF = false (*)
Evaluate with DGBF(ovo) = false (*)
Evaluate with DGBF(ova) = false (*)
Evaluate with CSBBoostXG(ovo) = false (*)
Evaluate with CSBBoostXG(ova) = false (*)
Evaluate with SGDC = false (*)
Evaluate with LinearForest = false (*)
Evaluate with LinearBoosting = false (*)
Boosting external learner = DecisionTree (*)
Maximum number of instances for training = 0 (*)
Maximum number of instances for learning = 0 (*)
Evaluate training error = false (*)
Sample using k-Means = false (*)
Stratified sampling = true (*)
Sample using k-Means subset size = 0 (*)
Test 1-NN slaves = 0 (*)
General distance metric = Euclidean (*)
Feature selection = false (*)
Instance selection = false (*)
Randomize inputs for feature selection = false (*)
Multi-class CI use cost matrix = false (*)
Duplicate instances = false (*)
Duplicate instances with noise = false (*)
Duplicate features = false (*)
Duplicate features ratio = 1 (*)
Save selection = false (*)
Maximum training set size = 0 (*)
Noise filtering k = 1 (*)
Second chances = 1 (*)
Use averaged accuracy for stop = true (*)
Evaluate with leave one out = false (*)
Heuristic search = SFS (*)
Plus l = 3 (*)
Minus r = 1 (*)
MultiBoost reinitialization iterations = 10 (*)
Boosting-Based IS ensemble method = MultiBoost (*)
Boosting-Based IS select best instance from every class = true (*)
Boosting-Based maximum instances per iteration = 0 (*)
MABUSE FS learner = C4.5 (*)
MABUSE FS maximum instances per training = 0 (*)
MABUSE FS/SIFS2 maximum features per iteration = 0 (*)
MABUSE FS ensemble method = AdaBoost (*)
MABUSE FS Performance measure = Kappa (*)
MABUSE FS/SIFS2 maximum losing streak = 10 (*)
MABUSE FS second chances = 2 (*)
MABUSE FS feature life = 1 (*)
MABUSE FS filter weighted sampling ratio = 1.000000 (*)
MABUSE FS conditional exclusion with margin = false (*)
MABUSE FS folds = 1 (*)
MABUSE FS ranking initial selection percentage = 0.000000 (*)
MABUSE FS/SIFS2 sort by weighted MI limit = 0 (*)
MABUSE FS minimum selected percentage = 0.000000 (*)
MABUSE use rankings for performance = false (*)
MABUSE average ranking weight = false (*)
MABUSE roulette selection = false (*)
MABUSE FS/SIFS2 maximum losing streak = 10 (*)
MABUSE FS/SIFS2 maximum features per iteration = 0 (*)
MABUSE FS/SIFS2 sort by weighted MI limit = 0 (*)
SIFS2 convergence using only accuracy = false (*)
SIFS2 best using only accuracy = true (*)
SIFS2 fitness evaluator = k-NN (*)
SIFS2 stratified sampling size = 0 (*)
SIFS2 stratum size = 0 (*)
SIFS2 when stratified evaluate with whole training set = false (*)
SIFS2 stratified additive instance selection = false (*)
SIFS2 stratified incremental selection = true (*)
SIFS2 stratified sampling size for performance evaluation = 0 (*)
SIFS2 features added group size = 1 (*)
SIFS2 random instance selection algorithm = false (*)
Instance reduction weight = 0.000000 (*)
Feature reduction weight = 0.000000 (*)
Exponential reduction feature = false (*)
Exponential reduction instance = false (*)
Exponential reduction = false (*)
Fitness as ratio = false (*)
CV for fitness as ratio = true (*)
Instance/feature selection algorithm = CHCGA (*)
Random FS algorithm = false (*)
Random FS algorithm low redundancy methods = false (*)
Random FS algorithm fast methods = false (*)
Random IS algorithm = false (*)
Multi-selection of instances = false (*)
Multi-selection of instances as local k = false (*)
Multi-select only minority class = false (*)
Multi-selection adjust initial percentage = false (*)
Pre-sample = false (*)
Pre-sample ratio = 0.000000 (*)
Pre-sample subset size = 1000 (*)
Pre-sample stratified = true (*)
Class-imbalanced = false (*)
Oversampling ratio = 1.000000 (*)
BalanceCascade s_i = 5 (*)
EasyEnsemble s_i = 5 (*)
CSBBoost minimum k = 2 (*)
CSBBoost maximum k = 10 (*)
sBoost k = 5 (*)
sBoost q = 5 (*)
ExtremeBoost size = 5 (*)
ExtremeBoostC choose best alpha = false (*)
MSmoothBost Lambda = 0.500000 (*)
Weighted Hybrid Ensemble ratio = 0.900000 (*)
Weighted Hybrid Ensemble p1 = 0.450000 (*)
Weighted Hybrid Ensemble p2 = 0.450000 (*)
Weighted Hybrid Ensemble DT probability = 0.750000 (*)
Weighted Hybrid Ensemble SMOTE k = 3 (*)
CUSBoost k = 10 (*)
CUSBoost undersampling ratio = 0.500000 (*)
iBRF k = 3 (*)
iBRF random undersampling ratio = 0.200000 (*)
iBRF SMOTE ratio = 0.800000 (*)
M-AdaBoost p = 0.500000 (*)
AdaBoost.NC lambda = 1.000000 (*)
JOUSBoost delta = 51 (*)
VadaBoost lambda = 0.500000 (*)
EmphasisBoost lambda = 0.350000 (*)
SmoothBoost kappa = 0.010000 (*)
SmoothBoost gamma = 0.250000 (*)
SmoothBoost theta = 0.250000 (*)
TLUSBoost t = 3 (*)
Cascade voting threshold = 10 (*)
SIOS missing rate = 0.350000 (*)
Random selection with replacement = false (*)
Adapt ratio = true (*)
Random selection minimum ratio = 0.010000 (*)
Random selection maximum ratio = 0.250000 (*)
Random selection ratio = 0.100000 (*)
T = 1 (*)
Resampling = true (*)
Alpha = 0.500000 (*)
Dynamic threshold = true (*)
Maximum number of thresholds = 0 (*)
Classifier = k-NN (*)
Votes evaluation measure = Accuracy (*)
Ensemble method = NoBoost (*)
Ensemble size = 50 (*)
Votes recording = ICPOS (*)
Random IS algorithm = false (*)
Use consider list = false (*)
Add k neighbors = 0 (*)
Boosting resampling method = WithoutReplacement (*)
Update w iterations = 1 (*)
Threshold evaluation iterations = 1 (*)
Alpha set to 1 = false (*)
Subsampling ratio = 0.100000 (*)
Weighted sampling for training = false (*)
BoostFS Classifier = C4.5 (*)
Evaluator = C4.5 (*)
Ensemble implementation = Resampling (*)
Select feature selection algorithm randomly = false (*)
Evaluate kappa-error = false (*)
Classifier = k-NN (*)
Maximum instance selection thresholds = 0 (*)
Maximum feature selection thresholds = 0 (*)
Maximum IS+FS selection thresholds = 0 (*)
Votes evaluator learner = k-NN (*)
Threshold evaluation instance reduction weight = 1.000000 (*)
Threshold evaluation feature reduction weight = 1.000000 (*)
Exponential evaluation of feature reduction = false (*)
Exponential evaluation of instance reduction = false (*)
Evaluation folds = 10 (*)
LIUBoost k = 5 (*)
LIUBoost delta = 0.010000 (*)
SSAdaBoost mu = 0.300000 (*)
SSAdaBoost alpha = 0.500000 (*)
SSAdaBoost delta = 0.950000 (*)
DPF-EL C = 5 (*)
IS+FS number of bites = 1 (*)
Feature subset evaluation = Classifier (*)
Feature subset combinations = Thresholding (*)
Maximum adjacency matrix size (GB) = 16.000000 (*)
Use edge list for vote graph = false (*)
Classification performance as percentage = false (*)
Select with sorted graph = false (*)
Save graph = false (*)
Sampling ratio = 0.000000 (*)
Accuracy weight for threshold = 1.000000 (*)
G-mean weight for threshold = 0.000000 (*)
Kappa weight for threshold = 1.000000 (*)
auROC weight for threshold = 0.000000 (*)
auPRC weight for threshold = 0.000000 (*)
Reduction weight for threshold = 1.000000 (*)
Evaluation subset size = 0 (*)
BoostFS CI weight initialization = false (*)
Fixed number of thresholds = false (*)
Maximum number of thresholds = 0 (*)
Performance file = "None" (*)
Save performance = false (*)
Wrapper FS learner = k-NN (*)
eReweightBoost r = 5 (*)
DBUBoost k = 20 (*)
HBoost models = "0010111" (*)
FlexBoost K = 0.500000 (*)
EPIBoost epsilon = 0.800000 (*)
EPIBoost r = 0 (*)
EPIBoost early finish = false (*)
Learning From Bites bite size = 1000 (*)
Learning From Bites subset size = 10000 (*)
Learning From Bites iterations = 250 (*)
External partitioning approach = false (*)
External subset size = 1000 (*)
Internal partitioning approach = false (*)
Internal subset size = 1000 (*)
Internal partitioning method = Random (*)
External partitioning method = Random (*)
External stratified partition = false (*)
Internal stratified partition = false (*)
Use PCs = false (*)
Internal subspace size = 0 (*)
External subspace size = 0 (*)
DELAK phi = 0.500000 (*)
Instance selection algorithm = DROP3 (*)
Sampling size = 0 (*)
Discretizer = FAYYAD (*)
Cross-validate hyper-parameters = false (*)
Cross-validate hyper-parameters for testing = false (*)
Cross-validate external hyper-parameters for testing = false (*)
Cross-validate k for testing = false (*)
Cross-validation performance measure = Accuracy (*)
Sampling for cross-validation of hyper-parameters = Stratified (*)
Sampling size for cross-validation of hyper-parameters = 0 (*)
k-NN k = 1 (*)
Minimum k = 1 (*)
Maximum k = 25 (*)
Adaptive distance measure = false (*)
k-NN with weighted votes = false (*)
Symmetric k-NN = false (*)
k-Means maximum iterations = 100 (*)
Evolutionary local k = false (*)
Optimal local k = false (*)
Edited k-NN = false (*)
Show average NN margin = false (*)
LASSO alpha = 0.001000 (*)
BBOB maximum size = 100 (*)
Gaussian NBC use logs = true (*)
Naive Bayes learner discretization = FANDI (*)
Naive Bayes learner intervals = 100 (*)
DECORATE R = 1.000000 (*)
Random forest trees = 50 (*)
Random forest percentage of features = 0.500000 (*)
External classifiers keep files = false (*)
Use 'ranger' for random forest = false (*)
RULES/RRULES bins = 10 (*)
RULES/RRULES discretize with frequency = false (*)
Oblique decision trees maximum depth = 5 (*)
Bit mutation = 0.100000
CHC wrapper = k-NN (*)
EBUS GS GM = false (*)
LOO for k-NN performance = true (*)
Evolutionary scheme = CHCGA (*)
Generations = 10000
Individuals = 100 (*)
Initial feature percentage = 0.500000 (*)
Initial percentage = 0.250000 (*)
Maximum evaluations = 0 (*)
Classification weight = 0.500000 (*)
Performance measure = Accuracy (*)
Reduction weight = 0.000000 (*)
Reinitialize at distance 0 = true (*)
Time limit = 3600
CHC maximum re-initialization attempts = 0 (*)
DE NP = 100 (*)
DE F = 0.800000 (*)
DE CR = 0.900000 (*)
DE Multi-IS and k = false (*)
DE evolve threshold = false (*)
DE separate thresholds = false (*)
DE local search IS probability = 0.000000 (*)
DE local search k probability = 0.000000 (*)
DE undersampling mutation probability = 0.000000 (*)
Subset size = 0 (*)
Minimum subset size = 0 (*)
Electors = 10 (*)
Weight votes = false (*)
Votes evaluator = k-NN (*)
Get best round = false (*)
Angle step = 1.000000 (*)
Fusion subsets = false (*)
Fusion from first = false (*)
Percent for error estimation = 0.100000 (*)
Get threshold with all = true (*)
Feature sampling = false (*)
Instance partition = false (*)
Partition = Random (*)
Number of thresholds = 0 (*)
Keep class distribution = true (*)
Subspace size = 5 (*)
Subsampling for evaluation = false (*)
Democratic evaluation of threshold = true (*)
Democratic evaluation rounds = 1 (*)
Democratic evaluation of threshold with CV = true (*)
RNN IS mutation = 0.000000 (*)
DROP3 mutation = 0.000000 (*)
ICF mutation = 0.000000 (*)
IB3 mutation = 0.000000 (*)
MSS mutation = 0.000000 (*)
RNN standard mutation = 0.000000 (*)
Add IS mutation = 0.000000 (*)
Boost-based IS mutation = 0.000000 (*)
RFE mutation = 0.000000 (*)
RNN FS mutation = 0.000000 (*)
FOCUS mutation = 0.000000 (*)
Add FS mutation = 0.000000 (*)
RNN IS+FS mutation = 0.000000 (*)
IMOEA mutation = 0.000000 (*)
Add IS+FS mutation = 0.000000 (*)
IS+FS maximum cache size (GB) = 4.000000 (*)
RMHC-PF1 iterations = 1000 (*)
Asymmetric AdaBoost K = 0.000000 (*)
C+ = 0.000000 (*)
C- = 0.000000 (*)
Compare votes = false (*)
Compare class = false (*)
Feature partition = false (*)
Individual filter = true (*)
Random undersampling for threshold evaluation = false (*)
Random undersampling for threshold evaluation ratio = 0.100000 (*)
Random undersampling for threshold evaluation minimum size = 10000 (*)
Random undersampling for threshold evaluation maximum size = 10000 (*)
ELM Hidden nodes = 100 (*)
ELM Transfer function = HypTan (*)
ELM A = 1.000000 (*)
ELM B = 0.666600 (*)
Validation percentage = 0.100000 (*)
Use validation data for votes threshold = false (*)
Only accept mutation if better = true (*)
LogitBoost use regression tree = true (*)
LogitBoost weight trim quantile = 0.000000 (*)
FilterBoost epsilon = 0.010000 (*)
FilterBoost delta = 0.950000 (*)
FilterBoost tau = 0.050000 (*)
FilterBoost mt = 1000 (*)
FilterBoost maximum size = 100 (*)
FilterBoost Cn = 300 (*)
MPBoost feature ratio = 0.500000 (*)
MPBoost instance ratio = 0.100000 (*)
MPBoost mu = 0.500000 (*)
MPBoost minimum ensemble size = 30 (*)
MPBoost maximum ensemble size = 100 (*)
RSM probability = 0.500000 (*)
EPIC p = 0.400000 (*)
CatBoost iterations = 5 (*)
CatBoost learning rate = 0.100000 (*)
IS+FS use cache = true (*)
IS+FS stepwise cache = false (*)
IS use cache = true (*)
FS use cache = true (*)
IS+FS mutation random order = false (*)
IS+FS inputs first = false (*)
IS+FS use fitness = true (*)
Maximum features Focus = 10 (*)
IMOEA seconds = 10 (*)
GGA elitism = 0.100000 (*)
GGA crossover probability = 0.600000 (*)
GGA mutation = 0.100000 (*)
Early stopping = false (*)
Percent for early stopping = 0.250000 (*)
Filter noise = true (*)
LVQ Npc = 25 (*)
GCNN rho = 0.990000 (*)
CNN class-imbalanced dataset = false (*)
RMHC m = 3 (*)
RMHC iterations = 1000 (*)
HMNEI epsilon = 0.100000 (*)
Proximity graph = Gabriel (*)
Validate parameters = false (*)
Cross-validation percentage = 0.100000 (*)
SMOTE undersampling ratio = 1.000000 (*)
Smote k = 5 (*)
Smote N = 2 (*)
SMOTE variant = SMOTE (*)
WeightBoost beta = 0.500000 (*)
AdaBoost.SECC eta = 0.500000 (*)
EvoSMOTE evaluate new instances = false (*)
EvoSMOTE use only minority class = true (*)
EvoSMOTE use majority class threshold = 5 (*)
EvoSMOTE undersamplig mutation rate = 0.000000 (*)
EvoSMOTE OSS mutation rate = 0.000000 (*)
EvoSMOTE instance selection mutation rate = 0.000000 (*)
EvoSMOTE CNNTL mutation rate = 0.000000 (*)
EvoSMOTE minimum N = 1 (*)
EvoSMOTE maximum N = 10 (*)
EvoSMOTE  wrapper evaluation interval = 0 (*)
Instance selection symmetric evolution = false (*)
EvoSMOTE roulette initialization = false (*)
Best individual fitness improvement generations = 0 (*)
Best individual fitness improvement addition = 0.100000 (*)
EvoSMOTE noise mutation rate = 0.000000 (*)
EvoSMOTE noise mutation rate k = 5 (*)
EvoSMOTE use only one weight = false (*)
Accuracy fitness weight = 0.000000 (*)
G-mean fitness weight = 0.000000 (*)
F1 fitness weight = 0.000000 (*)
auROC fitness weight = 0.000000 (*)
Kappa fitness weight = 0.000000 (*)
Recall macro fitness weight = 0.000000 (*)
Recall macro g-mean fitness weight = 0.000000 (*)
Average accuracy fitness weight = 0.000000 (*)
Oversampled instances fitness weight = 0.000000 (*)
Correlation coefficient fitness weight = 0.000000 (*)
Sensitivity fitness weight = 0.000000 (*)
Specificity fitness weight = 0.000000 (*)
Precision fitness weight = 0.000000 (*)
Negative predictive value fitness weight = 0.000000 (*)
Balance weight = 0.000000 (*)
Class-imbalance method = None (*)
Random selection ratio = 0.100000 (*)
Sampling with replacement = true (*)
LVF Iterations = 100000 (*)
Floating algorithms attempts = 3 (*)
K-S CBF k = 10 (*)
K-S CBF alpha = 0.050000 (*)
G-flip maximum iterations = 1000 (*)
Simba T = 0 (*)
iSimba lambda = 0.100000 (*)
iSimba T = 0 (*)
iSimba delta = 0.100000 (*)
Elitism percentage = 0.000000 (*)
Initial set probability = 0.500000 (*)
Mutation = 0.100000 (*)
GA alpha = 0.750000 (*)
Random undersampling = false (*)
Random undersampling ratio = 1.000000 (*)
Positive class = -1 (*)
Feature ranking = false (*)
Threshold evaluation = Voor (*)
Feature ranking algorithm = MutualInformation (*)
Unbalanced correlation score lambda = 3.000000 (*)
Threshold evaluation reduction weight = 0.500000 (*)
Threshold evaluation accuracy weight = 0.500000 (*)
Coefficient as geometric mean = false (*)
Coefficient as harmonic mean = false (*)
Kernel = 0
SVM type = 0 (*)
Gamma = 1.000000 (*)
Polynomial degree = 3 (*)
C = 0.001000
R = 0.000000 (*)
Tolerance = 0.001000 (*)
SVM Cache size = 1000000 (*)
C1 = 1.000000 (*)
C2 = 1.000000 (*)
Soften thresholds = false (*)
Prune tree = false (*)
Trials = 1 (*)
RForest split measure = GainRatio (*)
LocalBoost beat = 5.000000 (*)
LocalBoost k = 20 (*)
Nodes = 100 (*)
Initial weight = 0.100000 (*)
Transfer output = Log (*)
Transfer hidden = HypTan (*)
Number of epochs = 10000 (*)
Epoch size = 10 (*)
Eta = 0.500000 (*)
Alpha = 0.100000 (*)
Lambda = 0.000100 (*)
Flat spot = 0.100000 (*)
A output = 1.000000 (*)
B output = 0.666600 (*)
A = 1.000000 (*)
B = 0.666600 (*)
Asymptote = 0.200000 (*)
Maximum nodes = 100 (*)
DT minimum samples to split = 10 (*)
DT maximum depth = 20 (*)
DT grow full tree = false (*)
IBT split measure = GainRatio (*)
DT print tree = false (*)
Multi-classifier base learner = DT(C4.5) (*)
Multi-classifier output coding base learner = DT(C4.5) (*)
IBT ensemble method = NoBoost (*)
IBT alpha method = NoBoost (*)
IBT do not stop until complete = true (*)
IBT do not use internal nodes = false (*)
IBT eMultiBoost I = 10 (*)
IBT fixed node weight = false (*)
IBT predict with probabilities = false (*)
Feature selection algorithm = LVF (*)
Random undersampling for training = false (*)
Use G measure = false (*)
CV folds = 3 (*)
ReliefF m = 0 (*)
ReliefF k = 10 (*)
ReliefF threshold = 0.050000 (*)
ReliefF alpha = 0.000000 (*)
Evolutionary local k+IS = false (*)
Local k mutation rate = 0.000000 (*)
Local k mutation with no selection = 0.000000 (*)
Local k repetitions = 1 (*)
Local k lambda = 3 (*)
Local k alpha accuracy = 1.000000 (*)
Local k alpha regularization = 0.000000 (*)
Local k alpha margin = 0.000000 (*)
Local k error estimation method = k-FOLDCV (*)
Folds for training error estimation = 10 (*)
Local k memetic mutation ratio = 0.100000 (*)
Local k memetic bit mutation ratio = 0.250000 (*)
Minimum local k = 1 (*)
Maximum local k = 25 (*)
Number of bootstrapping samples = 10 (*)
Percentage for hold out method = 0.100000 (*)
Reevaluate individuals each generation = false (*)
Margin method = Normalized Votes Difference (*)
Artificial data number of neighbors = 1 (*)
Artificial data copies per instance = 1 (*)
Populate for optimal k = false (*)
Local k assign nearest value to no neighbors instances = false (*)
k values for local k = "3" (*)
Leave no neighbors unchanged = false (*)
FCBF Delta = -1.000000 (*)
LR number of epochs = 1000 (*)
LR learning coefficient = 0.010000 (*)
Regression tree delta = 0.010000 (*)
mRMR n = 50 (*)
mRMR evaluator = k-NN (*)
mRMR search heuristic = SFS (*)
wMI epsilon = 0.000000 (*)
Random FS selection probability = 0.100000 (*)
FAST Theta = -1.000000 (*)
MICIBM k = -1 (*)
BCEI alpha = 0.300000 (*)
BCEI beta = 0.600000 (*)
UFS epsilon = 0.100000 (*)
UFS R2max = 0.500000 (*)
Pre-undersampling = false (*)
Undersampling method = Random (*)
Random sampling = false (*)
Random sampling ratio = 0.100000 (*)
LS factor old individuals = 0.000000 (*)
LS only use selected = true (*)
Multi-Label method = ML-kNN (*)
Data partition method = NoDataPartition (*)
ML print base learner training error = false (*)
Minimum cluster size = 1 (*)
Multi-label undersample binary learners = false (*)
Cross-validate multi-label hyper-parameters = false (*)
BR classification model = DT(C4.5) (*)
BR RkNN oversampling = false (*)
BR RkNN k = 5 (*)
ML-TREE classification model = DT(C4.5) (*)
ML-FOREST number of trees = 50 (*)
ML-FOREST use MLCut = false (*)
DBR classification model = DT(C4.5) (*)
STA classification model = DT(C4.5) (*)
NS classification model = DT(C4.5) (*)
CC classification model = DT(C4.5) (*)
CC random order of labels = false (*)
RAkEL classification model = DT(C4.5) (*)
RAkEL k = 3 (*)
RAkEL m = 0 (*)
Fixed seed for RAkEL powersets = false (*)
LP classification model = DT(C4.5) (*)
LP undersample majority class = false (*)
CLR classification model = DT(C4.5) (*)
HOMER k = 3 (*)
HOMER classification model = DT(C4.5) (*)
HOMER balanced k-means iterations = 100 (*)
HOMER maximum depth = 10 (*)
PEVUS remove null label instances = false (*)
PEVUS random initialization = false (*)
Evaluate ML with ML-kNN = false (*)
Evaluate ML with BR = false (*)
Evaluate ML with CC = false (*)
Evaluate ML with LAML-kNN = false (*)
Evaluate ML with ML-k'sNN = false (*)
Evaluate ML with LAML-k'sNN = false (*)
Evaluate ML with RAkEL-D = false (*)
Evaluate ML with RAkEL-O = false (*)
Evaluate ML with LP = false (*)
Evaluate ML with CLR = false (*)
Evaluate ML with HOMER-B = false (*)
Evaluate ML with HOMER-K = false (*)
Evaluate ML with HOMER-R = false (*)
Evaluate ML with ML-DT = false (*)
Evaluate ML with ML-BASICBP = false (*)
Evaluate ML with BP-MLL = false (*)
Evaluate ML with ML-RF = false (*)
Evaluate ML with DBR = false (*)
Evaluate ML with STA = false (*)
Evaluate ML with NS = false (*)
Evaluate ML with ECC = false (*)
Evaluate ML with IBLR-ML = false (*)
Evaluate ML with IBLR-ML+ = false (*)
Evaluate ML with PCT = false (*)
Evaluate ML with E-PCT = false (*)
Evaluate ML with RF-PCT = false (*)
Evaluate ML with TSVM = false (*)
Evaluate ML with TSCCM = false (*)
Evaluate ML with TSPCCM = false (*)
Evaluate ML with ML-TREE = false (*)
Evaluate ML with ML-FOREST = false (*)
Evaluate ML with MLRS = false (*)
Evaluate ML with MLRS-LC = false (*)
Evaluate ML with OPCT = false (*)
Evaluate ML with Stacked ML-kNN = false (*)
Evaluate ML with DML-kNN = false (*)
Evaluate ML with Mr.KNN = false (*)
Evaluate ML with MLC-DWkNN.D = false (*)
Evaluate ML with MLC-DWkNN.M = false (*)
Evaluate ML with MLC-DWkNN.Z = false (*)
Evaluate ML with FRONEC-1.H = false (*)
Evaluate ML with FRONEC-2.H = false (*)
Evaluate ML with FRONEC-3.H = false (*)
Evaluate ML with FRONEC-1.Ld = false (*)
Evaluate ML with FRONEC-2.Ld = false (*)
Evaluate ML with FRONEC-3.Ld = false (*)
Evaluate ML with FV-kNN = false (*)
Evaluate ML with NSSC = false (*)
Evaluate ML with E-DT = false (*)
Evaluate ML with Extra-PCT = false (*)
Evaluate ML with PRUNEDSETS = false (*)
Evaluate ML with EPS = false (*)
Evaluate ML with PPT = false (*)
Evaluate ML with PPT-n = false (*)
Evaluate ML with PPT-c = false (*)
Evaluate ML with Extra-MLDT = false (*)
Evaluate ML with ZERO = false (*)
Evaluate ML with PYRAMID = false (*)
Evaluate ML with MASSIVE = true
Show ML datasets pairwise imbalance ratio = false (*)
Simulate missing labels ratio = 0.000000 (*)
Simulate noisy labels ratio = 0.000000 (*)
Evaluate ML with Bagging = false (*)
Evaluate ML with RealAdaBoost.RE = false (*)
Evaluate ML with DiscreteAdaBoost.RE = false (*)
Evaluate ML with RealAdaBoost.MH = false (*)
Evaluate ML with DiscreteAdaBoost.MH = false (*)
Evaluate ML with ML-kNN-py = false (*)
Evaluate ML with RAkELd = false (*)
Evaluate ML with RAkELo = false (*)
Evaluate ML with MLARAM = false (*)
Evaluate ML with Classifier Chain = false (*)
Evaluate ML with Binary Relevance = false (*)
Evaluate ML with Label Powerset = false (*)
Evaluate ML with BRkNNa = false (*)
Evaluate ML with BRkNNb = false (*)
Evaluate ML with MLTSVM = false (*)
Evaluate ML with MLMLP = false (*)
Evaluate ML with Deep Learning = false (*)
Evaluate ML with BOOMER = false (*)
Evaluate ML with BCC = false (*)
Evaluate ML with CDN = false (*)
Evaluate ML with HASEL = false (*)
Evaluate ML with PSt = false (*)
Evaluate ML with CT = false (*)
Evaluate ML with MCC = false (*)
Evaluate ML with NSR = false (*)
Evaluate ML with SCC = false (*)
Evaluate ML with PS = false (*)
Evaluate ML with LC = false (*)
Evaluate ML with CDT = false (*)
Evaluate ML with DBPNN = false (*)
Evaluate ML with FW = false (*)
Evaluate ML with Maniac = false (*)
Evaluate ML with PCC = false (*)
Evaluate ML with PMCC = false (*)
Exponential average alpha = 1.000000 (*)
ML ensemble threshold = 0.500000 (*)
ML ensemble set threshold by resubstitution = true (*)
CDN iterations = 1000 (*)
CDN collection iterations = 100 (*)
NSR P = 2 (*)
NSR N = 2 (*)
LImb k = 5 (*)
LImb p = 0.100000 (*)
ML-ESO do not copy unselected instances = false (*)
ML-ESO do not evaluate copies = true (*)
ML-ESO random oversampling ratio = 0.500000 (*)
ML-ESO random undersampling ratio = 0.500000 (*)
ML-ESO random evaluation ratio = 1.000000 (*)
ML-ESO use only one weight = false (*)
ML-ESO label propagation = Evolutionary (*)
ML-ESO leave one out = true (*)
ML ensemble size = 50 (*)
ML resampling method = ResamplingPerLabel (*)
ML ensemble method = None (*)
ML base classifier = ML-DT (*)
Mr.KNN f = 2.000000 (*)
Mr.KNN loo = true (*)
Option PCT epsilon = 0.100000 (*)
Option PCT O = 10 (*)
Option PCT l = 2 (*)
MLRS s = 1.000000 (*)
MLRS k = 10 (*)
MLRS beta = 0.900000 (*)
PS b = 2 (*)
PS p = 1 (*)
PS labels upper limit = 10 (*)
PS strategy = A (*)
PS classification model = DT(C4.5) (*)
EPS size = 30 (*)
EPS t = 0.300000 (*)
PPT-c k = 5 (*)
PPT-c t = 0.200000 (*)
PPT-c with no loss = true (*)
ML external base method = kNearestNeighbor (*)
ML external fitness evaluator = BinaryRelevance (*)
ML external fitness evaluation interval = 0 (*)
ML-DT adaptive threshold = false (*)
ML-RF threshold = 0.500000 (*)
ML-RF real votes = false (*)
ML-RF threshold shift = 0.000000 (*)
ML-TREE lambda = 0.500000 (*)
ML-TREE maximum depth = 5 (*)
MLMLP hidden nodes = 0 (*)
ML-kNN k = 3 (*)
ML-kNN s = 1.000000 (*)
Use neighbors cache = true (*)
SML-kNN second layer k = 10 (*)
DML-kNN delta = 3 (*)
LAML-kNN number of clusters = 10 (*)
LAML-kNN minimum number of clusters = 1 (*)
LAML-kNN maximum number of clusters = 1 (*)
LAML-k'sNN instance selection = false (*)
ML TSA threshold = -0.400000 (*)
ECC size = 10 (*)
ECC trained with bagging = true (*)
IBLR number of neighbors = 10 (*)
PYRAMID set threshold by resubstitution = true (*)
PYRAMID set threshold by G-mean = false (*)
PYRAMID fixed depth = 0 (*)
PYRAMID depth ratio = 1.000000 (*)
PYRAMID voting = false
PYRAMID prune = false (*)
PYRAMID reverse prune = false (*)
PYRAMID weight votes = false (*)
PYRAMID random search attempts = 100 (*)
PYRAMID train with PPT = false (*)
PYRAMID training set size = 5000
PYRAMID discard zero class = false (*)
PYRAMID remove duplicated powersets = false (*)
PYRAMID classification model (+0) = SVM
PYRAMID classification model (+10) = SVM
PYRAMID classification model (+25) = SVM
PYRAMID classification model (+50) = SVM
PYRAMID allow redundant classifiers = false (*)
PYRAMID random classifiers = false (*)
PYRAMID random powersets = false (*)
PYRAMID clustering method = Balanced-k-Means (*)
ML CHC initialization probability = 0.100000
ML CHC minimum distance factor = 4.000000 (*)
ML CHC random mutation = 0.050000
ML IS return overall best performing = false (*)
k ML memetic mutation ratio = 0.000000 (*)
ML Multi-selection mutation ratio = 0.000000 (*)
ML maximum intelligent oversampling ratio = 0 (*)
Massive ML method n = 1000
Massive evaluation percentage = 0.000000 (*)
Massive roulette initialization = false (*)
Massive maximum repetitions = 5
Massive drop labelsets = true
Massive retrain models at finish = false (*)
Massive use cache = true
Massive use Poisson distribution = true (*)
Poisson lambda = 4
ML Class-imbalance method = None (*)
ML Random oversampling m/M ratio = 0.100000 (*)
ML Random undersampling M/m ratio = 10.000000 (*)
ML PRUS/PROS applied to every sample = false (*)
ML PRUS ratio is relative = false (*)
ML PROS ratio is relative = false (*)
Repopulation evaluate with original labels = false (*)
Repopulation initial ratio = 0.025000 (*)
Repopulation with false labels = false (*)
Repopulate with instance selection = false (*)
Evolutionary ML evaluation set size = 0 (*)
Evolutionary ML training set size = 0 (*)
Evolutionary ML maximum evaluation set size = 0 (*)
Evolutionary ML maximum training set size = 0 (*)
Evolutionary ML set size increment = 0 (*)
ML mutate elite = false (*)
ML CC local optimization of best individual = false (*)
ML instance selection algorithm = None (*)
ML relative multi-selection = false (*)
SBC Nc = 2.000000 (*)
Initialize all minority class = false (*)
ML feature selection algorithm = None (*)
Binary learner = k-NN (*)
k-NN k = 3 (*)
BP-MLL alpha = 0.100000 (*)
BP-MLL hidden nodes = 1.000000 (*)
Subpopulation fitness weight fitness = 0.000000 (*)
Subpopulation fitness weight isolated = 0.000000 (*)
Subpopulation fitness weight difference = 0.000000 (*)
Subpopulation fitness weight substitution = 0.000000 (*)
Subpopulation fitness weight reduction = 0.000000 (*)
Subpopulation fitness weight one label = 0.000000 (*)
Subpopulation random fitness = false (*)
Subpopulation performance measure = GMean (*)
Subpopulation binary learner = k-NN (*)
Initialize ML instance selection with undersampling = false (*)
Initialize ML instance selection with roulette = false (*)
Initialize ML instance selection with cardinality = false (*)
Evolve undersampled subpopulations = false (*)
ML Use hash table = false (*)
Three way instance selection = false (*)
Three way positive to negative initial probability = 0.100000 (*)
Three way negative to positive initial probability = 0.100000 (*)
Three way only negative to positive = false (*)
Reinitialize after these number of G's with no improvement = 10
Three way evaluate with original labels = true (*)
Subpopulation memetic Drop3 mutation = 0.000000 (*)
Subpopulation memetic OSS mutation = 0.000000 (*)
Partial selection mutation = 0.000000 (*)
Partial selection mutation bit ratio = 0.100000 (*)
Partial selection single-label instance selection mutation = 0.000000 (*)
Biased HUX 1 ratio = 0.000000 (*)
Evaluation weight reduction = 0.000000
Evaluation weight repopulation reduction = 0.000000 (*)
Evaluation weight multi-selection reduction = 0.000000 (*)
Evaluation weight per label reduction = 0.000000 (*)
Evaluation weight EB subset accuracy = 0.000000
Evaluation weight EB Hamming loss = 0.000000
Evaluation weight EB Jaccard coefficient = 0.000000 (*)
Evaluation weight EB accuracy = 0.000000
Evaluation weight EB precision = 0.000000
Evaluation weight EB recall = 0.000000
Evaluation weight EB F1 = 0.000000
Evaluation weight LB accuracy macro = 0.000000
Evaluation weight LB precision macro = 0.000000
Evaluation weight LB recall macro = 0.000000
Evaluation weight LB F1 macro = 0.000000
Evaluation weight LB accuracy micro = 0.000000 (*)
Evaluation weight LB precision micro = 0.000000
Evaluation weight LB recall micro = 0.000000
Evaluation weight LB F1 micro = 0.000000
Evaluation weight EB one error = 0.000000
Evaluation weight EB coverage = 0.000000
Evaluation weight EB ranking loss = 1.000000
Evaluation weight EB average precision = 0.000000
Evaluation weight LB AUC macro = 0.000000
Evaluation weight LB AUC micro = 0.000000
Evaluation weight LB AUC instance = 0.000000
Evaluation weight LB F1 instance = 0.000000
Evaluation weight number of predicted labels = 0.000000
Evaluation k regularization weight = 0.000000 (*)
Evaluation oversampling weight = 0.000000 (*)
Evaluation false labels weight = 0.000000 (*)
Gaussian reduction = false (*)
Combine metrics using G-mean = false (*)
ML CC combine best subindividuals = false (*)
ML CC population evolutionary scheme = CHCGA (*)
ML CC return overall best = false (*)
Population size = 100
Population CHC initialization probability = 0.500000 (*)
Subpopulations size = 100 (*)
Subpopulations CHC initialization probability = 0.250000 (*)
Subpopulation CHC minimum distance factor = 8 (*)
Np = 10 (*)
SSGA Mutation = 0.010000 (*)
Nsp = 100 (*)
Reevaluate individuals each Nsp generation = false (*)
Merge method = QuorumSensing (*)
Minimum quorum sensing threshold = 0.000000 (*)
Maximum quorum sensing threshold = 1.000000 (*)
ML CC weighted subpopulations for quorum sensing = false (*)
Evolve only one subpopulation per cycle = false (*)
Select subpopulation by relevance = false (*)
ML CC allow negative subpopulation individual index = false (*)
ML-kNN minimum k = 1 (*)
ML-kNN maximum k = 100 (*)
ML-k'sNN minimum k = 1 (*)
ML-k'sNN maximum k = 100 (*)
Regularize ML-k'sNN = false (*)
ML elitism = 0.100000
ML IS k fixed by CV = false (*)
ML use HUX = true (*)
ML-k'sNN memetic mutation = 0.000000 (*)
ML-k'sNN local search mutation = 0.000000 (*)
ML-k'sNN local search mutation steps = 100 (*)
ML evaluate with leave-one-out = true (*)
Evaluate ML-k'sNN with leave one out = true (*)
ML CC memetic mutation = 0.000000 (*)
ML CC memetic mutation for subpopulations = 0.000000 (*)
ML CC memetic mutation for subpopulations bit ratio = 0.100000 (*)
Evaluation weight negative subpopulations = 0.000000 (*)
ML instance selection per label = false (*)
ML-k'sNN method = PerpendicularSearch (*)
ML-k'sNN local optimization ratio = 0.000000 (*)
ML-k'sNN rubber band penalty = 0.000000 (*)
ML-k'sNN different k's penalty = 0.000000 (*)
ML-k'sNN rounds = 1 (*)
ML-k'sNN sampling with replacement = true (*)
ML-k'sNN resolve ties with closer to cross-validated k = true (*)
ML-k'sNN use window for best K = false (*)
ML-k'sNN window size = 5 (*)
ML-k'sNN per label sampling = false (*)
ML-k'sNN update AdaBoostMH weights with current best = false (*)
ML instance selection per label with the same neighbors = false (*)
ML-k'sNN stepwise attempts = 1 (*)
ML-k'sNN penalty from global CV valued = 0.000000 (*)
OpenMP allow nested loops = false (*)
───────────────────────────
───────────────────────────
Random seed: 597334891
@Instances: 16091
@Features: 500
@Labels: 983
@Label cardinality: 19.036542
@Label density: 0.019366
@Label diversity: 15805
@Proportion of distinct labels: 0.982226
@MeanIR: 71.133810
@CVIR: 0.740520
@Multi-label instances: 0.998695

CI None method: 14482 -> 14482 instances
@CI reduction: 0.000000
None CI method REMOVED 0 instances
Pyramid m: 1000
