List your company
ResearchFormerly WizWand

Classification on Cardiac Phenotyping 5-fold CV (test)

81AccuracyCW-B
74.7676.387879.62Jun 29, 2026
Updated 4mo ago

Evaluation results

MethodLinks
CW-BValidation Protocol=5-fold stratified cross-validation, Boosting iterations=600, Depth=5, Learning rate=0.07, Subsampling=0.95, l2=1.02026.068172736769330.0103
XGBValidation Protocol=5-fold stratified cross-validation, Boosting iterations=1200, Depth=7, Learning rate=0.03, Subsampling=0.95, l2=1.02026.068069666567350.0108
CBValidation Protocol=5-fold stratified cross-validation, Boosting iterations=1200, Depth=8, Learning rate=0.032026.067870706666340.0136
BCValidation Protocol=5-fold stratified cross-validation, Architecture=MLP(256), Optimizer=Adam, Learning rate=10−32026.067768676264380.0109
DQNValidation Protocol=5-fold stratified cross-validation, Architecture=MLP(256), Optimizer=Adam, Learning rate=10−32026.067668686664340.007
MLPValidation Protocol=5-fold stratified cross-validation, Architecture=MLP(256,128), Features=early stopping, Optimizer=Adam, Learning rate=10−32026.067665635860420.018
STKValidation Protocol=5-fold stratified cross-validation, Base learners=CatBoost and XGBoost, Meta learner=logistic regression, Training mode=out of fold training2026.067569716966310.0069