Classification on Wine (test)
100AccuracySklearn
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SklearnModel architecture=Linear model, Loss function=Logistic regression, Solver=L-BFGS2022.05 | 100 | — | — | |
| Cross-entropyModel architecture=Linear model, Loss function=Cross-entropy2022.05 | 100 | — | — | |
| HingeModel architecture=Linear model, Loss function=Hinge2022.05 | 100 | — | — | |
| EXACTModel architecture=Linear model, Loss function=EXACT2022.05 | 100 | — | — | |
| LDAType=Euclidean, Number of components (nc)=3, Subsampling strategy=stratified, max 2000 samples2026.02 | 100 | — | — | |
| R-PGAType=Riemannian, Number of components (nc)=3, Subsampling strategy=stratified, max 2000 samples2026.02 | 100 | — | — | |
| R-RPCAType=Riemannian, Number of components (nc)=3, Subsampling strategy=stratified, max 2000 samples2026.02 | 100 | — | — | |
| R-LDAType=Riemannian, Number of components (nc)=3, Subsampling strategy=stratified, max 2000 samples2026.02 | 100 | — | — | |
| R-IsomapType=Riemannian, Number of components (nc)=3, Subsampling strategy=stratified, max 2000 samples2026.02 | 100 | — | — | |
| KANSample size=178, Validation protocol=5-fold cross-validation2026.05 | 98.3 | — | — | |
| PCAType=Euclidean, Number of components (nc)=3, Subsampling strategy=stratified, max 2000 samples2026.02 | 98.2 | — | — | |
| R-LEType=Riemannian, Number of components (nc)=3, Subsampling strategy=stratified, max 2000 samples2026.02 | 98.2 | — | — | |
| ReMLPSelection Category=Dynamic, Optimization Strategy=RL2025.08 | 98.15 | — | — | |
| MLPSelection Category=Dynamic, Optimization Strategy=RL2025.08 | 97.84 | — | — | |
| Soft LearningSample size=178, Validation protocol=5-fold cross-validation2026.05 | 97.8 | — | — | |
| RFSample size=178, Validation protocol=5-fold cross-validation2026.05 | 97.7 | — | — | |
| XBNet2021.06 | 97.22 | — | — | |
| XGBoost2021.06 | 97.22 | — | — | |
| CatBoostSample size=178, Validation protocol=5-fold cross-validation2026.05 | 97.2 | — | — | |
| IsomapType=Euclidean, Number of components (nc)=3, Subsampling strategy=stratified, max 2000 samples2026.02 | 96.3 | — | — | |
| VIPSelection Category=Dynamic, Optimization Strategy=Greedy2025.08 | 96.29 | — | — | |
| GBTSample size=178, Validation protocol=5-fold cross-validation2026.05 | 96.1 | — | — | |
| Static CMISelection Category=Static2025.08 | 95.37 | — | — | |
| R-ONPPType=Riemannian, Number of components (nc)=3, Subsampling strategy=stratified, max 2000 samples2026.02 | 94.4 | — | — | |
| CARTSelection Category=Dynamic, Optimization Strategy=Greedy2025.08 | 92.28 | — | — | |
| Tuned MLPSample size=178, Validation protocol=5-fold cross-validation2026.05 | 92.1 | — | — | |
| CARTSelection Category=Dynamic, Optimization Strategy=RL2025.08 | 91.36 | — | — | |
| FtreeSelection Category=Dynamic, Optimization Strategy=RL2025.08 | 90.74 | — | — | |
| XGBoost2026.06 | 88.64 | — | — | |
| XGBoostDeferral rate=0%2026.06 | 88.64 | — | 0.0022 | |
| Random Forest2026.06 | 88.48 | — | — | |
| Random ForestDeferral rate=0%2026.06 | 88.48 | — | 0.0042 | |
| MDT+XGBmax_test_deferral_rate=25%2026.06 | 87.88 | — | — | |
| MDT+XGBDeferral rate=≤25%2026.06 | 87.88 | — | 0.0052 | |
| FTreeSelection Category=Dynamic, Optimization Strategy=Greedy2025.08 | 87.34 | — | — | |
| TabNetSelection Category=Static2025.08 | 86.73 | — | — | |
| ReMLPSelection Category=Dynamic, Optimization Strategy=Greedy2025.08 | 84.57 | — | — | |
| FIGS2026.06 | 83.91 | — | — | |
| FIGSDeferral rate=0%2026.06 | 83.91 | — | 0.0039 | |
| Non-Deferral Tree2026.06 | 83.4 | — | — | |
| Non-Deferral TreeDeferral rate=0%2026.06 | 83.4 | — | 0.0015 | |
| CWCFSelection Category=Dynamic, Optimization Strategy=RL2025.08 | 83.33 | — | — | |
| DIMESelection Category=Dynamic, Optimization Strategy=Greedy2025.08 | 79.01 | — | — | |
| INVASESelection Category=Dynamic, Optimization Strategy=Greedy2025.08 | 78.71 | — | — | |
| MLPSelection Category=Dynamic, Optimization Strategy=Greedy2025.08 | 76.54 | — | — | |
| ArrowFlowN=178, Feat=13, Cls=3, ArrowFlow Config=[128] e=64 p=12026.04 | — | 2.8 | — | |
| DTE-1variant=DTE-12025.12 | — | 8.2 | — | |
| DTE-3variant=DTE-32025.12 | — | 2.4 | — | |
| Forest2025.12 | — | 2 | — | |
| KNNN=178, Feat=13, Cls=32026.04 | — | 0 | — | |
| MLPN=178, Feat=13, Cls=32026.04 | — | 2.8 | — | |
| RFN=178, Feat=13, Cls=32026.04 | — | 0 | — | |
| S-NN2025.12 | — | 2.4 | — | |
| SVMN=178, Feat=13, Cls=32026.04 | — | 2.8 | — | |
| Tree2025.12 | — | 10.3 | — | |
| XGBN=178, Feat=13, Cls=32026.04 | — | 2.8 | — |