Classification on ECG
99.84AccuracyCNN with 11 layers
Evaluation results
| Method | Links | ||
|---|---|---|---|
| CNN with 11 layersBounds=✗2026.03 | 99.84 | — | |
| CNN with frequency channel attentionBounds=✗2026.03 | 99.6 | — | |
| SVM with tunable Q-wavelet transformBounds=✗2026.03 | 99.27 | — | |
| Long short-term memory model with focal lossBounds=✗2026.03 | 99.26 | — | |
| Optimized decision treeBounds=✗2026.03 | 98.77 | — | |
| TSLANetAvg Params=762.32K, Avg GFlops=2.12G2025.08 | 98.51 | — | |
| LITEAvg Params=16.84K, Avg GFlops=0.27G2025.08 | 98.4 | — | |
| MambaAvg Params=235.81K, Avg GFlops=5.11G2025.08 | 98.27 | — | |
| TimesNetAvg Params=56.69M, Avg GFlops=4.07T2025.08 | 98.21 | — | |
| GP classification with boundsBounds=✓2026.03 | 98 | — | |
| PRISMAvg Params=13.59K, Avg GFlops=0.04G2025.08 | 97.76 | — | |
| MiniROCKETAvg Params=72.55K, Avg GFlops=0.55G2025.08 | 97.76 | — | |
| FiLMAvg Params=12.59M, Avg GFlops=0.58G2025.08 | 97.63 | — | |
| LightTSAvg Params=1.85M, Avg GFlops=0.07G2025.08 | 97.51 | — | |
| iTransformerAvg Params=491.73K, Avg GFlops=0.05G2025.08 | 96.89 | — | |
| PatchTSTAvg Params=485.79K, Avg GFlops=1.63G2025.08 | 93.21 | — | |
| DLinearAvg Params=3.64M, Avg GFlops=0.12G2025.08 | 90.41 | — | |
| InstructTime++Variant=Adapt2026.01 | 46.93 | 64.89 | |
| InstructTime++Variant=Universal2026.01 | 42.45 | 58.72 | |
| InstructTimeVariant=Adapt2026.01 | 41.21 | 55.47 | |
| FormerTime2026.01 | 37.12 | 52.33 | |
| InstructTimeVariant=Universal2026.01 | 34.02 | 48.2 | |
| MiniROCKET2026.01 | 26.89 | 39 | |
| TimeMAE2026.01 | 25.46 | 38.34 | |
| Patch Transformer2026.01 | 24.65 | 38.83 | |
| GPT-As-Classifier2026.01 | 22.53 | 35.57 | |
| Transformer2026.01 | 18.21 | 36.91 | |
| TS-TCC2026.01 | 17.78 | 37.8 | |
| TCN2026.01 | 10.14 | 16.54 | |
| MCDCNN2026.01 | 9.29 | 17.35 |