Knowledge Distillation on CIFAR-100
71.72AccuracyNAYER
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| NAYERTeacher model=WRN40-2, Student model=WRN16-2, GPU=NVIDIA A100, Training iterations (E)=3002023.09 | 71.72 | 7.22 | 4.47 | |
| NAYERTeacher model=WRN40-2, Student model=WRN16-2, GPU=NVIDIA A100, Training iterations (E)=2002023.09 | 71.22 | 4.03 | 7.93 | |
| NAYERTeacher model=WRN40-2, Student model=WRN16-2, GPU=NVIDIA A100, Training iterations (E)=1002023.09 | 70.2 | 2.15 | 14.88 | |
| SpaceshipNetTeacher model=WRN40-2, Student model=WRN16-2, GPU=NVIDIA A1002023.09 | 69.95 | 22.37 | 1.39 | |
| CMITeacher model=WRN40-2, Student model=WRN16-2, GPU=NVIDIA A1002023.09 | 68.75 | 24.01 | 1.3 | |
| FMTeacher model=WRN40-2, Student model=WRN16-2, GPU=NVIDIA A100, Training iterations (E)=3002023.09 | 68.25 | 7.56 | 4.29 | |
| FMTeacher model=WRN40-2, Student model=WRN16-2, GPU=NVIDIA A100, Training iterations (E)=2002023.09 | 67.75 | 4.42 | 7.46 | |
| FMTeacher model=WRN40-2, Student model=WRN16-2, GPU=NVIDIA A100, Training iterations (E)=1002023.09 | 67.15 | 2.23 | 14.17 | |
| DFQTeacher model=WRN40-2, Student model=WRN16-2, GPU=NVIDIA A1002023.09 | 64.79 | 3.31 | 9.73 | |
| MADTeacher model=WRN40-2, Student model=WRN16-2, GPU=NVIDIA A1002023.09 | 64.05 | 26.45 | 1.78 | |
| DeepInvTeacher model=WRN40-2, Student model=WRN16-2, GPU=NVIDIA A1002023.09 | 61.34 | 31.23 | 1 | |
| ZSKTTeacher model=WRN40-2, Student model=WRN16-2, GPU=NVIDIA A1002023.09 | 54.59 | 3.44 | 9.08 |