Image Classification on CIFAR10 standard (test) (Accuracy and Memory Saving)
91.37AccuracyE2E
Evaluation results
| Method | Links | ||
|---|---|---|---|
| E2EBackbone=Swin-Tiny Transformer, Training Protocol=End-to-End2023.09 | 91.37 | 0 | |
| Parallel TRGLBackbone=Swin-Tiny Transformer, Number of modules=4, Training Protocol=Parallel2023.09 | 86.48 | 29 | |
| InfoProLBackbone=Swin-Tiny Transformer, Number of modules=4, Training Protocol=Parallel2023.09 | 86.28 | 45 | |
| Parallel VanGLBackbone=Swin-Tiny Transformer, Number of modules=4, Training Protocol=Parallel2023.09 | 83.94 | 33 | |
| InfoProBackbone=Swin-Tiny Transformer, Number of modules=4, Training Protocol=Parallel2023.09 | 83.38 | 38 |