Pixel-level Image Classification on Image 4K Long Range Arena (LRA) (test)
40.4Accuracy (%)Efficient Attention + GFSA
Evaluation results
| Method | Links | |
|---|---|---|
| Efficient Attention + GFSARuntime (s per 1,000 steps)=135.8, GPU usage (GB)=1.332023.12 | 40.4 | |
| Transformer+GFSARuntime (s per 1,000 steps)=737.2, GPU usage (GB)=11.22023.12 | 40.2 | |
| Efficient AttentionRuntime (s per 1,000 steps)=121.1, GPU usage (GB)=1.142023.12 | 40.2 | |
| YOSO-ERuntime (s per 1,000 steps)=114.2, GPU usage (GB)=1.422023.12 | 39.8 | |
| TransformerRuntime (s per 1,000 steps)=345.1, GPU usage (GB)=5.882023.12 | 38.2 | |
| LinformerRuntime (s per 1,000 steps)=158.5, GPU usage (GB)=3.452023.12 | 37.8 |