Make sure to set up your environment according to the semantic segmentation README.
Training and evaluation is identical to NAT.
| Backbone | Network | # of Params | FLOPs | mIoU | mIoU (multi-scale) | Pre-training | Checkpoint | Config file |
|---|---|---|---|---|---|---|---|---|
| DiNAT-Mini | UPerNet | 50M | 900G | 45.8 | 47.2 | ImageNet-1K | Download | config.py |
| DiNAT-Tiny | UPerNet | 58M | 934G | 47.8 | 48.8 | ImageNet-1K | Download | config.py |
| DiNAT-Small | UPerNet | 82M | 1010G | 48.9 | 49.9 | ImageNet-1K | Download | config.py |
| DiNAT-Base | UPerNet | 123M | 1137G | 49.6 | 50.4 | ImageNet-1K | Download | config.py |
| DiNAT-Large | UPerNet | 238M | 2335G | 54.0 | 54.9 | ImageNet-22K | Download | config.py |
| Backbone | Network | # of Params | FLOPs | mIoU | mIoU (multi-scale) | Pre-training | Checkpoint | Config file |
|---|---|---|---|---|---|---|---|---|
| DiNATs-Tiny | UPerNet | 60M | 941G | 46.0 | 47.4 | ImageNet-1K | Download | config.py |
| DiNATs-Small | UPerNet | 81M | 1030G | 48.6 | 49.9 | ImageNet-1K | Download | config.py |
| DiNATs-Base | UPerNet | 121M | 1173G | 49.4 | 50.2 | ImageNet-1K | Download | config.py |
| DiNATs-Large | UPerNet | 234M | 2466G | 53.4 | 54.6 | ImageNet-22K | Download | config.py |