| semantic-segmentation-on-ade20k | DAT-B++ | #106 | Validation mIoU: 51.5 |
| semantic-segmentation-on-ade20k | DAT-S++ | #109 | Validation mIoU: 51.2 |
| semantic-segmentation-on-ade20k | DAT-T++ | #125 | Validation mIoU: 50.3 |
| image-classification-on-imagenet | DAT-B++ (384x384) | #200 | Top 1 Accuracy: 85.9%Number of params: 94MGFLOPs: 49.7 |
| image-classification-on-imagenet | DAT-B++ (224x224) | #300 | Top 1 Accuracy: 84.9%Number of params: 93MGFLOPs: 16.6 |
| image-classification-on-imagenet | DAT-S++ | #327 | Top 1 Accuracy: 84.6%Number of params: 53MGFLOPs: 9.4 |
| image-classification-on-imagenet | DAT-T++ | #408 | Top 1 Accuracy: 83.9%Number of params: 24MGFLOPs: 4.3 |
| object-detection-on-coco-val2017 | DAT-S++ (RetinaNet 3x) | #213 | box AP: 50.2AP50: 71.5AP75: 54.0APS: 34.7APM: 54.6APL: 65.3… |
| object-detection-on-coco-val2017 | DAT-T++ (RetinaNet 3x) | #235 | box AP: 49.2AP50: 70.3AP75: 53.0APS: 32.7APM: 53.4APL: 64.7… |