kyxt
/
code_test
Not watched
1
0
Fork 0
Code
Releases 0 Wiki evaluate Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain HPC
You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
9 Commits
1 Branch
316 MB
21 Download
Tree: 501f3c46ea
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from '501f3c46ea'
${ noResults }
code_test/configs/swin/README.md

README.md 2.9 kB
Raw Normal View History

first commit
2 years ago
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
 12345678910111213141516171819202122232425 
  1. # Swin Transformer: Hierarchical Vision Transformer using Shifted Windows

  2. 

  3. ## Introduction

  4. 

  5. <!-- [ALGORITHM] -->

  6. 

  7. ```latex

  8. @article{liu2021Swin,

  9.     title={Swin Transformer: Hierarchical Vision Transformer using Shifted Windows},

  10.     author={Liu, Ze and Lin, Yutong and Cao, Yue and Hu, Han and Wei, Yixuan and Zhang, Zheng and Lin, Stephen and Guo, Baining},

  11.     journal={arXiv preprint arXiv:2103.14030},

  12.     year={2021}

  13. }

  14. ```

  15. 

  16. ## Results and models

  17. 

  18. ### Mask R-CNN

  19. 

  20. | Backbone | Pretrain    | Lr schd | Multi-scale crop     |   FP16   |Mem (GB) | Inf time (fps) | box AP | mask AP |  Config  |   Download  |

  21. | :------: | :---------: | :-----: | :-------------------:| :------: |:------: | :------------: | :----: | :-----: | :------: |  :--------: |

  22. |  Swin-T  | ImageNet-1K |    1x   |        no            |    no    |   7.6   |                |  42.7  |  39.3   | [config](./mask_rcnn_swin-t-p4-w7_fpn_1x_coco.py)             | [model](https://download.openmmlab.com/mmdetection/v2.0/swin/mask_rcnn_swin-t-p4-w7_fpn_1x_coco/mask_rcnn_swin-t-p4-w7_fpn_1x_coco_20210902_120937-9d6b7cfa.pth)  &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/swin/mask_rcnn_swin-t-p4-w7_fpn_1x_coco/mask_rcnn_swin-t-p4-w7_fpn_1x_coco_20210902_120937.log.json) |

  23. |  Swin-T  | ImageNet-1K |    3x   |        yes           |    no    |   10.2  |                |  46.0  |  41.6   | [config](./mask_rcnn_swin-t-p4-w7_fpn_ms-crop-3x_coco.py)     | [model](https://download.openmmlab.com/mmdetection/v2.0/swin/mask_rcnn_swin-t-p4-w7_fpn_ms-crop-3x_coco/mask_rcnn_swin-t-p4-w7_fpn_ms-crop-3x_coco_20210906_131725-bacf6f7b.pth)  &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/swin/mask_rcnn_swin-t-p4-w7_fpn_ms-crop-3x_coco/mask_rcnn_swin-t-p4-w7_fpn_ms-crop-3x_coco_20210906_131725.log.json) |

  24. |  Swin-T  | ImageNet-1K |    3x   |        yes           |    yes   |   7.8   |                |  46.0  |  41.7   | [config](./mask_rcnn_swin-t-p4-w7_fpn_fp16_ms-crop-3x_coco.py)| [model](https://download.openmmlab.com/mmdetection/v2.0/swin/mask_rcnn_swin-t-p4-w7_fpn_fp16_ms-crop-3x_coco/mask_rcnn_swin-t-p4-w7_fpn_fp16_ms-crop-3x_coco_20210908_165006-90a4008c.pth)  &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/swin/mask_rcnn_swin-t-p4-w7_fpn_fp16_ms-crop-3x_coco/mask_rcnn_swin-t-p4-w7_fpn_fp16_ms-crop-3x_coco_20210908_165006.log.json) |

  25. |  Swin-S  | ImageNet-1K |    3x   |        yes           |    yes   |   11.9  |                |  48.2  |  43.2   | [config](./mask_rcnn_swin-s-p4-w7_fpn_fp16_ms-crop-3x_coco.py)| [model](https://download.openmmlab.com/mmdetection/v2.0/swin/mask_rcnn_swin-s-p4-w7_fpn_fp16_ms-crop-3x_coco/mask_rcnn_swin-s-p4-w7_fpn_fp16_ms-crop-3x_coco_20210903_104808-b92c91f1.pth)  &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/swin/mask_rcnn_swin-s-p4-w7_fpn_fp16_ms-crop-3x_coco/mask_rcnn_swin-s-p4-w7_fpn_fp16_ms-crop-3x_coco_20210903_104808.log.json) |

No Description

Contributors (3)
All