metafile.yml 961 B

1234567891011121314151617181920212223242526272829
  1. Collections:
  2. - Name: CSRA
  3. Metadata:
  4. Training Data: PASCAL VOC 2007
  5. Architecture:
  6. - Class-specific Residual Attention
  7. Paper:
  8. URL: https://arxiv.org/abs/2108.02456
  9. Title: 'Residual Attention: A Simple but Effective Method for Multi-Label Recognition'
  10. README: configs/csra/README.md
  11. Code:
  12. Version: v0.24.0
  13. URL: https://github.com/open-mmlab/mmpretrain/blob/v0.24.0/mmcls/models/heads/multi_label_csra_head.py
  14. Models:
  15. - Name: resnet101-csra_1xb16_voc07-448px
  16. Metadata:
  17. FLOPs: 4120000000
  18. Parameters: 23550000
  19. In Collection: CSRA
  20. Results:
  21. - Dataset: PASCAL VOC 2007
  22. Metrics:
  23. mAP: 94.98
  24. OF1: 90.80
  25. CF1: 89.16
  26. Task: Multi-Label Classification
  27. Weights: https://download.openmmlab.com/mmclassification/v0/csra/resnet101-csra_1xb16_voc07-448px_20220722-29efb40a.pth
  28. Config: configs/csra/resnet101-csra_1xb16_voc07-448px.py