30 lines
968 B
YAML
30 lines
968 B
YAML
Collections:
|
|
- Name: CSRA
|
|
Metadata:
|
|
Training Data: PASCAL VOC 2007
|
|
Architecture:
|
|
- Class-specific Residual Attention
|
|
Paper:
|
|
URL: https://arxiv.org/abs/1911.11929
|
|
Title: 'Residual Attention: A Simple but Effective Method for Multi-Label Recognition'
|
|
README: configs/csra/README.md
|
|
Code:
|
|
Version: v0.24.0
|
|
URL: https://github.com/open-mmlab/mmclassification/blob/v0.24.0/mmcls/models/heads/multi_label_csra_head.py
|
|
|
|
Models:
|
|
- Name: resnet101-csra_1xb16_voc07-448px
|
|
Metadata:
|
|
FLOPs: 4120000000
|
|
Parameters: 23550000
|
|
In Collections: CSRA
|
|
Results:
|
|
- Dataset: PASCAL VOC 2007
|
|
Metrics:
|
|
mAP: 94.98
|
|
OF1: 90.80
|
|
CF1: 89.16
|
|
Task: Multi-Label Classification
|
|
Weights: https://download.openmmlab.com/mmclassification/v0/csra/resnet101-csra_1xb16_voc07-448px_20220722-29efb40a.pth
|
|
Config: configs/csra/resnet101-csra_1xb16_voc07-448px.py
|