Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

add dn_detr algo #119

Merged
merged 163 commits into from
Jul 27, 2022
Merged

add dn_detr algo #119

merged 163 commits into from
Jul 27, 2022

Conversation

tuofeilunhifi
Copy link
Collaborator

@tuofeilunhifi tuofeilunhifi commented Jul 6, 2022

  1. support dn-detr-r50: 44.39
  2. refactor set_criterion

@tuofeilunhifi tuofeilunhifi changed the title [WIP] dn_detr [WIP] add dn_detr algo Jul 6, 2022
@tuofeilunhifi tuofeilunhifi changed the title [WIP] add dn_detr algo add dn_detr algo Jul 14, 2022
@@ -119,3 +119,15 @@

data = dict(
imgs_per_gpu=2, workers_per_gpu=2, train=train_dataset, val=val_dataset)

# evaluation
eval_config = dict(initial=True, interval=1, gpu_collect=False)
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

set initial to False

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

done

@@ -1,8 +1,7 @@
# model settings
model = dict(
type='Detection',
pretrained=
'http://pai-vision-data-hz.oss-cn-zhangjiakou.aliyuncs.com/pretrained_models/easycv/resnet/torchvision/resnet50.pth',
pretrained=True,
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

if user set pretrained=True, and the model does not have a default url, an error should be raised directly

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

done

## DETR

| Algorithm | Config | bbox_mAP<sup>val<br/><sub>0.5:0.95</sub> | AP<sup>val<br/><sub>50</sub> | Download |
| ---------- | ------------------------------------------------------------ | ------------------------ | --------------- | ------------------------------------------------------------ |
| DETR-r50 | [detr-r50](https://github.com/alibaba/EasyCV/tree/master/configs/detection/detr/detr_r50_8x2_150e_coco.py) | 39.92 | 60.52 | [model](https://pai-vision-data-hz.oss-cn-zhangjiakou.aliyuncs.com/EasyCV/modelzoo/detection/detr/epoch_150.pth) - [log](https://pai-vision-data-hz.oss-cn-zhangjiakou.aliyuncs.com/EasyCV/modelzoo/detection/detr/20220609_101243.log.json) |
| DAB-DETR-r50 | [dab-detr-r50](https://github.com/alibaba/EasyCV/tree/master/configs/detection/dab_detr/dab_detr_r50_8x2_50e_coco.py) | 42.52 | 63.03 | [model](https://pai-vision-data-hz.oss-cn-zhangjiakou.aliyuncs.com/EasyCV/modelzoo/detection/dab_detr/epoch_50.pth) - [log](https://pai-vision-data-hz.oss-cn-zhangjiakou.aliyuncs.com/EasyCV/modelzoo/detection/dab_detr/20220610_122811.log.json) |
| DAB-DETR-r50 | [dab-detr-r50](https://github.com/alibaba/EasyCV/tree/master/configs/detection/dab_detr/dab_detr_r50_8x2_50e_coco.py) | 42.52 | 63.03 | [model](https://pai-vision-data-hz.oss-cn-zhangjiakou.aliyuncs.com/EasyCV/modelzoo/detection/dab_detr/dab_detr_epoch_50.pth) - [log](https://pai-vision-data-hz.oss-cn-zhangjiakou.aliyuncs.com/EasyCV/modelzoo/detection/dab_detr/20220610_122811.log.json) |
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

add Params, flops, inference time for added model

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

done

@tuofeilunhifi tuofeilunhifi merged commit 9517bb8 into alibaba:master Jul 27, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants