Skip to content

[Enhance] Add auto scale lr for maskformer_swin_l #8385

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.

Sign up for GitHub

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

Open
wants to merge 7 commits into
base: dev
Choose a base branch
from
Open

[Enhance] Add auto scale lr for maskformer_swin_l #8385

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
e7d30a2
add auto scale lr for maskformer_swin_l
chhluo Jul 19, 2022
91bfa61
update auto_scale_lr for strong_baseline
chhluo Jul 25, 2022
b7f2b5a
batch_size=16 not 64 for lad
chhluo Jul 26, 2022
4e1d895
add 2 x 8 in config file name
chhluo Jul 26, 2022
402add8
reemove comments
chhluo Jul 26, 2022
15b6d81
add batch_size
chhluo Jul 27, 2022
367bc32
fix typos
chhluo Jul 27, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 0 additions & 1 deletion configs/faster_rcnn/metafile.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -276,7 +276,6 @@ Models:
Dataset: COCO
Metrics:
box AP: 37.9
# re-release
Weights: https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_iou_1x_coco/faster_rcnn_r50_fpn_iou_1x_coco_20200506_095954-938e81f0.pth

- Name: faster_rcnn_r50_fpn_giou_1x_coco
Expand Down
13 changes: 7 additions & 6 deletions configs/lad/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,17 +20,18 @@ Distillation.

### PAA with LAD

| Teacher | Student | Training schedule | AP (val) | Config | Download |
| :-----: | :-----: | :---------------: | :------: | :---------------------------------------------------------------------------------------------------------: | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
| -- | R-50 | 1x | 40.4 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/paa/paa_r50_fpn_1x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/paa/paa_r50_fpn_1x_coco/paa_r50_fpn_1x_coco_20200821-936edec3.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/paa/paa_r50_fpn_1x_coco/paa_r50_fpn_1x_coco_20200821-936edec3.log.json) |
| -- | R-101 | 1x | 42.6 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/paa/paa_r101_fpn_1x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/paa/paa_r101_fpn_1x_coco/paa_r101_fpn_1x_coco_20200821-0a1825a4.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/paa/paa_r101_fpn_1x_coco/paa_r101_fpn_1x_coco_20200821-0a1825a4.log.json) |
| R-101 | R-50 | 1x | 41.4 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/lad/lad_r50_paa_r101_fpn_coco_1x.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/lad/lad_r50_paa_r101_fpn_coco_1x/lad_r50_paa_r101_fpn_coco_1x_20220708_124246-74c76ff0.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/lad/lad_r50_paa_r101_fpn_coco_1x/lad_r50_paa_r101_fpn_coco_1x_20220708_124246.log.json) |
| R-50 | R-101 | 1x | 43.2 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/lad/lad_r101_paa_r50_fpn_coco_1x.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/lad/lad_r101_paa_r50_fpn_coco_1x/lad_r101_paa_r50_fpn_coco_1x_20220708_124357-9407ac54.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/lad/lad_r101_paa_r50_fpn_coco_1x/lad_r101_paa_r50_fpn_coco_1x_20220708_124357.log.json) |
| Teacher | Student | Training schedule | AP (val) | Config | Download |
| :-----: | :-----: | :---------------: | :------: | :-------------------------------------------------------------------------------------------------------------: | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
| -- | R-50 | 1x | 40.4 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/paa/paa_r50_fpn_1x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/paa/paa_r50_fpn_1x_coco/paa_r50_fpn_1x_coco_20200821-936edec3.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/paa/paa_r50_fpn_1x_coco/paa_r50_fpn_1x_coco_20200821-936edec3.log.json) |
| -- | R-101 | 1x | 42.6 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/paa/paa_r101_fpn_1x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/paa/paa_r101_fpn_1x_coco/paa_r101_fpn_1x_coco_20200821-0a1825a4.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/paa/paa_r101_fpn_1x_coco/paa_r101_fpn_1x_coco_20200821-0a1825a4.log.json) |
| R-101 | R-50 | 1x | 41.4 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/lad/lad_r50_paa_r101_fpn_2x8_coco_1x.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/lad/lad_r50_paa_r101_fpn_coco_1x/lad_r50_paa_r101_fpn_coco_1x_20220708_124246-74c76ff0.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/lad/lad_r50_paa_r101_fpn_coco_1x/lad_r50_paa_r101_fpn_coco_1x_20220708_124246.log.json) |
| R-50 | R-101 | 1x | 43.2 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/lad/lad_r101_paa_r50_fpn_2x8_coco_1x.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/lad/lad_r101_paa_r50_fpn_coco_1x/lad_r101_paa_r50_fpn_coco_1x_20220708_124357-9407ac54.pth) \| [log](https://download.openmmlab.com/mmdetection/v2.0/lad/lad_r101_paa_r50_fpn_coco_1x/lad_r101_paa_r50_fpn_coco_1x_20220708_124357.log.json) |

## Note

- Meaning of Config name: lad_r50(student model)\_paa(based on paa)\_r101(teacher model)\_fpn(neck)\_coco(dataset)\_1x(12 epoch).py
- Results may fluctuate by about 0.2 mAP.
- 2 GPUs are used, 8 samples per GPU.

## Citation

Expand Down
5 changes: 0 additions & 5 deletions configs/lad/lad_r101_paa_r50_fpn_coco_1x.py → ...s/lad/lad_r101_paa_r50_fpn_2x8_coco_1x.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -119,8 +119,3 @@
data = dict(samples_per_gpu=8, workers_per_gpu=4)
optimizer = dict(lr=0.01)
fp16 = dict(loss_scale=512.)

# NOTE: `auto_scale_lr` is for automatically scaling LR,
# USER SHOULD NOT CHANGE ITS VALUES.
# base_batch_size = (8 GPUs) x (8 samples per GPU)
auto_scale_lr = dict(base_batch_size=64)
5 changes: 0 additions & 5 deletions configs/lad/lad_r50_paa_r101_fpn_coco_1x.py → ...s/lad/lad_r50_paa_r101_fpn_2x8_coco_1x.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -118,8 +118,3 @@
data = dict(samples_per_gpu=8, workers_per_gpu=4)
optimizer = dict(lr=0.01)
fp16 = dict(loss_scale=512.)

# NOTE: `auto_scale_lr` is for automatically scaling LR,
# USER SHOULD NOT CHANGE ITS VALUES.
# base_batch_size = (8 GPUs) x (8 samples per GPU)
auto_scale_lr = dict(base_batch_size=64)
8 changes: 4 additions & 4 deletions configs/lad/metafile.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,9 +19,9 @@ Collections:
Version: v2.19.0

Models:
- Name: lad_r101_paa_r50_fpn_coco_1x
- Name: lad_r101_paa_r50_fpn_2x8_coco_1x
In Collection: Label Assignment Distillation
Config: configs/lad/lad_r101_paa_r50_fpn_coco_1x.py
Config: configs/lad/lad_r101_paa_r50_fpn_2x8_coco_1x.py
Metadata:
Training Memory (GB): 12.4
Epochs: 12
Expand All @@ -31,9 +31,9 @@ Models:
Metrics:
box AP: 43.2
Weights: https://download.openmmlab.com/mmdetection/v2.0/lad/lad_r101_paa_r50_fpn_coco_1x/lad_r101_paa_r50_fpn_coco_1x_20220708_124357-9407ac54.pth
- Name: lad_r50_paa_r101_fpn_coco_1x
- Name: lad_r50_paa_r101_fpn_2x8_coco_1x
In Collection: Label Assignment Distillation
Config: configs/lad/lad_r50_paa_r101_fpn_coco_1x.py
Config: configs/lad/lad_r50_paa_r101_fpn_2x8_coco_1x.py
Metadata:
Training Memory (GB): 8.9
Epochs: 12
Expand Down
5 changes: 5 additions & 0 deletions configs/maskformer/maskformer_swin-l-p4-w12_mstrain_64x1_300e_coco.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -65,3 +65,8 @@
warmup_ratio=1e-6,
warmup_iters=1500)
runner = dict(type='EpochBasedRunner', max_epochs=300)

# NOTE: `auto_scale_lr` is for automatically scaling LR,
# USER SHOULD NOT CHANGE ITS VALUES.
# base_batch_size = (64 GPUs) x (1 samples per GPU)
auto_scale_lr = dict(base_batch_size=64)
5 changes: 5 additions & 0 deletions configs/strong_baselines/mask_rcnn_r50_caffe_fpn_syncbn-all_rpn-2conv_lsj_100e_coco.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -78,3 +78,8 @@
train=dict(dataset=dict(pipeline=train_pipeline)),
val=dict(pipeline=test_pipeline),
test=dict(pipeline=test_pipeline))

# NOTE: `auto_scale_lr` is for automatically scaling LR,
# USER SHOULD NOT CHANGE ITS VALUES.
# base_batch_size = (64 GPUs) x (1 samples per GPU)
auto_scale_lr = dict(base_batch_size=64)
5 changes: 5 additions & 0 deletions configs/strong_baselines/mask_rcnn_r50_fpn_syncbn-all_rpn-2conv_lsj_100e_coco.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,3 +20,8 @@
conv_out_channels=256,
norm_cfg=head_norm_cfg),
mask_head=dict(norm_cfg=head_norm_cfg)))

# NOTE: `auto_scale_lr` is for automatically scaling LR,
# USER SHOULD NOT CHANGE ITS VALUES.
# base_batch_size = (64 GPUs) x (1 samples per GPU)
auto_scale_lr = dict(base_batch_size=64)