[Enhance] Add `hparams` argument in `AutoAugment` and `RandAugment` and some other improvement. by mzr1996 · Pull Request #398 · open-mmlab/mmpretrain

mzr1996 · 2021-08-09T09:20:01Z

Motivation

In AutoAugment and RandAugment, there are many common arguments in sub-policies. Now we have to set these arguments one by one.

Modification

Add hparams argument in AutoAugment and RandAugment, so that we can specify common arguments by once.
Use hparams and base variable to simplify RandAug settings in swin-transformer.
Now pad_val supports using Sequence instead of only tuple.
Use smaller test images to speed up unit tests.

BC-breaking (Optional)

No.

Use cases (Optional)

>>> policies = [
...     dict(
...         type='Translate',
...         magnitude_key='magnitude',
...         magnitude_range=(0, 1),
...         pad_val=128,
...         direction='horizontal'),
...     dict(
...         type='Rotate',
...         magnitude_key='angle',
...         magnitude_range=(0, 90))
... ]
>>> t = RandAugment(
...       policies=policies, 
...       num_policies=2, 
...       magnititude_level=12, 
...       hparams=dict(pad_val=15))
>>> # hparams won't override setting in policies config
>>> t.policies[0]['pad_val']
128
>>> # use hparams if not set in policies config
>>> t.policies[1]['pad_val']
15

Checklist

Before PR:

Pre-commit or other linting tools are used to fix the potential lint issues.
Bug fixes are fully covered by unit tests, the case that causes the bug should be added in the unit tests.
The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness.
The documentation has been modified accordingly, like docstring or example tutorials.

After PR:

If the modification has potential influence on downstream or other related projects, this PR should be tested with those projects, like MMDet or MMSeg.
CLA has been signed and all committers have signed the CLA in this PR.

And `pad_val` supports sequence instead of tuple only.

codecov · 2021-08-09T09:46:07Z

Codecov Report

Merging #398 (83ee8f8) into master (a747458) will increase coverage by 0.21%.
The diff coverage is 90.62%.

@@            Coverage Diff             @@
##           master     #398      +/-   ##
==========================================
+ Coverage   76.51%   76.72%   +0.21%     
==========================================
  Files          97       97              
  Lines        5224     5238      +14     
  Branches      856      861       +5     
==========================================
+ Hits         3997     4019      +22     
+ Misses       1108     1098      -10     
- Partials      119      121       +2

Flag	Coverage Δ
unittests	`76.72% <90.62%> (+0.21%)`	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
mmcls/datasets/pipelines/auto_augment.py	`79.13% <90.62%> (+0.35%)`	⬆️
mmcls/models/heads/linear_head.py	`82.60% <0.00%> (-3.60%)`	⬇️
mmcls/models/heads/vision_transformer_head.py	`91.89% <0.00%> (-1.14%)`	⬇️
mmcls/models/heads/multi_label_head.py	`64.70% <0.00%> (-0.92%)`	⬇️
mmcls/models/heads/stacked_head.py	`96.92% <0.00%> (-0.18%)`	⬇️
mmcls/apis/test.py	`24.13% <0.00%> (ø)`
mmcls/apis/train.py	`22.72% <0.00%> (ø)`
mmcls/datasets/voc.py	`38.46% <0.00%> (ø)`
mmcls/utils/logger.py	`75.00% <0.00%> (ø)`
... and 87 more

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update a747458...83ee8f8. Read the comment docs.

Ezra-Yu

Do we need to provide auto_aug configuration file similar to rand_aug.py?

mmcls/datasets/pipelines/auto_augment.py

configs/_base_/datasets/pipeline/rand_aug.py

mmcls/datasets/pipelines/auto_augment.py

mzr1996 · 2021-08-20T03:53:06Z

Do we need to provide auto_aug configuration file similar to rand_aug.py?

AutoAugment configuration will be added in ViT(#395)

Ezra-Yu

LGTM

…nd some other improvement. (open-mmlab#398) * Add hparams argument in `AutoAugment` and `RandAugment`. And `pad_val` supports sequence instead of tuple only. * Add unit tests for `AutoAugment` and `hparams` in `RandAugment`. * Use smaller test image to speed up uni tests. * Use hparams to simplify RandAugment config in swin-transformer. * Rename augment config name from `pipeline` to `pipelines`. * Add some commnet ad docstring.

* Add RepVGG code. * Add se_module as plugin. * Add the repvggA0 primitive config * Change repvggA0.py to fit mmcls * Add RepVGG configs * Add repvgg_to_mmcls * Add tools/deployment/convert_repvggblock_param_to_deploy.py * Change configs/repvgg/README.md * Streamlining the number of configuration files. * Fix lints * Delete plugins * Delete code about plugin. * Modify the code for using se module. * Modify config to fit repvgg with se. * Change se_cfg to allow loading of pre-training parameters. * Reduce the complexity of the configuration file. * Finsh unitest for repvgg. * Fix bug about se in repvgg_to_mmcls. * Rename convert_repvggblock_param_to_deploy.py to reparameterize_repvgg.py, and delete setting about device. * test commit * test commit * test commit command * Modify repvgg.py to make the code more readable. * Add value=0 in F.pad() * Add se_cfg to arch_settings. * Fix bug. * modeify some attr name and Update unit tests * rename stage_0 to stem and branch_identity to branch_norm * update unit tests * add m.eval in unit tests * [Enhance] Enhence SE layer to support custom squeeze channels. (#417) * add enhenced SE * Update * rm basechannel * fix docstring * Update se_layer.py fix docstring * [Docs] Add algorithm readme and update meta yml (#418) * Add README.md for models without checkpoints. * Update model-index.yml * Update metafile.yml of seresnet * [Enhance] Add `hparams` argument in `AutoAugment` and `RandAugment` and some other improvement. (#398) * Add hparams argument in `AutoAugment` and `RandAugment`. And `pad_val` supports sequence instead of tuple only. * Add unit tests for `AutoAugment` and `hparams` in `RandAugment`. * Use smaller test image to speed up uni tests. * Use hparams to simplify RandAugment config in swin-transformer. * Rename augment config name from `pipeline` to `pipelines`. * Add some commnet ad docstring. * [Feature] Support classwise weight in losses (#388) * Add classwise weight in losses:CE,BCE,softBCE * Update unit test * rm some extra code * rm some extra code * fix broadcast * fix broadcast * update unit tests * use new_tensor * fix lint * [Enhance] Better result visualization (#419) * Imporve result visualization to support wait time and change the backend to matplotlib. * Add unit test for visualization * Add adaptive dpi function * Rename `imshow_cls_result` to `imshow_infos`. * Support str in `imshow_infos` * Improve docstring. * Bump version to v0.15.0 (#426) * [CI] Add PyTorch 1.9 and Python 3.9 build workflow, and remove some CI. (#422) * Add PyTorch 1.9 build workflow, and remove some CI. * Add Python 3.9 CI * Show Python 3.9 support. * [Enhance] Rename the option `--options` in some tools to `--cfg-options`. (#425) * [Docs] Fix sphinx version (#429) * [Docs] Add `CITATION.cff` (#428) * Add CITATION.cff * Fix typo in setup.py * Change author in setup.py * modeify some attr name and Update unit tests * rename stage_0 to stem and branch_identity to branch_norm * update unit tests * add m.eval in unit tests * Update unit tests * refactor * refactor * Alignment inference accuracy * Update configs, readme and metafile * Update readme * return tuple and fix metafile * fix unit test * rm regnet and classifiers changes * update auto_aug * update metafile & readme * use delattr * rename cfgs * Update checkpoint url * Update readme * Rename config files. * Update readme and metafile * add comment * Update mmcls/models/backbones/repvgg.py Co-authored-by: Ma Zerun <mzr1996@163.com> * Update docstring * Improve docstring. * Update unittest_testblock Co-authored-by: Ezra-Yu <1105212286@qq.com> Co-authored-by: Ma Zerun <mzr1996@163.com>

…nd some other improvement. (open-mmlab#398) * Add hparams argument in `AutoAugment` and `RandAugment`. And `pad_val` supports sequence instead of tuple only. * Add unit tests for `AutoAugment` and `hparams` in `RandAugment`. * Use smaller test image to speed up uni tests. * Use hparams to simplify RandAugment config in swin-transformer. * Rename augment config name from `pipeline` to `pipelines`. * Add some commnet ad docstring.

* Add RepVGG code. * Add se_module as plugin. * Add the repvggA0 primitive config * Change repvggA0.py to fit mmcls * Add RepVGG configs * Add repvgg_to_mmcls * Add tools/deployment/convert_repvggblock_param_to_deploy.py * Change configs/repvgg/README.md * Streamlining the number of configuration files. * Fix lints * Delete plugins * Delete code about plugin. * Modify the code for using se module. * Modify config to fit repvgg with se. * Change se_cfg to allow loading of pre-training parameters. * Reduce the complexity of the configuration file. * Finsh unitest for repvgg. * Fix bug about se in repvgg_to_mmcls. * Rename convert_repvggblock_param_to_deploy.py to reparameterize_repvgg.py, and delete setting about device. * test commit * test commit * test commit command * Modify repvgg.py to make the code more readable. * Add value=0 in F.pad() * Add se_cfg to arch_settings. * Fix bug. * modeify some attr name and Update unit tests * rename stage_0 to stem and branch_identity to branch_norm * update unit tests * add m.eval in unit tests * [Enhance] Enhence SE layer to support custom squeeze channels. (open-mmlab#417) * add enhenced SE * Update * rm basechannel * fix docstring * Update se_layer.py fix docstring * [Docs] Add algorithm readme and update meta yml (open-mmlab#418) * Add README.md for models without checkpoints. * Update model-index.yml * Update metafile.yml of seresnet * [Enhance] Add `hparams` argument in `AutoAugment` and `RandAugment` and some other improvement. (open-mmlab#398) * Add hparams argument in `AutoAugment` and `RandAugment`. And `pad_val` supports sequence instead of tuple only. * Add unit tests for `AutoAugment` and `hparams` in `RandAugment`. * Use smaller test image to speed up uni tests. * Use hparams to simplify RandAugment config in swin-transformer. * Rename augment config name from `pipeline` to `pipelines`. * Add some commnet ad docstring. * [Feature] Support classwise weight in losses (open-mmlab#388) * Add classwise weight in losses:CE,BCE,softBCE * Update unit test * rm some extra code * rm some extra code * fix broadcast * fix broadcast * update unit tests * use new_tensor * fix lint * [Enhance] Better result visualization (open-mmlab#419) * Imporve result visualization to support wait time and change the backend to matplotlib. * Add unit test for visualization * Add adaptive dpi function * Rename `imshow_cls_result` to `imshow_infos`. * Support str in `imshow_infos` * Improve docstring. * Bump version to v0.15.0 (open-mmlab#426) * [CI] Add PyTorch 1.9 and Python 3.9 build workflow, and remove some CI. (open-mmlab#422) * Add PyTorch 1.9 build workflow, and remove some CI. * Add Python 3.9 CI * Show Python 3.9 support. * [Enhance] Rename the option `--options` in some tools to `--cfg-options`. (open-mmlab#425) * [Docs] Fix sphinx version (open-mmlab#429) * [Docs] Add `CITATION.cff` (open-mmlab#428) * Add CITATION.cff * Fix typo in setup.py * Change author in setup.py * modeify some attr name and Update unit tests * rename stage_0 to stem and branch_identity to branch_norm * update unit tests * add m.eval in unit tests * Update unit tests * refactor * refactor * Alignment inference accuracy * Update configs, readme and metafile * Update readme * return tuple and fix metafile * fix unit test * rm regnet and classifiers changes * update auto_aug * update metafile & readme * use delattr * rename cfgs * Update checkpoint url * Update readme * Rename config files. * Update readme and metafile * add comment * Update mmcls/models/backbones/repvgg.py Co-authored-by: Ma Zerun <mzr1996@163.com> * Update docstring * Improve docstring. * Update unittest_testblock Co-authored-by: Ezra-Yu <1105212286@qq.com> Co-authored-by: Ma Zerun <mzr1996@163.com>

mzr1996 added 4 commits August 9, 2021 17:11

Add hparams argument in AutoAugment and RandAugment.

2a72931

And `pad_val` supports sequence instead of tuple only.

Add unit tests for AutoAugment and hparams in RandAugment.

199fd24

Use smaller test image to speed up uni tests.

61cc015

Use hparams to simplify RandAugment config in swin-transformer.

b840bce

mzr1996 requested a review from Ezra-Yu August 9, 2021 09:20

Ezra-Yu reviewed Aug 12, 2021

View reviewed changes

mmcls/datasets/pipelines/auto_augment.py Show resolved Hide resolved

configs/_base_/datasets/pipeline/rand_aug.py Show resolved Hide resolved

mmcls/datasets/pipelines/auto_augment.py Outdated Show resolved Hide resolved

mzr1996 added 2 commits August 20, 2021 11:49

Rename augment config name from pipeline to pipelines.

829cfe4

Add some commnet ad docstring.

83ee8f8

mzr1996 requested a review from Ezra-Yu August 20, 2021 03:53

mzr1996 mentioned this pull request Aug 23, 2021

[Refactor] Refator ViT (Continue #295) #395

Merged

6 tasks

Ezra-Yu approved these changes Aug 24, 2021

View reviewed changes

mzr1996 merged commit 6a0a76a into open-mmlab:master Aug 24, 2021

mzr1996 deleted the aug-add-hparams branch December 7, 2022 02:07

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[Enhance] Add `hparams` argument in `AutoAugment` and `RandAugment` and some other improvement.#398

[Enhance] Add `hparams` argument in `AutoAugment` and `RandAugment` and some other improvement.#398
mzr1996 merged 6 commits intoopen-mmlab:masterfrom
mzr1996:aug-add-hparams

mzr1996 commented Aug 9, 2021

Uh oh!

codecov bot commented Aug 9, 2021 •

edited

Loading

Uh oh!

Ezra-Yu left a comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

mzr1996 commented Aug 20, 2021

Uh oh!

Ezra-Yu left a comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

mzr1996 commented Aug 9, 2021

Motivation

Modification

BC-breaking (Optional)

Use cases (Optional)

Checklist

Uh oh!

codecov bot commented Aug 9, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Ezra-Yu left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

mzr1996 commented Aug 20, 2021

Uh oh!

Ezra-Yu left a comment

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

codecov bot commented Aug 9, 2021 •

edited

Loading