8000 [Feature] Support ConvNeXt by hhaAndroid · Pull Request #7281 · open-mmlab/mmdetection · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

[Feature] Support ConvNeXt #7281

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

Merged
merged 30 commits into from
May 24, 2022
Merged

[Feature] Support ConvNeXt #7281

merged 30 commits into from
May 24, 2022

Conversation

hhaAndroid
Copy link
Collaborator
@hhaAndroid hhaAndroid commented Feb 28, 2022

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.

Motivation

Support ConvNeXt.

Modification

Please briefly describe what modification is made in this PR.

BC-breaking (Optional)

Does the modification introduce changes that break the backward-compatibility of the downstream repos?
If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR.

Use cases (Optional)

If this PR introduces a new feature, it is better to list some use cases here, and update the documentation.

Checklist

  1. Pre-commit or other linting tools are used to fix the potential lint issues.
  2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness.
  3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMCls.
  4. The documentation has been modified accordingly, like docstring or example tutorials.

@hhaAndroid hhaAndroid added the WIP Working in progress label Feb 28, 2022
@codecov
Copy link
codecov bot commented Feb 28, 2022

Codecov Report

Merging #7281 (d2ae0a2) into dev (a828499) will increase coverage by 0.05%.
The diff coverage is 80.95%.

@@            Coverage Diff             @@
##              dev    #7281      +/-   ##
==========================================
+ Coverage   64.88%   64.93%   +0.05%     
==========================================
  Files         352      355       +3     
  Lines       28504    28608     +104     
  Branches     4822     4849      +27     
==========================================
+ Hits        18494    18577      +83     
- Misses       9028     9042      +14     
- Partials      982      989       +7     
Flag Coverage Δ
unittests 64.91% <80.95%> (+0.06%) ⬆️

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

Impacted Files Coverage Δ
mmdet/core/optimizers/builder.py 33.33% <33.33%> (ø)
...re/optimizers/layer_decay_optimizer_constructor.py 90.24% <90.24%> (ø)
mmdet/apis/train.py 13.68% <100.00%> (ø)
mmdet/core/__init__.py 100.00% <100.00%> (ø)
mmdet/core/optimizers/__init__.py 100.00% <100.00%> (ø)
mmdet/utils/setup_env.py 88.00% <0.00%> (-4.00%) ⬇️
mmdet/datasets/builder.py 52.08% <0.00%> (-1.05%) ⬇️
mmdet/core/bbox/assigners/max_iou_assigner.py 73.68% <0.00%> (+1.31%) ⬆️

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 a828499...d2ae0a2. Read the comment docs.

@ZwwWayne
Copy link
Collaborator

Need to resolve conflicts.

@hhaAndroid hhaAndroid removed the WIP Working in progress label Apr 11, 2022
@hhaAndroid hhaAndroid requested a review from ZwwWayne April 26, 2022 07:54
@jbwang1997
Copy link
Collaborator

Other parts look good to me.

@rishabh279
Copy link

Hello @hhaAndroid, thanks a lot for working on this, may I know what are your plans to integrate and finalize this? Very much excited to use Convnext as backbone!

@hhaAndroid
Copy link
Collaborator Author

onvnext as backbone!

If you just want the backbone, you can refer to the configuration and use the mmcls directly. like this:

backbone=dict(
        _delete_=True,
        type='mmcls.ConvNeXt',
        arch='tiny',
        out_indices=[0, 1, 2, 3],
        drop_path_rate=0.4,
        layer_scale_init_value=1.0,
        gap_before_final_norm=False,
        init_cfg=dict(
            type='Pretrained', checkpoint=checkpoint_file,
            prefix='backbone.')),

@ZwwWayne
Copy link
Collaborator

Is there any progress?

@hhaAndroid hhaAndroid requested a review from jbwang1997 May 16, 2022 02:56
@MiXaiLL76
Copy link

any update?

@ZwwWayne ZwwWayne merged commit 1fd48f7 into open-mmlab:dev May 24, 2022
ZwwWayne pushed a commit that referenced this pull request Jul 18, 2022
* update

* update

* fix init_cfg

* update

* update

* update

* update

* update1

* final

* update

* update

* fix lint

* fix backbone config

* update cascade_mask_rcnn

* update and fix lint

* update

* fix DefaultOptimizerConstructor error

* update

* update

* update

* fix year

* update

* fix lint
ZwwWayne pushed a commit to ZwwWayne/mmdetection that referenced this pull request Jul 19, 2022
* update

* update

* fix init_cfg

* update

* update

* update

* update

* update1

* final

* update

* update

* fix lint

* fix backbone config

* update cascade_mask_rcnn

* update and fix lint

* update

* fix DefaultOptimizerConstructor error

* update

* update

* update

* fix year

* update

* fix lint
ZwwWayne pushed a commit to ZwwWayne/mmdetection that referenced this pull request Jul 19, 2022
* update

* update

* fix init_cfg

* update

* update

* update

* update

* update1

* final

* update

* update

* fix lint

* fix backbone config

* update cascade_mask_rcnn

* update and fix lint

* update

* fix DefaultOptimizerConstructor error

* update

* update

* update

* fix year

* update

* fix lint
@hhaAndroid hhaAndroid deleted the convnext branch July 19, 2022 07:03
SakiRinn pushed a commit to SakiRinn/mmdetection-locount that referenced this pull request Mar 17, 2023
* update

* update

* fix init_cfg

* update

* update

* update

* update

* update1

* final

* update

* update

* fix lint

* fix backbone config

* update cascade_mask_rcnn

* update and fix lint

* update

* fix DefaultOptimizerConstructor error

* update

* update

* update

* fix year

* update

* fix lint
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
Development

Successfully merging this pull request may close these issues.

5 participants
0