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.

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

Jump to bottom

[Feature] Add configs for AOT-GAN #681

Merged
merged 20 commits into from
May 30, 2022
Merged

[Feature] Add configs for AOT-GAN #681

merged 20 commits into from
May 30, 2022

Conversation

quincylin1
Copy link
Contributor

added aotgan config

@quincylin1 quincylin1 changed the title Aot gan configs [Feature] Add configs for AOT-GAN Dec 21, 2021
@codecov
Copy link

codecov bot commented Dec 21, 2021 •
edited
Loading

Codecov Report

Merging #681 (3bccc5b) into master (b77cc3e) will increase coverage by 0.01%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #681      +/-   ##
==========================================
+ Coverage   83.50%   83.51%   +0.01%     
==========================================
  Files         222      222              
  Lines       12813    12813              
  Branches     2094     2094              
==========================================
+ Hits        10699    10701       +2     
+ Misses       1784     1783       -1     
+ Partials      330      329       -1
Flag Coverage Δ
unittests 83.47% <100.00%> (+<0.01%) ⬆️

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

Impacted Files Coverage Δ
mmedit/models/inpaintors/aot_inpaintor.py 89.62% <100.00%> (ø)
...ls/components/stylegan2/generator_discriminator.py 86.14% <0.00%> (+1.20%) ⬆️

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 b77cc3e...3bccc5b. Read the comment docs.

ckkelvinchan
ckkelvinchan reviewed Mar 31, 2022
View reviewed changes
Copy link
Member

@ckkelvinchan ckkelvinchan left a comment

Choose a reason for hiding this comment

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

Hi @quincylin1 , would you mind also providing the weights converted from the official weights?

configs/inpainting/AOT-GAN/AOT-GAN_512x512_places.py Outdated
load_from = None
resume_from = None
workflow = [('train', 10000)]
exp_name = 'AOT-GAN_512x512_places'
Copy link
Member

Choose a reason for hiding this comment

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

We have a naming convention for the models. Would you mind use the convention below?

{model}_[model setting]_{backbone}_[refiner]_[norm setting]_[misc]_[gpu x batch_per_gpu]_{schedule}_{dataset}

An example can also be found here

Copy link
Member

Choose a reason for hiding this comment

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

Hi @quincylin1 , would you mind also providing the weights converted from the official weights?

The checkpoint can be named as follows:

{config_name}_{date}-{hash}.pth

For the hash, you can use tools/publish_model.py to generate.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

We have a naming convention for the models. Would you mind use the convention below?

{model}_[model setting]_{backbone}_[refiner]_[norm setting]_[misc]_[gpu x batch_per_gpu]_{schedule}_{dataset}

An example can also be found here

image

@ckkelvinchan just wondering if this name is ok for this config: AOT-GAN_512x512_4x12_places.py

Copy link
Member

Choose a reason for hiding this comment

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

Yeah I think it works.

@wangruohui wangruohui added this to the v0.15.0 milestone Apr 1, 2022
ckkelvinchan
ckkelvinchan reviewed Apr 3, 2022
View reviewed changes
tests/test_models/test_backbones/test_vfi_backbones/test_tof_vfi_net.py Outdated
assert torch.is_tensor(output)
assert output.shape == (1, 3, 256, 256)
Copy link
Member

Choose a reason for hiding this comment

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

Is it a bug from previous PR? @Yshuo-Li

Copy link
Collaborator

Choose a reason for hiding this comment

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

Yes, it is.

An error in GPU unit test of TOFlow.

Is it a bug from previous PR? @Yshuo-Li

@ckkelvinchan
Copy link
Member

By the way, can @wangruohui or @Yshuo-Li help with the checkpoint uploading?

@quincylin1 Are you able to provide us with the

  1. converted official weights
  2. weights trained with your code and the corresponding log?

Thanks!

@quincylin1
Copy link
Contributor Author

Hi @quincylin1 , would you mind also providing the weights converted from the official weights?

Hi @ckkelvinchan, slightly confused here, what do you mean by weights converted from the official weights?

@ckkelvinchan
Copy link
Member

Hi @quincylin1 , would you mind also providing the weights converted from the official weights?

Hi @ckkelvinchan, slightly confused here, what do you mean by weights converted from the official weights?

I remembered you have tried loading the official weights to your model and obtained same performance?

@wangruohui
Copy link
Member

Hello @quincylin1 ,

Any updates on the model weights?

Yshuo-Li
Yshuo-Li reviewed May 26, 2022
View reviewed changes
configs/inpainting/AOT-GAN/metafile.yml Outdated
Name: AOT-GAN_512x512_4x12_places
Results:
- Dataset: PLACES
Metrics: {}
Copy link
Collaborator

Choose a reason for hiding this comment

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

Please check it again.

Copy link
Member

Choose a reason for hiding this comment

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

Fixed

@wangruohui wangruohui merged commit f4bfe94 into open-mmlab:master May 30, 2022
wangruohui added a commit to wangruohui/mmediting that referenced this pull request Jul 7, 2022
@quincylin1 @wangruohui
[Feature] Add configs and readme for AOT-GAN (open-mmlab#681)
9b7dbe0 
* add aot config

* fixed aot config

* modified colorjitter in config

* rename config and fix unittest

* renamed exp

* fix conflicts

* fix eval_result

* add aot gan raedme

* fix

* add detailed table

* add chinese readme

* fix

* add aot-gan in readme

* add results

* add results

* Update test_losses.py

Co-authored-by: wangruohui <12756472+wangruohui@users.noreply.github.com>
wangruohui added a commit to wangruohui/mmediting that referenced this pull request Jul 7, 2022
@quincylin1 @wangruohui
[Feature] Add configs and readme for AOT-GAN (open-mmlab#681)
b26d2de 
* add aot config

* fixed aot config

* modified colorjitter in config

* rename config and fix unittest

* renamed exp

* fix conflicts

* fix eval_result

* add aot gan raedme

* fix

* add detailed table

* add chinese readme

* fix

* add aot-gan in readme

* add results

* add results

* Update test_losses.py

Co-authored-by: wangruohui <12756472+wangruohui@users.noreply.github.com>
Yshuo-Li pushed a commit to Yshuo-Li/mmediting that referenced this pull request Jul 15, 2022
@quincylin1 @wangruohui
[Feature] Add configs and readme for AOT-GAN (open-mmlab#681)
1014a71 
* add aot config

* fixed aot config

* modified colorjitter in config

* rename config and fix unittest

* renamed exp

* fix conflicts

* fix eval_result

* add aot gan raedme

* fix

* add detailed table

* add chinese readme

* fix

* add aot-gan in readme

* add results

* add results

* Update test_losses.py

Co-authored-by: wangruohui <12756472+wangruohui@users.noreply.github.com>
Yshuo-Li pushed a commit to Yshuo-Li/mmediting that referenced this pull request Jul 15, 2022
@quincylin1 @wangruohui
[Feature] Add configs and readme for AOT-GAN (open-mmlab#681)
53f5c50 
* add aot config

* fixed aot config

* modified colorjitter in config

* rename config and fix unittest

* renamed exp

* fix conflicts

* fix eval_result

* add aot gan raedme

* fix

* add detailed table

* add chinese readme

* fix

* add aot-gan in readme

* add results

* add results

* Update test_losses.py

Co-authored-by: wangruohui <12756472+wangruohui@users.noreply.github.com>
Yshuo-Li pushed a commit to Yshuo-Li/mmediting that referenced this pull request Jul 16, 2022
@quincylin1 @wangruohui
[Feature] Add configs and readme for AOT-GAN (open-mmlab#681)
ff61c9e 
* add aot config

* fixed aot config

* modified colorjitter in config

* rename config and fix unittest

* renamed exp

* fix conflicts

* fix eval_result

* add aot gan raedme

* fix

* add detailed table

* add chinese readme

* fix

* add aot-gan in readme

* add results

* add results

* Update test_losses.py

Co-authored-by: wangruohui <12756472+wangruohui@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@wangruohui wangruohui wangruohui left review comments

@ckkelvinchan ckkelvinchan ckkelvinchan left review comments

@Yshuo-Li Yshuo-Li Yshuo-Li approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
v0.15.0
Development

Successfully merging this pull request may close these issues.
4 participants
@quincylin1 @ckkelvinchan @wangruohui @Yshuo-Li