[Enhance] Merge EditVal/TestLoop and EditPreprocessor to dev-1.x

[LeoXing1996]

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

Merge refactored loops and preprocessor to dev-1.x.

Modification

1. EditDataSample is refactored in [Refactor] Refactor PackEditInputs and EditDataSample #1573
2. EditDatapreprocessor is refactored in [Enhancement] Refactor DataPreprocessor #1537
3. Metric and Loops are refactored in [Enhancement] Refactor metric and loop #1540
4. Support stack and split for EditDataSample [Enhancement] Support stack and split for EditDataSampe #1669
5. Revise the forward logic for GAN models for easy outstanding [Enhancement] Make forward logic more clear for GAN models #1670

Hightlights

1. GenValLoopandMultiValLoophas been merged toEditValLoop, GenTestLoopandMultiTestLoophas been merged toEditTestLoop`. Use case:

    Case 1: metrics on a single dataset

    >>> # add the following lines in your config
    >>> # 1. use `EditValLoop` instead of `ValLoop` in MMEngine
    >>> val_cfg = dict(type='EditValLoop')
    >>> # 2. specific EditEvaluator instead of Evaluator in MMEngine
    >>> val_evaluator = dict(
    >>>     type='EditEvaluator',
    >>>     metrics=[
    >>>         dict(type='PSNR', crop_border=2, prefix='Set5'),
    >>>         dict(type='SSIM', crop_border=2, prefix='Set5'),
    >>>     ])
    >>> # 3. define dataloader
    >>> val_dataloader = dict(...)

    Case 2: different metrics on different datasets

    >>> # add the following lines in your config
    >>> # 1. use `EditValLoop` instead of `ValLoop` in MMEngine
    >>> val_cfg = dict(type='EditValLoop')
    >>> # 2. specific a list EditEvaluator
    >>> # do not forget to add prefix for each metric group
    >>> div2k_evaluator = dict(
    >>>     type='EditEvaluator',
    >>>     metrics=dict(type='SSIM', crop_border=2, prefix='DIV2K'))
    >>> set5_evaluator = dict(
    >>>     type='EditEvaluator',
    >>>     metrics=[
    >>>         dict(type='PSNR', crop_border=2, prefix='Set5'),
    >>>         dict(type='SSIM', crop_border=2, prefix='Set5'),
    >>>     ])
    >>> # define evaluator config
    >>> val_evaluator = [div2k_evaluator, set5_evaluator]
    >>> # 3. specific a list dataloader for each metric groups
    >>> div2k_dataloader = dict(...)
    >>> set5_dataloader = dict(...)
    >>> # define dataloader config
    >>> val_dataloader = [div2k_dataloader, set5_dataloader]

2. Support stack and split for EditDataSample, Use case:

# Example for `split`
gen_sample = EditDataSample()
gen_sample.fake_img = outputs  # tensor
gen_sample.noise = noise  # tensor
gen_sample.sample_kwargs = deepcopy(sample_kwargs)  # dict
gen_sample.sample_model = sample_model  # string
# set allow_nonseq_value as True to copy non-sequential data (sample_kwargs and sample_model for this example)
batch_sample_list = gen_sample.split(allow_nonseq_value=True)  

# Example for `stack`
data_sample1 = EditDataSample()
data_sample1.set_gt_label(1)
data_sample1.set_tensor_data({'img': torch.randn(3, 4, 5)})
data_sample1.set_data({'mode': 'a'})
data_sample1.set_metainfo({
    'channel_order': 'rgb',
    'color_flag': 'color'
})
data_sample2 = EditDataSample()
data_sample2.set_gt_label(2)
data_sample2.set_tensor_data({'img': torch.randn(3, 4, 5)})
data_sample2.set_data({'mode': 'b'})
data_sample2.set_metainfo({
    'channel_order': 'rgb',
    'color_flag': 'color'
})
data_sample_merged = EditDataSample.stack([data_sample1, data_sample2])

Who can help? @ them here!

BC-breaking (Optional)

1. For BaseGAN/CondGAN models, real images are passed from data_samples.gt_img instead of inputs['img']
2. GenValLoop and MultiValLoop has been merged to EditValLoop, GenTestLoop and MultiTestLoop has been merged to EditTestLoop.
3. GenDataPreprocessor has been merged into EditDataPreprocessor,
   • No changes are required other than changing the type field in config.
   • Users do not need to define input_view and output_view since we will infer the shape of mean automatically.
   • In evaluation stage, all tensors will be converted to BGR (for three-channel images) and [0, 255].
4. PixelData has been removed

Use cases (Optional)

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

Checklist

Before PR:

• I have read and followed the workflow indicated in the CONTRIBUTING.md to create this PR.
• Pre-commit or linting tools indicated in CONTRIBUTING.md are used to fix the potential lint issues.
• Bug fixes are covered by unit tests, the case that causes the bug should be added in the unit tests.
• New functionalities are covered by complete unit tests. If not, please add more unit test to ensure the correctness.
• The documentation has been modified accordingly, including docstring or example tutorials.

After PR:

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

[codecov]

Codecov Report

Patch coverage: 90.42% and project coverage change: -0.43 ⚠️

Comparison is base (a6d75e9) 88.22% compared to head (2b1bbdc) 87.80%.

❗ Current head 2b1bbdc differs from pull request most recent head c504704. Consider uploading reports for the commit c504704 to get more accurate results

Additional details and impacted files

@@             Coverage Diff             @@
##           dev-1.x    #1656      +/-   ##
===========================================
- Coverage    88.22%   87.80%   -0.43%     
===========================================
  Files          401      399       -2     
  Lines        26383    26368      -15     
  Branches      4083     4114      +31     
===========================================
- Hits         23276    23152     -124     
- Misses        2204     2329     +125     
+ Partials       903      887      -16     

Flag	Coverage Δ
unittests	87.80% <90.42%> (-0.43%)	⬇️

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

Impacted Files	Coverage Δ
mmedit/datasets/transforms/aug_pixel.py	96.38% <ø> (ø)
mmedit/datasets/transforms/crop.py	85.38% <ø> (ø)
mmedit/engine/hooks/pggan_fetch_data_hook.py	95.74% <ø> (ø)
mmedit/evaluation/metrics/base_gen_metric.py	88.88% <ø> (ø)
mmedit/evaluation/metrics/equivariance.py	80.89% <ø> (ø)
mmedit/models/editors/biggan/biggan_generator.py	83.67% <0.00%> (-1.75%)	⬇️
...edit/models/editors/guided_diffusion/classifier.py	17.06% <ø> (ø)
mmedit/models/editors/stylegan1/stylegan1.py	90.24% <ø> (ø)
...edit/models/editors/stylegan2/stylegan2_modules.py	87.93% <0.00%> (ø)
mmedit/utils/io_utils.py	22.64% <ø> (ø)
... and 98 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[zengyh1900]

This PR involves significant changes.
Please list as detailed as possible use_case, migration notes, etc.

[zengyh1900]

Hi @LeoXing1996 feel free to discuss any comments. If I misunderstand some usage, please correct me.

[zengyh1900]

lgtm