[Feature] Support CUB dataset

[okotaku]

Motivation

Support CUB dataset.

Augmentation and optimizer is based on A Novel Plug-in Module for Fine-Grained Visual Classification.

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.

[codecov]

Codecov Report

Merging #703 (013e987) into dev (c1534f9) will increase coverage by 0.05%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##              dev     #703      +/-   ##
==========================================
+ Coverage   84.98%   85.04%   +0.05%     
==========================================
  Files         122      123       +1     
  Lines        7573     7603      +30     
  Branches     1304     1311       +7     
==========================================
+ Hits         6436     6466      +30     
  Misses        945      945              
  Partials      192      192              

Flag	Coverage Δ
unittests	84.96% <100.00%> (+0.05%)	⬆️

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

Impacted Files	Coverage Δ
mmcls/datasets/__init__.py	100.00% <100.00%> (ø)
mmcls/datasets/cub.py	100.00% <100.00%> (ø)

---

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 c1534f9...013e987. Read the comment docs.

[Ezra-Yu]

Thank you for your awesome contribution, I will run some experiments to get several checkpoints and logs. That would take a little bit of time.

[Ezra-Yu]

I have reproduced the baseline r50(88.2) and swin-l(91.9) result in A Novel Plug-in Module for Fine-Grained Visual Classification.

Could I commit the configs and doc to your branch directly?

[okotaku]

@Ezra-Yu Sure.

[Ezra-Yu]

I can not commit to your branch directly, maybe you should check the Allow edits and access to secrets by maintainers.

for example:

[okotaku]

@Ezra-Yu I could not run Allow edits and access to secrets by maintainers because it was not forked with my personal account. Instead, I gave you write permission and invited as collaborators. Please make sure this is okay.

[Ezra-Yu]

@Ezra-Yu I could not run Allow edits and access to secrets by maintainers because it was not forked with my personal account. Instead, I gave you write permission and invited as collaborators. Please make sure this is okay.

It works.

[Ezra-Yu]

LGTM.

[okotaku]

@Ezra-Yu

I noticed that the following url for the weight is NOT FOUND.
https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_mill_3rdparty_in21k_20220307-bdb3a68b.pth

[Ezra-Yu]

@okotaku Yes, the weights and logs have not been uploaded to the server, I will urge this to be done.

[okotaku]

@Ezra-Yu @mzr1996 This PR was released in v0.22.0, but weights and logs do not yet exist.

[mzr1996]

@Ezra-Yu @mzr1996 This PR was released in v0.22.0, but weights and logs do not yet exist.

My bad, I will upload it soon.

[mzr1996]

@Ezra-Yu @mzr1996 This PR was released in v0.22.0, but weights and logs do not yet exist.

Done.