use torch.nn.functional.one_hot rewrite convert_to_one_hot

[HuDi2018]

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

Torch has the implementation of one-hot, here we can replace the one-hot implemented by MMCLS itself

Modification

as described above

BC-breaking (Optional)

Does the modification introduce changes that break the backward compatibility of the downstream repositories?
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

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.

[CLAassistant]

All committers have signed the CLA.

[Ezra-Yu]

Thank you for your contribution, please fix the lint refer to this guide before we review it.

[Ezra-Yu]

After installing pre-commit, the pre-commit does not work until you create a commit,

try to run

pre-commit run --all-files

twice in '$MMClassification' to check if there is any style error and fix it.

[HuDi2018]

cannot understand this error:

yapf.....................................................................Failed
- hook id: yapf
- files were modified by this hook

@Ezra-Yu

[codecov]

Codecov Report

Merging #696 (68f1542) into dev (43024cd) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##              dev     #696   +/-   ##
=======================================
  Coverage   85.03%   85.03%           
=======================================
  Files         121      121           
  Lines        7389     7392    +3     
  Branches     1277     1278    +1     
=======================================
+ Hits         6283     6286    +3     
  Misses        916      916           
  Partials      190      190           

Flag	Coverage Δ
unittests	85.03% <100.00%> (+<0.01%)	⬆️

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

Impacted Files	Coverage Δ
mmcls/models/losses/utils.py	56.25% <100.00%> (-1.33%)	⬇️
mmcls/models/backbones/resnet.py	100.00% <0.00%> (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 43024cd...68f1542. Read the comment docs.

[Ezra-Yu]

That means the pre-commit would modify the code automatically.

you can let the L117 in one line:

    one_hot_targets = F.one_hot(targets.long().squeeze(-1), num_classes=classes)

and run pre-commit run --all-files twice in command, the pre-commit would modify this line automatically.
it would be:

    one_hot_targets = F.one_hot(
        targets.long().squeeze(-1), num_classes=classes)

[HuDi2018]

That means the pre-commit would modify the code automatically.

you can let the L117 in one line:

    one_hot_targets = F.one_hot(targets.long().squeeze(-1), num_classes=classes)

and run pre-commit run --all-files twice in command, the pre-commit would modify this line automatically. it would be:

    one_hot_targets = F.one_hot(
        targets.long().squeeze(-1), num_classes=classes)

doesn't seem like a necessary change, currently I cannot install&use pre-commit successfully for some permission problem, so is there a work-around?

[Ezra-Yu]

me permission problem, s

OK, get it. We would modify it later when reviewing.

[HuDi2018]

me permission problem, s

OK, get it. We would modify it later when reviewing.

that's great help

[Ezra-Yu]

LGTM.