[Feature] add eva02 backbone

[zzc98]

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

Add the EVA02 backbone.
paper: https://arxiv.org/abs/2303.11331
code: https://github.com/baaivision/EVA/tree/master/EVA-02

Modification

• Add new backbone network EVA02
• Complete the network implementation and configuration files
• Complete the weight conversion according to the official repository
• Align the accuracy on the ImageNet1k validation set

Use cases

model = dict(
    type='ImageClassifier',
    backbone=dict(
        type='EVA02',
        arch='base',
        img_size=448,
        sub_ln=True,
        out_type='avg_featmap'),
    neck=None,
    head=dict(
        type='LinearClsHead',
        num_classes=1000,
        in_channels=768,
        loss=dict(
            type='LabelSmoothLoss', label_smooth_val=0.1, mode='original'),
    ),
    init_cfg=[
        dict(type='TruncNormal', layer='Linear', std=.02),
        dict(type='Constant', layer='LayerNorm', val=1., bias=0.),
    ],
    train_cfg=dict(augments=[
        dict(type='Mixup', alpha=0.8),
        dict(type='CutMix', alpha=1.0)
    ])
)

More details can be found in configs/eva02/README.md.

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

Patch coverage has no change and project coverage change: +0.91 🎉

Comparison is base (c9a0cb0) 84.37% compared to head (9fe90e6) 85.28%.

❗ Current head 9fe90e6 differs from pull request most recent head c481819. Consider uploading reports for the commit c481819 to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##              dev    #1450      +/-   ##
==========================================
+ Coverage   84.37%   85.28%   +0.91%     
==========================================
  Files         142      240      +98     
  Lines        9925    18095    +8170     
  Branches     1621     2814    +1193     
==========================================
+ Hits         8374    15433    +7059     
- Misses       1277     2138     +861     
- Partials      274      524     +250     

Flag	Coverage Δ
unittests	85.28% <ø> (+0.91%)	⬆️

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

see 382 files with indirect coverage changes

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

[Ezra-Yu]

please rm _base_.models.eva folder and put the models cfg into every file of config/eva02/eva02_xxxx.py.

[Ezra-Yu]

LGTM.

[fangyixiao18]

also check config names