[Feature] Support EfficientFormer

[Ezra-Yu]

Motivation

Support EfficientFormer, Arxiv | Github

Use cases (Optional)

    Example:
        >>> from mmcls.models import EfficientFormer
        >>> import torch
        >>> inputs = torch.rand((1, 3, 224, 224))
        >>> # build EfficientFormer backbone for classification task
        >>> model = EfficientFormer(arch="l1")
        >>> model.eval()
        >>> level_outputs = model(inputs)
        >>> for level_out in level_outputs:
        ...     print(tuple(level_out.shape))
        (1, 448, 49)
        >>> # build EfficientFormer backbone for downstream task
        >>> EfficientFormer(
        >>>    arch="l3",
        >>>    out_indices=(0, 1, 2, 3),
        >>>    reshape_last_feat=True)
        >>> model.eval()
        >>> level_outputs = model(inputs)
        >>> for level_out in level_outputs:
        ...     print(tuple(level_out.shape))

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 #954 (4dcea5a) into dev (b5bb86a) will increase coverage by 0.26%.
The diff coverage is 92.53%.

@@            Coverage Diff             @@
##              dev     #954      +/-   ##
==========================================
+ Coverage   85.86%   86.13%   +0.26%     
==========================================
  Files         136      140       +4     
  Lines        9306     9674     +368     
  Branches     1611     1677      +66     
==========================================
+ Hits         7991     8333     +342     
- Misses       1076     1090      +14     
- Partials      239      251      +12     

Flag	Coverage Δ
unittests	86.06% <92.53%> (+0.27%)	⬆️

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

Impacted Files	Coverage Δ
mmcls/datasets/voc.py	31.81% <28.57%> (-1.52%)	⬇️
mmcls/models/heads/efficientformer_head.py	92.10% <92.10%> (ø)
mmcls/models/heads/multi_label_csra_head.py	92.15% <92.15%> (ø)
mmcls/models/backbones/efficientformer.py	94.49% <94.49%> (ø)
mmcls/models/backbones/__init__.py	100.00% <100.00%> (ø)
mmcls/models/heads/__init__.py	100.00% <100.00%> (ø)
mmcls/datasets/__init__.py	100.00% <0.00%> (ø)
mmcls/models/backbones/regnet.py	98.05% <0.00%> (ø)
mmcls/datasets/stanford_cars.py	94.11% <0.00%> (ø)

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

[mzr1996]

x.dim() == 3? why do we need this condition?

[Ezra-Yu]

in case the vit-num is 0 in arch.

[mzr1996]

LGTM