efficientformer_l7

→EfficientFormer

efficientformer_l7(pretrained: bool = False, overrides: object = {})

source edit

EfficientFormer-L7 backbone (Li et al., 2022).

Builds the canonical EfficientFormer-L7 configuration: depths=(6, 6, 18, 8), embed_dims=(96, 192, 384, 768), drop_path_rate=0.2. Approximately 81.5M parameters — the largest variant in the paper.

Model Size

Parameters

pretrainedbool= False

If True, loads ImageNet-1k pretrained weights when available. Defaults to False.

**overridesobject= {}

Keyword overrides on top of the canonical L7 config.

Returns

EfficientFormer

An EfficientFormer backbone returning a flat $(B, 768)$ feature.

Notes

EfficientFormer-L7 reaches 83.3% top-1 on ImageNet-1k (Li et al., 2022, Table 4) — the headline result of the paper.

Examples

>>> import lucid
>>> from lucid.models.vision.efficientformer import efficientformer_l7
>>> model = efficientformer_l7()
>>> x = lucid.randn(1, 3, 224, 224)
>>> model.forward_features(x).shape
(1, 768)

Used by 1

lucid.models.vision.efficientformer

>>> import lucid >>> from lucid.models.vision.efficientformer import efficientformer_l7 >>> model = efficientformer_l7() >>> x = lucid.randn(1, 3, 224, 224) >>> model.forward_features(x).shape (1, 768)