MaxViT-Large backbone (Tu et al., 2022).
Builds the canonical MaxViT-Large configuration:
depths=(2, 6, 14, 2), dims=(128, 256, 512, 1024).
209.7M parameters headless (211.8M with the classifier
head; paper Table 1 lists 212M).
Model Size
Parameters
pretrainedbool= FalseNo checkpoint is published for
maxvit_large, so True raises
NotImplementedError rather than return randomly initialised weights.
maxvit_large_cls is the checkpointed variant.**overridesobject= {}Keyword overrides on top of the canonical MaxViT-Large config.
Returns
MaxViTA MaxViT backbone returning a flat
feature.
Notes
MaxViT-Large reaches 85.2% top-1 on ImageNet-1k at 224x224 (Tu et al., 2022, Table 6).
Examples
>>> import lucid
>>> from lucid.models.vision.maxvit import maxvit_large
>>> model = maxvit_large()
>>> x = lucid.randn(1, 3, 224, 224)
>>> model.forward_features(x).shape
(1, 1024)