maxvit_large_cls(pretrained: bool | str = False, weights: MaxViTLargeWeights | None = None, overrides: object = {})MaxViT-Large image classifier (Tu et al., 2022).
Combines the maxvit_large backbone (depths=(2, 6, 14, 2),
dims=(128, 256, 512, 1024)) with the reference NormMLP head.
~211.8M parameters.
Model Size
Parameters
pretrainedbool or str= FalsePretrained-weight selector.
False → random init; True
→ the DEFAULT tag (MaxViTLargeWeights.IN1K); a tag
string → that specific checkpoint. Mutually exclusive with
weights (which wins if both are given).Explicit weights enum member. Takes precedence over
pretrained.**overridesobject= {}Keyword overrides on top of the canonical MaxViT-Large config.
Returns
MaxViTForImageClassificationClassifier whose logits has shape (B, num_classes).
Notes
MaxViT-Large reaches 85.2% top-1 on ImageNet-1k at 224x224
(Tu et al., 2022, Table 6). Pretrained weights are converted from
timm's maxvit_large_tf_224.in1k and hosted under
lucid-dl/maxvit-large.
Examples
>>> import lucid
>>> from lucid.models.vision.maxvit import maxvit_large_cls
>>> model = maxvit_large_cls(num_classes=1000)
>>> x = lucid.randn(1, 3, 224, 224)
>>> model(x).logits.shape
(1, 1000)