Pretrained weights for lucid.models.vit_base_16_cls.
Single ImageNet-1k checkpoint converted from reference_vision's
ViT_B_16_Weights.IMAGENET1K_V1 (Dosovitskiy et al., 2021;
~86.6M params, 81.072% top-1).
Examples
>>> from lucid.models.vision.vit._weights import ViTBase16Weights
>>> list(ViTBase16Weights.__members__)
['IMAGENET1K_V1', 'DEFAULT']
DEFAULT is an alias rather than a fourth entry, so a bare
pretrained=True and the tag it resolves to cannot drift apart.
>>> ViTBase16Weights.DEFAULT is ViTBase16Weights.IMAGENET1K_V1
True
>>> ViTBase16Weights.IMAGENET1K_V1.num_classes
1000