Pretrained weights for lucid.models.pvt_v2_b2_cls.
Single ImageNet-1k checkpoint converted from timm's
pvt_v2_b2.in1k (Wang et al., 2022; ~25.4M params, 82.0% top-1).
Examples
>>> from lucid.models.vision.pvt._weights import PVTv2B2Weights
>>> list(PVTv2B2Weights.__members__)
['IN1K', 'DEFAULT']
DEFAULT is an alias rather than a fourth entry, so a bare
pretrained=True and the tag it resolves to cannot drift apart.
>>> PVTv2B2Weights.DEFAULT is PVTv2B2Weights.IN1K
True
>>> PVTv2B2Weights.IN1K.num_classes
1000