class

InceptionV3Weights

extendsWeightsEnum

InceptionV3Weights()

Pretrained weights for lucid.models.inception_v3_cls.

Szegedy et al. CVPR 2016 (23.8 M params, top-1 77.29% on ImageNet-1k).

IMAGENET1K_V1WeightEntry

ImageNet-1k V1 checkpoint (top-1 77.294% / top-5 93.450%), sourced from torchvision/Inception_V3_Weights.IMAGENET1K_V1.

DEFAULTWeightEntry

Alias for IMAGENET1K_V1.

Reference: Szegedy, Vanhoucke, Ioffe, Shlens, Wojna, "Rethinking the Inception Architecture for Computer Vision", CVPR 2016 (arXiv:1512.00567).

>>> from lucid.models import inception_v3_cls
>>> model = inception_v3_cls(pretrained=True).eval()