Pretrained weight tags for lucid.models.resnet_18_cls.
Ships a single ImageNet-1k checkpoint (IMAGENET1K_V1) — the
canonical V1 weights distributed by the reference-framework model
zoo, re-hosted under huggingface.co/lucid-dl/resnet-18 with the
official acc@1=69.758 / acc@5=89.078 validation metrics.
Attributes
IMAGENET1K_V1WeightEntryImageNet-1k V1 checkpoint (top-1 69.758% / top-5 89.078%),
sourced from
torchvision/ResNet18_Weights.IMAGENET1K_V1.DEFAULTWeightEntryAlias for
IMAGENET1K_V1.Notes
Reference: He, Zhang, Ren, Sun, "Deep Residual Learning for Image Recognition", CVPR 2016 (arXiv:1512.03385).
Examples
>>> from lucid.models import resnet_18_cls
>>> model = resnet_18_cls(pretrained=True).eval()