Pretrained weights for lucid.models.se_resnet_101_cls.
SE-ResNet-101 (Hu et al. CVPR 2018; 49.3 M params, top-1 78.32%).
Attributes
IN1KWeightEntryImageNet-1k checkpoint (top-1 78.32%), sourced from
timm/legacy_seresnet101.in1k.DEFAULTWeightEntryAlias for
IN1K.Notes
Reference: Hu, Shen, Sun, "Squeeze-and-Excitation Networks", CVPR 2018 (arXiv:1709.01507).
Examples
>>> from lucid.models import se_resnet_101_cls
>>> model = se_resnet_101_cls(pretrained=True).eval()