vgg_11_bn_cls(pretrained: bool | str = False, weights: VGG11BNWeights | None = None, overrides: object = {})VGG-11 with BatchNorm — image classifier.
Same topology as vgg_11_cls but with
lucid.nn.BatchNorm2d after each Conv + ReLU pair.
BatchNorm was added in the timm / reference-framework
reimplementations of VGG (not in the original paper) and
substantially improves convergence speed and final accuracy.
Approximately 132.9 M parameters; top-1 ImageNet validation
accuracy of 70.37% (torchvision eval).
Model Size
Parameters
pretrainedbool or str= FalsePretrained-weight selector.
False → random init; True
→ the DEFAULT tag (VGG11BNWeights.IMAGENET1K_V1); a
tag string → that specific checkpoint. Mutually exclusive with
weights (which wins if both are given).Explicit weights enum member. Takes precedence over
pretrained.**overridesobject= {}Keyword overrides forwarded into
VGGConfig.Returns
VGGForImageClassificationClassifier with the VGG-11-BN configuration applied (or with
overrides merged on top of it), optionally initialised from
pretrained weights.
Notes
Pretrained weights are converted from torchvision's
VGG11_BN_Weights and hosted on the Hugging Face Hub under
lucid-dl/vgg-11-bn.
Examples
>>> import lucid
>>> from lucid.models.vision.vgg import vgg_11_bn_cls
>>> model = vgg_11_bn_cls(pretrained=True)
>>> x = lucid.randn(1, 3, 224, 224)
>>> out = model(x)
>>> out.logits.shape
(1, 1000)