AlexNet

Krizhevsky et al., 2012 — ImageNet Classification with Deep Convolutional Neural Networks (NIPS); single-stream channel widths from Krizhevsky 2014 (arXiv:1404.5997).

Lucid port of torchvision/AlexNet_Weights.IMAGENET1K_V1, converted to Lucid-native safetensors.

Available weights

Tag	acc@1	acc@5	Params	GFLOPs	Size	Source
IMAGENET1K_V1 (default)	56.522	79.066	61.1M	0.714	233.08 MB	torchvision

Usage

import lucid.models as models
from lucid.models.weights import AlexNetWeights

# default tag
model = models.alexnet_cls(pretrained=True)

# explicit tag (enum or string)
model = models.alexnet_cls(weights=AlexNetWeights.IMAGENET1K_V1)
model = models.alexnet_cls(pretrained="IMAGENET1K_V1")

# preprocessing travels with the weights
weights = AlexNetWeights.IMAGENET1K_V1
preprocess = weights.transforms()
logits = model(preprocess(image)[None]).logits

Conversion

Converted from torchvision/AlexNet_Weights.IMAGENET1K_V1 via python -m tools.convert_weights alexnet_cls --tag IMAGENET1K_V1. Key mapping + numerical parity verified against the source.

License

bsd-3-clause — inherited from the original weights.

Citation

@inproceedings{krizhevsky2012imagenet,
  title={ImageNet Classification with Deep Convolutional Neural Networks},
  author={Krizhevsky, Alex and Sutskever, Ilya and Hinton, Geoffrey E.},
  booktitle={NIPS}, year={2012}
}
@article{krizhevsky2014oneweirdtrick,
  title={One weird trick for parallelizing convolutional neural networks},
  author={Krizhevsky, Alex},
  journal={arXiv preprint arXiv:1404.5997}, year={2014}
}