ConvNeXt-Large

Liu et al., 2022 — A ConvNet for the 2020s (arXiv:2201.03545)

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

Available weights

Tag	acc@1	acc@5	Params	GFLOPs	Size	Source
IMAGENET1K_V1 (default)	84.414	96.976	197.8M	34.361	754.45 MB	torchvision

Usage

import lucid.models as models
from lucid.models.weights import ConvNeXtLargeWeights

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

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

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

Conversion

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

License

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

Citation

@inproceedings{liu2022convnet,
  title={A ConvNet for the 2020s},
  author={Liu, Zhuang and Mao, Hanzi and Wu, Chao-Yuan and Feichtenhofer, Christoph and Darrell, Trevor and Xie, Saining},
  booktitle={CVPR}, year={2022}
}