ConvNeXt-XLarge

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

Lucid port of timm/convnext_xlarge.fb_in22k_ft_in1k, converted to Lucid-native safetensors.

Available weights

Tag	acc@1	acc@5	Params	GFLOPs	Size	Source
FB_IN22K_FT_IN1K (default)	87.0	98.2	350.2M	—	1335.93 MB	timm

Usage

import lucid.models as models
from lucid.models.weights import ConvNeXtXLargeWeights

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

# explicit tag (enum or string)
model = models.convnext_xlarge_cls(weights=ConvNeXtXLargeWeights.FB_IN22K_FT_IN1K)
model = models.convnext_xlarge_cls(pretrained="FB_IN22K_FT_IN1K")

# preprocessing travels with the weights
weights = ConvNeXtXLargeWeights.FB_IN22K_FT_IN1K
preprocess = weights.transforms()
logits = model(preprocess(image)[None]).logits

Conversion

Converted from timm/convnext_xlarge.fb_in22k_ft_in1k via python -m tools.convert_weights convnext_xlarge --tag FB_IN22K_FT_IN1K. Key mapping + numerical parity verified against the source.

License

apache-2.0 — 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}
}