Views
No views yet
Hu et al., 2018 — Squeeze-and-Excitation Networks (arXiv:1709.01507)
timm/legacy_seresnet101.in1k,
converted to Lucid-native safetensors.| Tag | acc@1 | acc@5 | Params | GFLOPs | Size | Source |
|---|---|---|---|---|---|---|
IN1K (default) | 78.32 | — | 49.3M | — | 188.64 MB | timm |
1import lucid.models as models
2from lucid.models.weights import SEResNet101Weights
3
4# default tag
5model = models.se_resnet_101_cls(pretrained=True)
6
7# explicit tag (enum or string)
8model = models.se_resnet_101_cls(weights=SEResNet101Weights.IN1K)
9model = models.se_resnet_101_cls(pretrained="IN1K")
10
11# preprocessing travels with the weights
12weights = SEResNet101Weights.IN1K
13preprocess = weights.transforms()
14logits = model(preprocess(image)[None]).logitstimm/legacy_seresnet101.in1k via
python -m tools.convert_weights se_resnet_101 --tag IN1K.
Key mapping + numerical parity verified against the source.apache-2.0 — inherited from the original weights.@inproceedings{hu2018squeeze,
title={Squeeze-and-Excitation Networks},
author={Hu, Jie and Shen, Li and Sun, Gang},
booktitle={CVPR}, year={2018}
}