Inception-v4

Szegedy et al., 2017 — Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning (arXiv:1602.07261)

Lucid port of timm/inception_v4.tf_in1k, converted to Lucid-native safetensors.

Available weights

Tag	acc@1	acc@5	Params	GFLOPs	Size	Source
TF_IN1K (default)	80.144	94.982	42.7M	—	163.15 MB	timm

Usage

import lucid.models as models
from lucid.models.weights import InceptionV4Weights

# default tag
model = models.inception_v4_cls(pretrained=True)

# explicit tag (enum or string)
model = models.inception_v4_cls(weights=InceptionV4Weights.TF_IN1K)
model = models.inception_v4_cls(pretrained="TF_IN1K")

# preprocessing travels with the weights
weights = InceptionV4Weights.TF_IN1K
preprocess = weights.transforms()
out = model(preprocess(image)[None])
logits = out.logits  # (B, num_classes)

Conversion

Converted from timm/inception_v4.tf_in1k via python -m tools.convert_weights inception_v4 --tag TF_IN1K. Key set, tensor shapes and a strict load verified against a freshly built Lucid model.

License

apache-2.0 — inherited from the original weights.

Citation

@inproceedings{szegedy2017inception,
  title={Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning},
  author={Szegedy, Christian and Ioffe, Sergey and Vanhoucke, Vincent and Alemi, Alexander A.},
  booktitle={AAAI}, year={2017}
}