Structured forward output for InceptionV4ForImageClassification.
Inception-v4 trains with a single softmax head — the paper drops the auxiliary classifiers of GoogLeNet / Inception v3 — so this dataclass carries only the logits and an optional loss.
Parameters
Examples
>>> import lucid
>>> from lucid.models.vision.inception_v4 import inception_v4_cls
>>> model = inception_v4_cls().eval()
>>> out = model(lucid.randn(1, 3, 299, 299))
>>> out.logits.shape
(1, 1000)
>>> out.loss is None
True