data
InceptionResNetOutput
InceptionResNetOutput(logits: Tensor, loss: Tensor | None = None)Structured forward output for InceptionResNetV2ForImageClassification.
Inception-ResNet v2 does not use auxiliary classifiers (unlike Inception v3 / GoogLeNet) — the residual shortcuts already mitigate the vanishing-gradient problem that auxiliary heads were originally introduced to address — so this dataclass only carries the main logits and an optional loss tensor.
Parameters
Examples
>>> import lucid
>>> from lucid.models.vision.inception_resnet import inception_resnet_v2_cls
>>> model = inception_resnet_v2_cls().eval()
>>> x = lucid.randn(1, 3, 299, 299)
>>> out = model(x)
>>> out.logits.shape
(1, 1000)
>>> out.loss is None
True