efficientnet_b1

→EfficientNet

efficientnet_b1(pretrained: bool = False, overrides: object = {})

source edit

EfficientNet-B1 feature-extracting backbone (compound coefficient $\phi = 0.5$ ).

Builds an EfficientNet with depth multiplier 1.1 and width multiplier 1.0 — same channel counts as B0 but with block repeats scaled up by ceil(n · 1.1). Approximately 7.8M parameters and 79.1% ImageNet-1k top-1 accuracy at the 240×240 native resolution (Tan & Le, 2019, Table 2).

Model Size

Parameters

pretrainedbool= False

Reserved for future pretrained-weight loading. Currently ignored.

**overridesobject= {}

Keyword overrides forwarded into EfficientNetConfig.

Returns

EfficientNet

Backbone with the B1 configuration applied (or with overrides merged on top of it).

Notes

See Tan & Le, "EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks", ICML 2019 (arXiv:1905.11946), Table 2.

Examples

>>> import lucid
>>> from lucid.models.vision.efficientnet import efficientnet_b1
>>> model = efficientnet_b1()
>>> x = lucid.randn(1, 3, 240, 240)
>>> out = model(x)
>>> out.last_hidden_state.shape
(1, 1280, 1, 1)

Used by 1

lucid.models.vision.efficientnet

>>> import lucid >>> from lucid.models.vision.efficientnet import efficientnet_b1 >>> model = efficientnet_b1() >>> x = lucid.randn(1, 3, 240, 240) >>> out = model(x) >>> out.last_hidden_state.shape (1, 1280, 1, 1)