mobilenet_v1

→MobileNetV1

mobilenet_v1(pretrained: bool = False, overrides: object = {})

source edit

MobileNet-v1 backbone at width multiplier $\alpha = 1.0$ .

Builds a MobileNetV1 with the canonical paper topology: a 3×3 stem (stride 2) followed by 13 depthwise+pointwise blocks, yielding approximately 4.2M parameters. Howard et al., 2017 report a 70.6% ImageNet-1k top-1 validation accuracy with this configuration (Table 7). The default choice when the full accuracy budget is available.

Model Size

Parameters

pretrainedbool= False

Reserved for future pretrained-weight loading. Currently ignored — the returned model is randomly initialised.

**overridesobject= {}

Keyword overrides forwarded into MobileNetV1Config (e.g. in_channels=1 for grayscale input).

Returns

MobileNetV1

Backbone with the MobileNet-v1 ( $\alpha = 1.0$ ) configuration applied (or with overrides merged on top of it).

Notes

See Howard et al., "MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications", arXiv:1704.04861, 2017, Table 1 and Table 7.

Examples

>>> import lucid
>>> from lucid.models.vision.mobilenet import mobilenet_v1
>>> model = mobilenet_v1()
>>> x = lucid.randn(1, 3, 224, 224)
>>> out = model(x)
>>> out.last_hidden_state.shape
(1, 1024, 1, 1)

Used by 1

lucid.models.vision.mobilenet

>>> import lucid >>> from lucid.models.vision.mobilenet import mobilenet_v1 >>> model = mobilenet_v1() >>> x = lucid.randn(1, 3, 224, 224) >>> out = model(x) >>> out.last_hidden_state.shape (1, 1024, 1, 1)