mobilenet_v1_075

→MobileNetV1

mobilenet_v1_075(pretrained: bool = False, overrides: object = {})

source edit

MobileNet-v1 backbone at width multiplier $\alpha = 0.75$ .

Builds a MobileNetV1 with every channel count multiplied by 0.75 — approximately 2.6M parameters. Howard et al., 2017 report 68.4% ImageNet-1k top-1 accuracy with this configuration (Table 7), at roughly 60% of the FLOPs of the full-width model.

Model Size

Parameters

pretrainedbool= False

Reserved for future pretrained-weight loading. Currently ignored.

**overridesobject= {}

Keyword overrides forwarded into MobileNetV1Config.

Returns

MobileNetV1

Backbone with the MobileNet-v1 ( $\alpha = 0.75$ ) 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 7.

Examples

>>> import lucid
>>> from lucid.models.vision.mobilenet import mobilenet_v1_075
>>> model = mobilenet_v1_075()
>>> x = lucid.randn(1, 3, 224, 224)
>>> out = model(x)
>>> out.last_hidden_state.shape
(1, 768, 1, 1)

Used by 1

lucid.models.vision.mobilenet

>>> import lucid >>> from lucid.models.vision.mobilenet import mobilenet_v1_075 >>> model = mobilenet_v1_075() >>> x = lucid.randn(1, 3, 224, 224) >>> out = model(x) >>> out.last_hidden_state.shape (1, 768, 1, 1)