mobilenet_v3_small

lucid.models.mobilenet_v3_small(num_classes: int = 1000, **kwargs) → MobileNet_V3

The mobilenet_v3_small function constructs the MobileNet-v3 Small model. This preset uses the standard MobileNet-v3 Small bottleneck layout and last_channels=1024.

Total Parameters: 2,537,238

Function Signature

@register_model
def mobilenet_v3_small(num_classes: int = 1000, **kwargs) -> MobileNet_V3

Parameters

  • num_classes (int, optional): Number of output classes for classification. Default is 1000.

  • kwargs (dict, optional): Additional keyword arguments forwarded to MobileNetV3Config, excluding the preset bottleneck_cfg, last_channels, and stem_channels fields.

Returns

  • MobileNet_V3: A MobileNet-v3 Small model instance constructed from the default preset config.