mobilenet_v4_conv_medium¶

lucid.models.mobilenet_v4_conv_medium(num_classes: int = 1000, **kwargs) → MobileNet_V4¶

Overview¶

The mobilenet_v4_conv_medium function provides an instance of the MobileNet_V4 model configured for a medium-scale convolutional variant. This version is designed to offer a balanced compromise between computational efficiency and performance, making it well-suited for devices with moderate resource constraints.

Total Parameters: 9,715,512

Function Signature¶

@register_model
def mobilenet_v4_conv_medium(num_classes: int = 1000, **kwargs) -> MobileNet_V4

Parameters¶

num_classes (int, optional): Specifies the number of output classes. The default is 1000, typically used for ImageNet classification.
**kwargs (dict): Additional keyword arguments that allow further customization of the model configuration. These can be used to override default settings and fine-tune the architecture for specific application requirements.

Usage Example¶

>>> import lucid.models as models
>>> model = models.mobilenet_v4_conv_medium(num_classes=1000)
>>> print(model)

Details¶

The mobilenet_v4_conv_medium function instantiates a MobileNet-v4 model variant optimized for a medium-scale convolutional configuration. This variant is particularly suitable for scenarios where a balance between computational cost and model accuracy is required.

Note

For scenarios requiring different performance or resource trade-offs, consider using the base MobileNet_V4 class with a custom configuration or exploring other registered MobileNet_V4 variants.