mobilenet_v4_conv_small¶

lucid.models.mobilenet_v4_conv_small(num_classes: int = 1000, **kwargs) → MobileNet_V4¶

Overview¶

The mobilenet_v4_conv_small function returns an instance of the MobileNet_V4 model tailored for compact convolutional configurations. This variant is optimized for resource-constrained environments while delivering robust performance on classification tasks.

Total Parameters: 3,774,024

Function Signature¶

@register_model
def mobilenet_v4_conv_small(num_classes: int = 1000, **kwargs) -> MobileNet_V4

Parameters¶

num_classes (int, optional): Specifies the number of output classes. The default is 1000, commonly used for ImageNet classification.
**kwargs (dict): Additional keyword arguments to customize the model configuration. These may override the default settings used in the small convolutional variant.

Usage Example¶

>>> import lucid.models as models
>>> model = models.mobilenet_v4_conv_small(num_classes=1000)
>>> print(model)

Details¶

The mobilenet_v4_conv_small function instantiates a MobileNet-v4 model with a configuration optimized for efficiency in mobile and embedded systems. By emphasizing a compact convolutional design, it strikes a balance between low computational cost and high accuracy. This makes it an excellent choice for applications where resources are limited yet performance is critical.

Note

For scenarios requiring different architectural configurations, consider using the base MobileNet_V4 class with a custom configuration dictionary to better tailor the model to your specific requirements.