convnext_xlarge

lucid.models.convnext_xlarge(num_classes: int = 1000, **kwargs) → ConvNeXt

The convnext_xlarge function constructs the ConvNeXt-XLarge preset. This preset uses the default ConvNeXtConfig stage layout for the xlarge variant.

Total Parameters: 350,196,968

Function Signature

@register_model
def convnext_xlarge(num_classes: int = 1000, **kwargs) -> ConvNeXt

Parameters

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

  • kwargs (dict, optional): Additional keyword arguments forwarded to ConvNeXtConfig, excluding the preset depths and dims fields.

Returns

  • ConvNeXt: A ConvNeXt model instance constructed from the xlarge preset config.