convnext_v2_huge

lucid.models.convnext_v2_huge(num_classes: int = 1000, **kwargs) → ConvNeXt_V2

The convnext_v2_huge creates a ConvNeXt-v2 variant with a maximal configuration, optimized for state-of-the-art performance on large-scale and high-complexity image classification tasks. This model is designed for applications requiring abundant computational resources and top-tier accuracy.

Total Parameters: 660,289,640

Function Signature

@register_model
def convnext_v2_huge(num_classes: int = 1000, **kwargs) -> ConvNeXt_V2

Parameters

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

  • kwargs: Additional keyword arguments for customizing the ConvNeXt-v2 architecture.

Returns

  • model (ConvNeXt_V2): An instance of the ConvNeXt-v2 class configured as the huge variant.

Examples

Basic Example

import lucid.models as models

# Create convnext_v2_huge with default 21841 classes
model = models.convnext_v2_huge(num_classes=21841)

# Input tensor with shape (1, 3, 224, 224)
input_ = lucid.random.randn(1, 3, 224, 224)

# Perform forward pass
output = model(input_)

print(output.shape)  # Shape: (1, 21841)