swin_v2_giant¶
- lucid.models.swin_v2_giant(img_size: int = 224, num_classes: int = 1000, **kwargs) SwinTransformer_V2¶
The swin_v2_giant function initializes the largest variant of the Swin Transformer V2 model, optimized for extremely high-performance vision tasks such as large-scale image classification, object detection, and segmentation. This model is designed to handle complex data distributions by leveraging deeper layers, a higher number of attention heads, and more efficient normalization.
Total Parameters: 3,000,869,564
Function Signature¶
@register_model
def swin_v2_giant(img_size: int = 224, num_classes: int = 1000, **kwargs) -> SwinTransformer_V2
Parameters¶
img_size (int, optional): The size of the input image (assumes square images). Default is 224.
num_classes (int, optional): The number of output classes for classification. Default is 1000.
kwargs (dict, optional): Additional parameters for customization, including:
depths (list[int]): A list specifying the number of transformer blocks in each stage. The typical default configuration is [2, 2, 42, 4], indicating that the model has 4 stages with 2, 2, 18, and 2 blocks respectively.
num_heads (list[int]): A list specifying the number of attention heads in each stage. The common default is [6, 12, 24, 48], allowing the model to efficiently capture global dependencies and perform detailed feature extraction.
Returns¶
SwinTransformer_V2: An instance of the SwinTransformer_V2 class configured as an ultra-large vision transformer.
Examples¶
>>> import lucid.models as models
>>> model = models.swin_v2_giant()
>>> print(model)
SwinTransformer_V2(img_size=224, num_classes=1000, ...)