swin_v2_tiny¶

lucid.models.swin_v2_tiny(img_size: int = 224, num_classes: int = 1000, **kwargs) → SwinTransformer_V2¶

The swin_v2_tiny function instantiates a small Swin Transformer V2 model with a predefined architecture. This model builds upon the original Swin Transformer by introducing improved normalization techniques and more robust window attention, making it well-suited for vision tasks such as image recognition and segmentation.

Total Parameters: 28,349,842

Function Signature¶

@register_model
def swin_v2_tiny(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, 6, 2], indicating that the model has 4 stages with 2, 2, 6, and 2 blocks respectively.
- num_heads (list[int]): A list specifying the number of attention heads in each stage. The common default is [3, 6, 12, 24], corresponding to the number of heads used in each stage, enabling multi-scale feature extraction.

Returns¶

SwinTransformer_V2: An instance of the SwinTransformer_V2 class configured as a lightweight vision transformer.

Examples¶

>>> import lucid.models as models
>>> model = models.swin_v2_tiny()
>>> print(model)
SwinTransformer_V2(img_size=224, num_classes=1000, ...)