pvt_medium¶
The pvt_medium function instantiates the PVT-Medium model, a more powerful variant of the Pyramid Vision Transformer (PVT). It features greater depth and complexity, offering strong performance for difficult image recognition tasks while maintaining efficiency.
Total Parameters: 41,492,648
Function Signature¶
@register_model
def pvt_medium(img_size: int = 224, num_classes: int = 1000, **kwargs) -> PVT
Parameters¶
img_size (int, optional): The input image size. 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:
embed_dims (list[int]): A list specifying the embedding dimensions at different hierarchical levels. The default for PVT-Medium is [64, 128, 320, 512], offering deeper feature representations.
depths (list[int]): A list specifying the number of transformer blocks in each stage. The default values for PVT-Medium are [3, 4, 18, 3], providing deeper levels of abstraction.
num_heads (list[int]): The number of self-attention heads in each transformer stage. The default values are [1, 2, 5, 8], progressively capturing more complex relationships.
mlp_ratios (list[float]): The ratio of MLP hidden dimensions to the embedding dimension in each stage. Default values are [8.0, 8.0, 4.0, 4.0].
sr_ratios (list[float]): The spatial reduction ratios used in the self-attention mechanism. Default values are [8.0, 4.0, 2.0, 1.0].
Returns¶
PVT: An instance of the PVT class configured as a PVT-Medium vision transformer.
Examples¶
>>> import lucid.models as models
>>> model = models.pvt_medium()
>>> print(model)