braindecode.modules.CAT

class braindecode.modules.CAT(in_channels, reduction_rate, kernel_size, bias=False)

Attention Mechanism from [Wu2023].

Parameters:

• in_channels (int) – number of input feature channels

• reduction_rate (int) – reduction ratio of the fully-connected layers

• kernel_size (int) – kernel size of the convolutional layer

• bias (bool, default=False) – if True, adds a learnable bias will be used in the convolution,

Examples

>>> import torch
>>> from braindecode.modules import CAT
>>> module = CAT(in_channels=16, reduction_rate=4, kernel_size=3)
>>> inputs = torch.randn(2, 16, 1, 64)
>>> outputs = module(inputs)
>>> outputs.shape
torch.Size([2, 16, 1, 64])

References

[Wu2023]

Wu, Z. et al., 2023 CAT: Learning to Collaborate Channel and Spatial Attention from Multi-Information Fusion. IET Computer Vision 2023.

Methods

forward(x)

Apply the CAT block to the input tensor.

Parameters:

x (Pytorch.Tensor)

Return type:

Pytorch.Tensor