braindecode.modules.MaxNormLinear#

class braindecode.modules.MaxNormLinear(in_features, out_features, bias=True, max_norm_val=2, eps=1e-05, **kwargs)[source]#

Linear layer with MaxNorm constraining on weights.

Equivalent of Keras tf.keras.Dense(…, kernel_constraint=max_norm()) [1] and [2]. Implemented as advised in [3].

Parameters:

  • in_features (int) – Size of each input sample.

  • out_features (int) – Size of each output sample.

  • bias (bool, optional) – If set to False, the layer will not learn an additive bias. Default: True.

Examples

>>> import torch
>>> from braindecode.modules import MaxNormLinear
>>> module = MaxNormLinear(10, 5, max_norm_val=2)
>>> inputs = torch.randn(2, 10)
>>> outputs = module(inputs)
>>> outputs.shape
torch.Size([2, 5])

References

On this page