espnet.nets.pytorch_backend.transformer.layer_norm.LayerNorm

source

class espnet.nets.pytorch_backend.transformer.layer_norm.LayerNorm(nout, dim=-1)

Bases: LayerNorm

Layer normalization module.

• Parameters:
  • nout (int) – Output dim size.
  • dim (int) – Dimension to be normalized.

Construct an LayerNorm object.

forward(x)

Apply layer normalization.

• Parameters:x (torch.Tensor) – Input tensor.
• Returns: Normalized tensor.
• Return type: torch.Tensor