espnet.nets.pytorch_backend.rnn.encoders.VGG2L

source

class espnet.nets.pytorch_backend.rnn.encoders.VGG2L(in_channel=1)

Bases: Module

VGG-like module.

• Parameters:in_channel (int) – number of input channels

Initialize VGG2L.

forward(xs_pad, ilens, **kwargs)

Forward VGG2L.

• Parameters:
  • xs_pad (torch.Tensor) – batch of padded input sequences (B, Tmax, D)
  • ilens (torch.Tensor) – batch of lengths of input sequences (B)
• Returns: batch of padded hidden state sequences (B, Tmax // 4, 128 * D // 4)
• Return type: torch.Tensor