espnet.nets.pytorch_backend.transformer.subsampling.Conv2dSubsampling1

source

class espnet.nets.pytorch_backend.transformer.subsampling.Conv2dSubsampling1(idim, odim, dropout_rate, pos_enc=None)

Bases: Module

Similar to Conv2dSubsampling module, but without any subsampling performed.

• Parameters:
  • idim (int) – Input dimension.
  • odim (int) – Output dimension.
  • dropout_rate (float) – Dropout rate.
  • pos_enc (torch.nn.Module) – Custom position encoding layer.

Construct an Conv2dSubsampling1 object.

forward(x, x_mask)

Pass x through 2 Conv2d layers without subsampling.

• Parameters:

  • x (torch.Tensor) – Input tensor (#batch, time, idim).
  • x_mask (torch.Tensor) – Input mask (#batch, 1, time).

• Returns: Subsampled tensor (#batch, time’, odim). : where time’ = time - 4.

  torch.Tensor: Subsampled mask (#batch, 1, time’). : where time’ = time - 4.

• Return type: torch.Tensor