TransformerEncoder

class torch.nn.TransformerEncoder(encoder_layer, num_layers, norm=None, enable_nested_tensor=True, mask_check=True)

TransformerEncoder is a stack of N encoder layers. Users can build the BERT(https://arxiv.org/abs/1810.04805) model with corresponding parameters.

Parameters:

• encoder_layer – an instance of the TransformerEncoderLayer() class (required).

• num_layers – the number of sub-encoder-layers in the encoder (required).

• norm – the layer normalization component (optional).

• enable_nested_tensor – if True, input will automatically convert to nested tensor (and convert back on output). This will improve the overall performance of TransformerEncoder when padding rate is high. Default: True (enabled).

Examples::

>>> encoder_layer = nn.TransformerEncoderLayer(d_model=512, nhead=8)
>>> transformer_encoder = nn.TransformerEncoder(encoder_layer, num_layers=6)
>>> src = torch.rand(10, 32, 512)
>>> out = transformer_encoder(src)

forward(src, mask=None, src_key_padding_mask=None, is_causal=None)

Pass the input through the encoder layers in turn.

Parameters:

• src (Tensor) – the sequence to the encoder (required).

• mask (Optional[Tensor]) – the mask for the src sequence (optional).

• is_causal (Optional[bool]) – If specified, applies a causal mask as mask (optional) and ignores attn_mask for computing scaled dot product attention. Default: False.

• src_key_padding_mask (Optional[Tensor]) – the mask for the src keys per batch (optional).

Return type:

Tensor

Shape:

see the docs in Transformer class.