torch (version 0.2.0)

nn_multihead_attention: MultiHead attention

Description

Allows the model to jointly attend to information from different representation subspaces. See reference: Attention Is All You Need

Usage

nn_multihead_attention(
  embed_dim,
  num_heads,
  dropout = 0,
  bias = TRUE,
  add_bias_kv = FALSE,
  add_zero_attn = FALSE,
  kdim = NULL,
  vdim = NULL
)

Arguments

embed_dim

total dimension of the model.

num_heads

parallel attention heads.

dropout

a Dropout layer on attn_output_weights. Default: 0.0.

bias

add bias as module parameter. Default: True.

add_bias_kv

add bias to the key and value sequences at dim=0.

add_zero_attn

add a new batch of zeros to the key and value sequences at dim=1.

kdim

total number of features in key. Default: NULL

vdim

total number of features in value. Default: NULL. Note: if kdim and vdim are NULL, they will be set to embed_dim such that query, key, and value have the same number of features.

Shape

Inputs:

• query: \((L, N, E)\) where L is the target sequence length, N is the batch size, E is the embedding dimension.

• key: \((S, N, E)\), where S is the source sequence length, N is the batch size, E is the embedding dimension.

• value: \((S, N, E)\) where S is the source sequence length, N is the batch size, E is the embedding dimension.

• key_padding_mask: \((N, S)\) where N is the batch size, S is the source sequence length. If a ByteTensor is provided, the non-zero positions will be ignored while the position with the zero positions will be unchanged. If a BoolTensor is provided, the positions with the value of True will be ignored while the position with the value of False will be unchanged.

• attn_mask: 2D mask \((L, S)\) where L is the target sequence length, S is the source sequence length. 3D mask \((N*num_heads, L, S)\) where N is the batch size, L is the target sequence length, S is the source sequence length. attn_mask ensure that position i is allowed to attend the unmasked positions. If a ByteTensor is provided, the non-zero positions are not allowed to attend while the zero positions will be unchanged. If a BoolTensor is provided, positions with True is not allowed to attend while False values will be unchanged. If a FloatTensor is provided, it will be added to the attention weight.

Outputs:

• attn_output: \((L, N, E)\) where L is the target sequence length, N is the batch size, E is the embedding dimension.

• attn_output_weights: \((N, L, S)\) where N is the batch size, L is the target sequence length, S is the source sequence length.

Details

$$ \mbox{MultiHead}(Q, K, V) = \mbox{Concat}(head_1,\dots,head_h)W^O \mbox{where} head_i = \mbox{Attention}(QW_i^Q, KW_i^K, VW_i^V) $$

Examples

# NOT RUN {
if (torch_is_installed()) {
# }
# NOT RUN {
multihead_attn = nn_multihead_attention(embed_dim, num_heads)
out <- multihead_attn(query, key, value)
attn_output <- out[[1]]
attn_output_weights <- out[[2]]
# }
# NOT RUN {
}
# }