Attention(d_k[, n_heads, label, eps, dtype])
|
|
Decoder(d_vocab, n_heads, d_model, d_k, d_v, ...)
|
|
DecoderLayer(n_heads, d_model, d_k, d_v, ...)
|
|
Embedding(d_vocab, d_model[, dtype])
|
|
Encoder(d_vocab, n_heads, d_model, d_k, d_v, ...)
|
|
EncoderLayer(n_heads, d_model, d_k, d_v, ...)
|
|
MultiHeadAttention(n_heads, d_n, d_k, d_v, ...)
|
|
Transformer(enc_tokenizer, dec_tokenizer[, ...])
|
|
