def __init__(self, d_model, d_inner, n_head, d_k, d_v, dropout=0.1):
super(EncoderLayer, self).__init__()
self.slf_attn = MultiHeadAttention(n_head,
d_model,
d_k,
d_v,
dropout=dropout)
self.pos_ffn = PositionwiseFeedForward(d_model,
d_inner,
dropout=dropout)
def __init__(self,
headNum,
vectorDim,
entityDim,
hiddenDim,
dropout=0.1,
useAtt=False,
use_cuda=True):
super(TransformerLayer, self).__init__()
self.useAtt = useAtt
self.selfAttention = MultiHeadAttention(headNum,
vectorDim,
entityDim,
hiddenDim,
dropout=dropout,
useAtt=useAtt,
use_cuda=use_cuda)
self.posFeedForward = PositionwiseFeedForward(hiddenDim,
dropout=dropout,
use_cuda=use_cuda)
def __init__(self,
d_model,
d_inner,
n_head,
d_k,
d_v,
dropout=0.1,
rel_pos=False,
rel_pos_clip=None,
ex_mask=None):
super(EncoderLayer, self).__init__()
self.slf_attn = MultiHeadAttention(n_head,
d_model,
d_k,
d_v,
dropout=dropout,
rel_pos=rel_pos,
rel_pos_clip=rel_pos_clip,
ex_mask=ex_mask)
self.pos_ffn = PositionwiseFeedForward(d_model,
d_inner,
dropout=dropout)