K-BERT/uer/subencoders/rnn_subencoder.py
2019-12-12 19:37:32 +08:00

26 lines
992 B
Python
Executable File

import torch
import torch.nn as nn
class LstmSubencoder(nn.Module):
def __init__(self, args, vocab_size):
super(LstmSubencoder, self).__init__()
self.hidden_size= args.emb_size
self.layers_num = args.sub_layers_num
self.embedding_layer = nn.Embedding(vocab_size, args.emb_size)
self.rnn = nn.LSTM(input_size=args.emb_size,
hidden_size=self.hidden_size,
num_layers=self.layers_num,
dropout=args.dropout,
batch_first=True)
def forward(self, ids):
batch_size, _ = ids.size() # batch_size, max_length
hidden = (torch.zeros(self.layers_num, batch_size, self.hidden_size).to(ids.device),
torch.zeros(self.layers_num, batch_size, self.hidden_size).to(ids.device))
emb = self.embedding_layer(ids)
output, hidden = self.rnn(emb, hidden)
output = output.mean(1)
return output