最近参加了伯禹平台和Datawhale等举办的《动手学深度学习PyTorch版》课程,机器翻译及相关技术,注意力机制与Seq2seq模型做下笔记。
机器翻译和数据集
机器翻译(MT):将一段文本从一种语言自动翻译为另一种语言,用神经网络解决这个问题通常称为神经机器翻译(NMT)。 主要特征:输出是单词序列而不是单个单词。 输出序列的长度可能与源序列的长度不同。
使用Encoder-Decoder框架:
实现一个Encoder类
class Encoder(nn.Module):
def __init__(self, **kwargs):
super(Encoder, self).__init__(**kwargs)
def forward(self, X, *args):
raise NotImplementedError
实现一个Decoder类
class Decoder(nn.Module):
def __init__(self, **kwargs):
super(Decoder, self).__init__(**kwargs)
def init_state(self, enc_outputs, *args):
raise NotImplementedError
def forward(self, X, state):
raise NotImplementedError
实现一个拼接类把encode输出的state作为decode的初始state
class EncoderDecoder(nn.Module):
def __init__(self, encoder, decoder, **kwargs):
super(EncoderDecoder, self).__init__(**kwargs)
self.encoder = encoder
self.decoder = decoder
def forward(self, enc_X, dec_X, *args):
enc_outputs = self.encoder(enc_X, *args)
dec_state = self.decoder.init_state(enc_outputs, *args)
return self.decoder(dec_X, dec_state)
训练时:
左侧Encoder编码将输入序列转化成一个固定长度的向量编码,右侧Decoder解码将之前生成的固定向量再转化成输出序列。
<bos>是特殊字符表示句子的开头,在decoder训练时,用<bos>+前面的hidden state预测下一个单词。
预测时:用上一个预测出来的单词预测下一个
具体结构:
循环网络可以是深层循环网络(n层)
class Seq2SeqEncoder(d2l.Encoder):
def __init__(self, vocab_size, embed_size, num_hiddens, num_layers,
dropout=0, **kwargs):
super(Seq2SeqEncoder, self).__init__(**kwargs)
self.num_hiddens=num_hiddens
self.num_layers=num_layers
self.embedding = nn.Embedding(vocab_size, embed_size)
self.rnn = nn.LSTM(embed_size,num_hiddens, num_layers, dropout=dropout)
def begin_state(self, batch_size, device):
return [torch.zeros(size=(self.num_layers, batch_size, self.num_hiddens), device=device),
torch.zeros(size=(self.num_layers, batch_size, self.num_hiddens), device=device)]
def forward(self, X, *args):
X = self.embedding(X) # X shape: (batch_size, seq_len, embed_size)
X = X.transpose(0, 1) # RNN needs first axes to be time
# state = self.begin_state(X.shape[1], device=X.device)
out, state = self.rnn(X)
# The shape of out is (seq_len, batch_size, num_hiddens).
# state contains the hidden state and the memory cell
# of the last time step, the shape is (num_layers, batch_size, num_hiddens)
return out, state
class Seq2SeqDecoder(d2l.Decoder):
def __init__(self, vocab_size, embed_size, num_hiddens, num_layers,
dropout=0, **kwargs):
super(Seq2SeqDecoder, self).__init__(**kwargs)
self.embedding = nn.Embedding(vocab_size, embed_size)
self.rnn = nn.LSTM(embed_size,num_hiddens, num_layers, dropout=dropout)
self.dense = nn.Linear(num_hiddens,vocab_size)
def init_state(self, enc_outputs, *args):
return enc_outputs[1]
def forward(self, X, state):
X = self.embedding(X).transpose(0, 1)
out, state = self.rnn(X, state)
# Make the batch to be the first dimension to simplify loss computation.
out = self.dense(out).transpose(0, 1)
return out, state
计算损失函数时,不要把padding部分的损失也计算进去了
def SequenceMask(X, X_len,value=0):
maxlen = X.size(1)
mask = torch.arange(maxlen)[None, :].to(X_len.device) < X_len[:, None]
X[~mask]=value
return X
class MaskedSoftmaxCELoss(nn.CrossEntropyLoss):
# pred shape: (batch_size, seq_len, vocab_size)
# label shape: (batch_size, seq_len)
# valid_length shape: (batch_size, )
def forward(self, pred, label, valid_length):
# the sample weights shape should be (batch_size, seq_len)
weights = torch.ones_like(label)
weights = SequenceMask(weights, valid_length).float()
self.reduction='none'
output=super(MaskedSoftmaxCELoss, self).forward(pred.transpose(1,2), label)
return (output*weights).mean(dim=1)
Beam Search
简单greedy search:每次选择单词表中概率最大的单词作为下一个预测的输入,重复这个过程
缺点:没有考虑全局的最优解只考虑了当前的最优解
但是如果把所以的单词都试一遍选择整体分数最高的句子,搜索空间又太大了
集束搜索:
刚开始找最好的(beam=2)个:AC
接着在A之后和C之后的十个里面找最好的两个,BE
接着在AB之后和CE之后的十个里面找最好的两个,DD
注意力机制
由于解码的目标词语可能只与原输入的部分词语有关,而并不是与所有的输入有关,例如,当把“Hello world”翻译成“Bonjour le monde”时,“Hello”映射成“Bonjour”,“world”映射成“monde”,在seq2seq模型中,解码器只能隐式地从编码器的最终状态中选择相应的信息。然而,注意力机制可以将这种选择过程显式地建模。
可以参考另一个博主的这篇文章理解https://blog.csdn.net/hpulfc/article/details/80448570