一、编码器解码器架构：

1.定义：

Encoder负责对Input进行特征提取，输出特征矩阵State
Decoder接收State，负责进行预测并输出

2.在CNN中的体现：

3.在RNN中的体现：

4.代码：

from torch import nn

class Encoder(nn.Module):
    """编码器-解码器结构的基本编码器接口。"""
    def __init__(self, **kwargs):
        super(Encoder, self).__init__(**kwargs)

    def forward(self, X, *args):
        raise NotImplementedError
    
class Decoder(nn.Module):
    """编码器-解码器结构的基本解码器接口。"""
    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
    
class EncoderDecoder(nn.Module):
    """编码器-解码器结构的基类。"""
    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)    

二、Seq2seq：

1.模型架构：

这里以机器翻译任务为例：

1.1编码器：

编码器不管在训练阶段还是预测阶段都是用于提取特征，可以是单层RNN、多层RNN、双向RNN（双向RNN不仅可以提取上文序列特征，还可以提取下文的序列特征）

1.2解码器：

解码器在不同阶段作用不同，只能是单层RNN或多层RNN，不能是双向RNN（解码器用于预测，双向RNN不能预测）

• 训练阶段，解码器主要是为了特征提取，通过接收编码器的输出隐藏状态作为h0并接收预测的真实值Input，每个时间步使用隐藏状态ht-1进行特征提取并更新隐藏状态ht，然后将ht和当前时间步的真实值token（而非预测值，因为是要更好的学习）作为下一个时间步的输入，不断更新可学习参数。
• 预测阶段，解码器主要是为了执行预测任务，不再接收预测的真实值（因为不知道），仅接收编码器的输出隐藏状态作为h0，每个时间步使用隐藏状态ht-1进行预测并更新隐藏状态ht，然后将ht和当前时间步的预测值token作为下一个时间步的输入，进行下一个token的预测。

2.架构细节：

Seq2seq的编码器和解码器都是RNN

3.模型评估指标BLEU：

4.代码：

import collections
import math
import torch
from torch import nn
from d2l import torch as d2l

# 使用GRU作为编码器
class Seq2SeqEncoder(d2l.Encoder):
    def __init__(self, vocab_size, embed_size, num_hiddens, num_layers,
                 dropout=0, **kwargs):
        super(Seq2SeqEncoder, self).__init__(**kwargs)
        # 1
        self.embedding = nn.Embedding(vocab_size, embed_size)
        # 2 
        self.rnn = nn.GRU(embed_size, num_hiddens, num_layers,
                          dropout=dropout)

    def forward(self, X, *args):
        X = self.embedding(X)
        X = X.permute(1, 0, 2)
        output, state = self.rnn(X)
        return output, state
    
# 使用GRU作为解码器
class Seq2SeqDecoder(d2l.Decoder):
    def __init__(self, vocab_size, embed_size, num_hiddens, num_layers,
                 dropout=0, **kwargs):
        super(Seq2SeqDecoder, self).__init__(**kwargs)
        # 1
        self.embedding = nn.Embedding(vocab_size, embed_size)
        # 2
        self.rnn = nn.GRU(embed_size + num_hiddens, num_hiddens, num_layers,
                          dropout=dropout)
        # 3
        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).permute(1, 0, 2)
        context = state[-1].repeat(X.shape[0], 1, 1)
        X_and_context = torch.cat((X, context), 2)
        output, state = self.rnn(X_and_context, state)
        output = self.dense(output).permute(1, 0, 2)
        return output, state
    
# 在序列中屏蔽不相关的项,即屏蔽序列中之前使用<pad>填充的无效值   
def sequence_mask(X, valid_len, value=0):
    maxlen = X.size(1)
    mask = torch.arange((maxlen), dtype=torch.float32,
                        device=X.device)[None, :] < valid_len[:, None]
    X[~mask] = value
    return X

# 填充的无效值不参与损失值的计算，因为这些值的预测对错没有意义
class MaskedSoftmaxCELoss(nn.CrossEntropyLoss):
    def forward(self, pred, label, valid_len):
        weights = torch.ones_like(label)
        weights = sequence_mask(weights, valid_len)
        self.reduction = 'none'
        unweighted_loss = super(MaskedSoftmaxCELoss,
                                self).forward(pred.permute(0, 2, 1), label)
        weighted_loss = (unweighted_loss * weights).mean(dim=1)
        return weighted_loss
    
# 训练过程
def train_seq2seq(net, data_iter, lr, num_epochs, tgt_vocab, device):
    def xavier_init_weights(m):
        if type(m) == nn.Linear:
            nn.init.xavier_uniform_(m.weight)
        if type(m) == nn.GRU:
            for param in m._flat_weights_names:
                if "weight" in param:
                    nn.init.xavier_uniform_(m._parameters[param])

    net.apply(xavier_init_weights)
    net.to(device)
    optimizer = torch.optim.Adam(net.parameters(), lr=lr)
    loss = MaskedSoftmaxCELoss()
    net.train()
    animator = d2l.Animator(xlabel='epoch', ylabel='loss',
                            xlim=[10, num_epochs])
    for epoch in range(num_epochs):
        timer = d2l.Timer()
        metric = d2l.Accumulator(2)
        for batch in data_iter:
            X, X_valid_len, Y, Y_valid_len = [x.to(device) for x in batch]
            bos = torch.tensor([tgt_vocab['<bos>']] * Y.shape[0],
                               device=device).reshape(-1, 1)
            dec_input = torch.cat([bos, Y[:, :-1]], 1)
            Y_hat, _ = net(X, dec_input, X_valid_len)
            l = loss(Y_hat, Y, Y_valid_len)
            l.sum().backward()
            d2l.grad_clipping(net, 1)
            num_tokens = Y_valid_len.sum()
            optimizer.step()
            with torch.no_grad():
                metric.add(l.sum(), num_tokens)
        if (epoch + 1) % 10 == 0:
            animator.add(epoch + 1, (metric[0] / metric[1],))
    print(f'loss {metric[0] / metric[1]:.3f}, {metric[1] / timer.stop():.1f} '
          f'tokens/sec on {str(device)}')
    
# 预测过程    
def predict_seq2seq(net, src_sentence, src_vocab, tgt_vocab, num_steps,
                    device, save_attention_weights=False):
    net.eval()
    src_tokens = src_vocab[src_sentence.lower().split(' ')] + [
        src_vocab['<eos>']]
    enc_valid_len = torch.tensor([len(src_tokens)], device=device)
    src_tokens = d2l.truncate_pad(src_tokens, num_steps, src_vocab['<pad>'])
    enc_X = torch.unsqueeze(
        torch.tensor(src_tokens, dtype=torch.long, device=device), dim=0)
    enc_outputs = net.encoder(enc_X, enc_valid_len)
    dec_state = net.decoder.init_state(enc_outputs, enc_valid_len)
    dec_X = torch.unsqueeze(
        torch.tensor([tgt_vocab['<bos>']], dtype=torch.long, device=device),
        dim=0)
    output_seq, attention_weight_seq = [], []
    for _ in range(num_steps):
        Y, dec_state = net.decoder(dec_X, dec_state)
        dec_X = Y.argmax(dim=2)
        pred = dec_X.squeeze(dim=0).type(torch.int32).item()
        if save_attention_weights:
            attention_weight_seq.append(net.decoder.attention_weights)
        if pred == tgt_vocab['<eos>']:
            break
        output_seq.append(pred)
    return ' '.join(tgt_vocab.to_tokens(output_seq)), attention_weight_seq    
    
# 模型评估指标
def bleu(pred_seq, label_seq, k):  
    pred_tokens, label_tokens = pred_seq.split(' '), label_seq.split(' ')
    len_pred, len_label = len(pred_tokens), len(label_tokens)
    score = math.exp(min(0, 1 - len_label / len_pred))
    for n in range(1, k + 1):
        num_matches, label_subs = 0, collections.defaultdict(int)
        for i in range(len_label - n + 1):
            label_subs[''.join(label_tokens[i:i + n])] += 1
        for i in range(len_pred - n + 1):
            if label_subs[''.join(pred_tokens[i:i + n])] > 0:
                num_matches += 1
                label_subs[''.join(pred_tokens[i:i + n])] -= 1
        score *= math.pow(num_matches / (len_pred - n + 1), math.pow(0.5, n))
    return score

# 训练    
embed_size, num_hiddens, num_layers, dropout = 32, 32, 2, 0.1
batch_size, num_steps = 64, 10
lr, num_epochs, device = 0.005, 300, d2l.try_gpu()

train_iter, src_vocab, tgt_vocab = d2l.load_data_nmt(batch_size, num_steps)
encoder = Seq2SeqEncoder(len(src_vocab), embed_size, num_hiddens, num_layers,
                         dropout)
decoder = Seq2SeqDecoder(len(tgt_vocab), embed_size, num_hiddens, num_layers,
                         dropout)
net = d2l.EncoderDecoder(encoder, decoder)
train_seq2seq(net, train_iter, lr, num_epochs, tgt_vocab, device)    
    
# 预测    
engs = ['go .', "i lost .", 'he\'s calm .', 'i\'m home .']
fras = ['va !', 'j\'ai perdu .', 'il est calme .', 'je suis chez moi .']
for eng, fra in zip(engs, fras):
    translation, attention_weight_seq = predict_seq2seq(
        net, eng, src_vocab, tgt_vocab, num_steps, device)
    print(f'{eng} => {translation}, bleu {bleu(translation, fra, k=2):.3f}')

三、束搜索：

1.贪心搜索：

2.束搜索：