import torch
from torch import nn
from torch.nn import functional as F
from d2l import torch as d2l

batch_size, num_steps = 32, 35
train_iter, vocab = d2l.load_data_time_machine(batch_size, num_steps)

8.6.1 定义模型

num_hiddens = 256
rnn_layer = nn.RNN(len(vocab), num_hiddens)

state = torch.zeros((1, batch_size, num_hiddens))
state.shape  # （隐藏层数，批量大小，隐藏单元数）

torch.Size([1, 32, 256])

通过一个隐状态和一个输入可以用更新后的隐状态计算输出。

需要强调的是，rnn_layer的“输出”（Y）不涉及输出层的计算：它是指每个时间步的隐状态，这些隐状态可以用作后续输出层的输入。

X = torch.rand(size=(num_steps, batch_size, len(vocab)))
Y, state_new = rnn_layer(X, state)
Y.shape, state_new.shape

(torch.Size([35, 32, 256]), torch.Size([1, 32, 256]))

#@save
class RNNModel(nn.Module):
    """循环神经网络模型"""
    def __init__(self, rnn_layer, vocab_size, **kwargs):
        super(RNNModel, self).__init__(**kwargs)
        self.rnn = rnn_layer
        self.vocab_size = vocab_size
        self.num_hiddens = self.rnn.hidden_size
        # 如果RNN是双向的（之后将介绍），num_directions应该是2，否则应该是1
        if not self.rnn.bidirectional:
            self.num_directions = 1
            self.linear = nn.Linear(self.num_hiddens, self.vocab_size)
        else:
            self.num_directions = 2
            self.linear = nn.Linear(self.num_hiddens * 2, self.vocab_size)

    def forward(self, inputs, state):
        X = F.one_hot(inputs.T.long(), self.vocab_size)
        X = X.to(torch.float32)
        Y, state = self.rnn(X, state)
        # 全连接层首先将Y的形状改为(时间步数*批量大小,隐藏单元数)
        # 它的输出形状是(时间步数*批量大小,词表大小)。
        output = self.linear(Y.reshape((-1, Y.shape[-1])))
        return output, state

    def begin_state(self, device, batch_size=1):
        if not isinstance(self.rnn, nn.LSTM):
            # nn.GRU以张量作为隐状态
            return  torch.zeros((self.num_directions * self.rnn.num_layers,
                                 batch_size, self.num_hiddens),
                                device=device)
        else:
            # nn.LSTM以元组作为隐状态
            return (torch.zeros((
                self.num_directions * self.rnn.num_layers,
                batch_size, self.num_hiddens), device=device),
                    torch.zeros((
                        self.num_directions * self.rnn.num_layers,
                        batch_size, self.num_hiddens), device=device))

8.6.2 训练与预测

device = d2l.try_gpu()
net = RNNModel(rnn_layer, vocab_size=len(vocab))
net = net.to(device)
d2l.predict_ch8('time traveller', 10, net, vocab, device)

'time travellerffffffffff'

num_epochs, lr = 500, 1
d2l.train_ch8(net, train_iter, vocab, lr, num_epochs, device)  # 比自己写的跑得快

perplexity 1.3, 213489.4 tokens/sec on cuda:0
time traveller held in his han so withtre scon the thin one mige
travellericho for the prof read haly and hes it nople hat d

练习

（1）尝试使用高级API，能使循环神经网络模型过拟合吗？

略。

（2）如果在循环神经网络模型中增加隐藏层的数量会发生什么？能使模型正常工作吗？

num_hiddens1 = 1024
rnn_layer1 = nn.RNN(len(vocab), num_hiddens1)

net1 = RNNModel(rnn_layer1, vocab_size=len(vocab))
net1 = net1.to(device)

num_epochs, lr = 500, 1
d2l.train_ch8(net1, train_iter, vocab, lr, num_epochs, device)  # 效果更好了，但是曲线没那么平滑了

perplexity 1.0, 97329.8 tokens/sec on cuda:0
time travelleryou can show black is white by argument said filby
travelleryou can show black is white by argument said filby

（3）尝试使用循环神经网络实现 8.1 节的自回归模型。

略