8.6.rnn concise
nn.RNN
接口很简单。注意隐状态 state
是在 RNN 对象之外另开对象进行存储的。
batch_size, num_steps = 32, 35
train_iter, vocab = d2l.load_data_time_machine(batch_size, num_steps)
num_hiddens = 256
# RNN() 内部含有 W, b 参数之类的
rnn_layer = nn.RNN(len(vocab), num_hiddens)
# state 形状 [隐藏层数,批量大小,隐藏单元数]
state = torch.zeros((1, batch_size, num_hiddens))
# X.shape = [时间步,批量,词典大小(独热)]
X = torch.rand(size=(num_steps, batch_size, len(vocab)))
# Y.shape = [时间步,批量,隐藏单元数], state 形状不变
Y, state = self.rnn(X, state)
封装
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))
32:07