6.5.1 定义模型

Mxnet:

num_hiddens = 256
rnn_layer = rnn.RNN(num_hiddens)
rnn_layer.initialize()

batch_size = 2
state = rnn_layer.begin_state(batch_size=batch_size)
state[0].shape

num_steps = 35
X = nd.random.uniform(shape=(num_steps, batch_size, vocab_size))
Y, state_new = rnn_layer(X, state)
Y.shape, len(state_new), state_new[0].shape

# 本类已保存在d2lzh包中方便以后使用
class RNNModel(nn.Block):
    def __init__(self, rnn_layer, vocab_size, **kwargs):
        super(RNNModel, self).__init__(**kwargs)
        self.rnn = rnn_layer
        self.vocab_size = vocab_size
        self.dense = nn.Dense(vocab_size)

    def forward(self, inputs, state):
        # 将输入转置成(num_steps, batch_size)后获取one-hot向量表示
        X = nd.one_hot(inputs.T, self.vocab_size)
        Y, state = self.rnn(X, state)
        # 全连接层会首先将Y的形状变成(num_steps * batch_size, num_hiddens)，它的输出
        # 形状为(num_steps * batch_size, vocab_size)
        output = self.dense(Y.reshape((-1, Y.shape[-1])))
        return output, state

    def begin_state(self, *args, **kwargs):
        return self.rnn.begin_state(*args, **kwargs)

Pytorch:

num_hiddens = 256
rnn_layer = nn.RNN(input_size=vocab_size, hidden_size=num_hiddens)

num_steps = 35
batch_size = 2
state = None
X = torch.rand(num_steps, batch_size, vocab_size)
Y, state_new = rnn_layer(X, state)
print(Y.shape, len(state_new), state_new[0].shape)

# 本类已保存在d2lzh_pytorch包中方便以后使用
class RNNModel(nn.Module):
    def __init__(self, rnn_layer, vocab_size):
        super(RNNModel, self).__init__()
        self.rnn = rnn_layer
        self.hidden_size = rnn_layer.hidden_size * (2 if rnn_layer.bidirectional else 1) 
        self.vocab_size = vocab_size
        self.dense = nn.Linear(self.hidden_size, vocab_size)
        self.state = None

    def forward(self, inputs, state): # inputs: (batch, seq_len)
        # 获取one-hot向量表示
        X = d2l.to_onehot(inputs, vocab_size) # X是个list
        Y, self.state = self.rnn(torch.stack(X), state)
        # 全连接层会首先将Y的形状变成(num_steps * batch_size, num_hiddens)，它的输出
        # 形状为(num_steps * batch_size, vocab_size)
        output = self.dense(Y.view(-1, Y.shape[-1]))
        return output, self.state

6.5.2 训练模型

Mxnet:

# 本函数已保存在d2lzh包中方便以后使用
def predict_rnn_gluon(prefix, num_chars, model, vocab_size, ctx, idx_to_char,
                      char_to_idx):
    # 使用model的成员函数来初始化隐藏状态
    state = model.begin_state(batch_size=1, ctx=ctx)
    output = [char_to_idx[prefix[0]]]
    for t in range(num_chars + len(prefix) - 1):
        X = nd.array([output[-1]], ctx=ctx).reshape((1, 1))
        (Y, state) = model(X, state)  # 前向计算不需要传入模型参数
        if t < len(prefix) - 1:
            output.append(char_to_idx[prefix[t + 1]])
        else:
            output.append(int(Y.argmax(axis=1).asscalar()))
    return ''.join([idx_to_char[i] for i in output])

# 本函数已保存在d2lzh包中方便以后使用
def train_and_predict_rnn_gluon(model, num_hiddens, vocab_size, ctx,
                                corpus_indices, idx_to_char, char_to_idx,
                                num_epochs, num_steps, lr, clipping_theta,
                                batch_size, pred_period, pred_len, prefixes):
    loss = gloss.SoftmaxCrossEntropyLoss()
    model.initialize(ctx=ctx, force_reinit=True, init=init.Normal(0.01))
    trainer = gluon.Trainer(model.collect_params(), 'sgd',
                            {'learning_rate': lr, 'momentum': 0, 'wd': 0})

    for epoch in range(num_epochs):
        l_sum, n, start = 0.0, 0, time.time()
        data_iter = d2l.data_iter_consecutive(
            corpus_indices, batch_size, num_steps, ctx)
        state = model.begin_state(batch_size=batch_size, ctx=ctx)
        for X, Y in data_iter:
            for s in state:
                s.detach()
            with autograd.record():
                (output, state) = model(X, state)
                y = Y.T.reshape((-1,))
                l = loss(output, y).mean()
            l.backward()
            # 梯度裁剪
            params = [p.data() for p in model.collect_params().values()]
            d2l.grad_clipping(params, clipping_theta, ctx)
            trainer.step(1)  # 因为已经误差取过均值，梯度不用再做平均
            l_sum += l.asscalar() * y.size
            n += y.size

        if (epoch + 1) % pred_period == 0:
            print('epoch %d, perplexity %f, time %.2f sec' % (
                epoch + 1, math.exp(l_sum / n), time.time() - start))
            for prefix in prefixes:
                print(' -', predict_rnn_gluon(
                    prefix, pred_len, model, vocab_size, ctx, idx_to_char,
                    char_to_idx))

num_epochs, batch_size, lr, clipping_theta = 250, 32, 1e2, 1e-2
pred_period, pred_len, prefixes = 50, 50, ['分开', '不分开']
train_and_predict_rnn_gluon(model, num_hiddens, vocab_size, ctx,
                            corpus_indices, idx_to_char, char_to_idx,
                            num_epochs, num_steps, lr, clipping_theta,
                            batch_size, pred_period, pred_len, prefixes)

Pytorch:

# 本函数已保存在d2lzh_pytorch包中方便以后使用
def predict_rnn_pytorch(prefix, num_chars, model, vocab_size, device, idx_to_char,
                      char_to_idx):
    state = None
    output = [char_to_idx[prefix[0]]] # output会记录prefix加上输出
    for t in range(num_chars + len(prefix) - 1):
        X = torch.tensor([output[-1]], device=device).view(1, 1)
        if state is not None:
            if isinstance(state, tuple): # LSTM, state:(h, c)  
                state = (state[0].to(device), state[1].to(device))
            else:   
                state = state.to(device)
            
        (Y, state) = model(X, state)  # 前向计算不需要传入模型参数
        if t < len(prefix) - 1:
            output.append(char_to_idx[prefix[t + 1]])
        else:
            output.append(int(Y.argmax(dim=1).item()))
    return ''.join([idx_to_char[i] for i in output])

# 本函数已保存在d2lzh_pytorch包中方便以后使用
def train_and_predict_rnn_pytorch(model, num_hiddens, vocab_size, device,
                                corpus_indices, idx_to_char, char_to_idx,
                                num_epochs, num_steps, lr, clipping_theta,
                                batch_size, pred_period, pred_len, prefixes):
    loss = nn.CrossEntropyLoss()
    optimizer = torch.optim.Adam(model.parameters(), lr=lr)
    model.to(device)
    state = None
    for epoch in range(num_epochs):
        l_sum, n, start = 0.0, 0, time.time()
        data_iter = d2l.data_iter_consecutive(corpus_indices, batch_size, num_steps, device) # 相邻采样
        for X, Y in data_iter:
            if state is not None:
                # 使用detach函数从计算图分离隐藏状态, 这是为了
                # 使模型参数的梯度计算只依赖一次迭代读取的小批量序列(防止梯度计算开销太大)
                if isinstance (state, tuple): # LSTM, state:(h, c)  
                    state = (state[0].detach(), state[1].detach())
                else:   
                    state = state.detach()
    
            (output, state) = model(X, state) # output: 形状为(num_steps * batch_size, vocab_size)
            
            # Y的形状是(batch_size, num_steps)，转置后再变成长度为
            # batch * num_steps 的向量，这样跟输出的行一一对应
            y = torch.transpose(Y, 0, 1).contiguous().view(-1)
            l = loss(output, y.long())
            
            optimizer.zero_grad()
            l.backward()
            # 梯度裁剪
            d2l.grad_clipping(model.parameters(), clipping_theta, device)
            optimizer.step()
            l_sum += l.item() * y.shape[0]
            n += y.shape[0]
        
        try:
            perplexity = math.exp(l_sum / n)
        except OverflowError:
            perplexity = float('inf')
        if (epoch + 1) % pred_period == 0:
            print('epoch %d, perplexity %f, time %.2f sec' % (
                epoch + 1, perplexity, time.time() - start))
            for prefix in prefixes:
                print(' -', predict_rnn_pytorch(
                    prefix, pred_len, model, vocab_size, device, idx_to_char,
                    char_to_idx))

num_epochs, batch_size, lr, clipping_theta = 250, 32, 1e-3, 1e-2 # 注意这里的学习率设置
pred_period, pred_len, prefixes = 50, 50, ['分开', '不分开']
train_and_predict_rnn_pytorch(model, num_hiddens, vocab_size, device,
                            corpus_indices, idx_to_char, char_to_idx,
                            num_epochs, num_steps, lr, clipping_theta,
                            batch_size, pred_period, pred_len, prefixes)