.. _sec_rnn-concise:
循环神经网络的简洁实现
======================
虽然 :numref:`sec_rnn_scratch`
对了解循环神经网络的实现方式具有指导意义，但并不方便。
本节将展示如何使用深度学习框架的高级API提供的函数更有效地实现相同的语言模型。
我们仍然从读取时光机器数据集开始。
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
from mxnet import np, npx
from mxnet.gluon import nn, rnn
from d2l import mxnet as d2l
npx.set_np()
batch_size, num_steps = 32, 35
train_iter, vocab = d2l.load_data_time_machine(batch_size, num_steps)
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
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)
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
import tensorflow as tf
from d2l import tensorflow as d2l
batch_size, num_steps = 32, 35
train_iter, vocab = d2l.load_data_time_machine(batch_size, num_steps)
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
import warnings
from d2l import paddle as d2l
warnings.filterwarnings("ignore")
import paddle
from paddle import nn
from paddle.nn import functional as F
batch_size, num_steps = 32, 35
train_iter, vocab = d2l.load_data_time_machine(batch_size, num_steps)
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
num_hiddens = 256
rnn_layer = rnn.RNN(num_hiddens)
rnn_layer.initialize()
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
[07:46:28] ../src/storage/storage.cc:196: Using Pooled (Naive) StorageManager for CPU
初始化隐状态是简单的，只需要调用成员函数\ ``begin_state``\ 即可。
函数将返回一个列表（\ ``state``\ ），列表中包含了初始隐状态用于小批量数据中的每个样本，
其形状为（隐藏层数，批量大小，隐藏单元数）。
对于以后要介绍的一些模型（例如长-短期记忆网络），这样的列表还会包含其他信息。
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
state = rnn_layer.begin_state(batch_size=batch_size)
len(state), state[0].shape
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
(1, (1, 32, 256))
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
num_hiddens = 256
rnn_layer = nn.RNN(len(vocab), num_hiddens)
我们使用张量来初始化隐状态，它的形状是（隐藏层数，批量大小，隐藏单元数）。
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
state = torch.zeros((1, batch_size, num_hiddens))
state.shape
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
torch.Size([1, 32, 256])
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
num_hiddens = 256
rnn_cell = tf.keras.layers.SimpleRNNCell(num_hiddens,
kernel_initializer='glorot_uniform')
rnn_layer = tf.keras.layers.RNN(rnn_cell, time_major=True,
return_sequences=True, return_state=True)
state = rnn_cell.get_initial_state(batch_size=batch_size, dtype=tf.float32)
state.shape
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
TensorShape([32, 256])
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
num_hiddens = 256
rnn_layer = nn.SimpleRNN(len(vocab), num_hiddens, time_major=True)
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
W0818 09:37:14.584262 68518 gpu_resources.cc:61] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 11.8, Runtime API Version: 11.8
W0818 09:37:14.614424 68518 gpu_resources.cc:91] device: 0, cuDNN Version: 8.7.
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
state = paddle.zeros(shape=[1, batch_size, num_hiddens])
state.shape
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
[1, 32, 256]
.. raw:: html
.. raw:: html
.. raw:: html
此外，\ ``rnn_layer``\ 返回的更新后的隐状态（\ ``state_new``\ ）
是指小批量数据的最后时间步的隐状态。
这个隐状态可以用来初始化顺序分区中一个迭代周期内下一个小批量数据的隐状态。
对于多个隐藏层，每一层的隐状态将存储在（\ ``state_new``\ ）变量中。
至于稍后要介绍的某些模型（例如，长－短期记忆），此变量还包含其他信息。
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
X = np.random.uniform(size=(num_steps, batch_size, len(vocab)))
Y, state_new = rnn_layer(X, state)
Y.shape, len(state_new), state_new[0].shape
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
((35, 32, 256), 1, (1, 32, 256))
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
X = torch.rand(size=(num_steps, batch_size, len(vocab)))
Y, state_new = rnn_layer(X, state)
Y.shape, state_new.shape
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
(torch.Size([35, 32, 256]), torch.Size([1, 32, 256]))
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
X = tf.random.uniform((num_steps, batch_size, len(vocab)))
Y, state_new = rnn_layer(X, state)
Y.shape, len(state_new), state_new[0].shape
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
(TensorShape([35, 32, 256]), 32, TensorShape([256]))
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
X = paddle.rand(shape=[num_steps, batch_size, len(vocab)])
Y, state_new = rnn_layer(X, state)
Y.shape, state_new.shape
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
([35, 32, 256], [1, 32, 256])
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
#@save
class RNNModel(nn.Block):
"""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.dense = nn.Dense(vocab_size)
def forward(self, inputs, state):
X = npx.one_hot(inputs.T, self.vocab_size)
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.dense(Y.reshape(-1, Y.shape[-1]))
return output, state
def begin_state(self, *args, **kwargs):
return self.rnn.begin_state(*args, **kwargs)
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
#@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))
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
#@save
class RNNModel(tf.keras.layers.Layer):
def __init__(self, rnn_layer, vocab_size, **kwargs):
super(RNNModel, self).__init__(**kwargs)
self.rnn = rnn_layer
self.vocab_size = vocab_size
self.dense = tf.keras.layers.Dense(vocab_size)
def call(self, inputs, state):
X = tf.one_hot(tf.transpose(inputs), self.vocab_size)
# rnn返回两个以上的值
Y, *state = self.rnn(X, state)
output = self.dense(tf.reshape(Y, (-1, Y.shape[-1])))
return output, state
def begin_state(self, *args, **kwargs):
return self.rnn.cell.get_initial_state(*args, **kwargs)
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
#@save
class RNNModel(nn.Layer):
"""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 self.rnn.num_directions==1:
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, self.vocab_size)
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, batch_size=1):
if not isinstance(self.rnn, nn.LSTM):
# nn.GRU以张量作为隐状态
return paddle.zeros(shape=[self.num_directions * self.rnn.num_layers,
batch_size, self.num_hiddens])
else:
# nn.LSTM以元组作为隐状态
return (paddle.zeros(
shape=[self.num_directions * self.rnn.num_layers,
batch_size, self.num_hiddens]),
paddle.zeros(
shape=[self.num_directions * self.rnn.num_layers,
batch_size, self.num_hiddens]))
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
device = d2l.try_gpu()
net = RNNModel(rnn_layer, len(vocab))
net.initialize(force_reinit=True, ctx=device)
d2l.predict_ch8('time traveller', 10, net, vocab, device)
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
[07:46:29] ../src/storage/storage.cc:196: Using Pooled (Naive) StorageManager for GPU
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
'time travellervmoopwrrrr'
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
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)
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
'time travellerbbabbkabyg'
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
device_name = d2l.try_gpu()._device_name
strategy = tf.distribute.OneDeviceStrategy(device_name)
with strategy.scope():
net = RNNModel(rnn_layer, vocab_size=len(vocab))
d2l.predict_ch8('time traveller', 10, net, vocab)
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
'time travellerfgxfmzovt'
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
device = d2l.try_gpu()
net = RNNModel(rnn_layer, vocab_size=len(vocab))
d2l.predict_ch8('time traveller', 10, net, vocab, device)
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
'time travellercmppppppp'
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
num_epochs, lr = 500, 1
d2l.train_ch8(net, train_iter, vocab, lr, num_epochs, device)
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
perplexity 1.2, 144941.9 tokens/sec on gpu(0)
time travellerit s against reason said filby of course a solid b
travelleryou can show black is whine be move the endled the
.. figure:: output_rnn-concise_eff2f4_84_1.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
num_epochs, lr = 500, 1
d2l.train_ch8(net, train_iter, vocab, lr, num_epochs, device)
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
perplexity 1.3, 404413.8 tokens/sec on cuda:0
time travellerit would be remarkably convenient for the historia
travellery of il the hise fupt might and st was it loflers
.. figure:: output_rnn-concise_eff2f4_87_1.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
num_epochs, lr = 500, 1
d2l.train_ch8(net, train_iter, vocab, lr, num_epochs, strategy)
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
perplexity 1.3, 14777.4 tokens/sec on /GPU:0
time traveller spod is his filenctot on shichthes sonee tothit o
traveller hald ane taly the onsthete or sime ance to that b
.. figure:: output_rnn-concise_eff2f4_90_1.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
num_epochs, lr = 500, 1.0
d2l.train_ch8(net, train_iter, vocab, lr, num_epochs, device)
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
困惑度 1.3, 129676.7 词元/秒 Place(gpu:0)
time travellerit s againstritovemabsen thingst and why casnother
traveller acowist has neitly sealexith has andlysubl and fo
.. figure:: output_rnn-concise_eff2f4_93_1.svg
.. raw:: html
.. raw:: html
.. raw:: html
`Discussions `__
.. raw:: html
.. raw:: html
`Discussions `__
.. raw:: html
.. raw:: html
`Discussions `__
.. raw:: html
.. raw:: html
`Discussions `__
.. raw:: html
.. raw:: html