tf.nn.static_bidirectional_rnn
Aliases:
- tf.contrib.rnn.static_bidirectional_rnn
- tf.nn.static_bidirectional_rnn
tf.nn.static_bidirectional_rnn(
cell_fw,
cell_bw,
inputs,
initial_state_fw=None,
initial_state_bw=None,
dtype=None,
sequence_length=None,
scope=None
)
Defined in tensorflow/python/ops/rnn.py.
See the guide: RNN and Cells (contrib) > Recurrent Neural Networks
Creates a bidirectional recurrent neural network.
Similar to the unidirectional case above (rnn) but takes input and builds independent forward and backward RNNs with the final forward and backward outputs depth-concatenated, such that the output will have the format [time][batch][cell_fw.output_size + cell_bw.output_size]. The input_size of forward and backward cell must match. The initial state for both directions is zero by default (but can be set optionally) and no intermediate states are ever returned -- the network is fully unrolled for the given (passed in) length(s) of the sequence(s) or completely unrolled if length(s) is not given.
Args:
- cell_fw:用于前向传播的RNNCell.
- cell_bw: 用于反向传播的RNNCell.
- inputs: A length T list of inputs, each a tensor of shape [batch_size, input_size], or a nested tuple of such elements.(输入 数据为list类型,list中元素为Tensor,每个Tensor的shape为[batch_size, input_size],如有batch_size个文档,每个文档的单词数量为1000,每个单词的词向量维度为100,则该inputs为list(1000*tensor(batch_size*100)))
- initial_state_fw: (optional)前向RNN的初始状态。 This must be a tensor of appropriate type and shape [batch_size, cell_fw.state_size]. If cell_fw.state_size is a tuple, this should be a tuple of tensors having shapes [batch_size, s] for s in cell_fw.state_size.。
- initial_state_bw: (optional) Same as for initial_state_fw, but using the corresponding properties of cell_bw.
- dtype: (optional) The data type for the initial state. Required if either of the initial states are not provided.
- sequence_length: (optional) An int32/int64 vector, size [batch_size], containing the actual lengths for each of the sequences.
- scope: VariableScope for the created subgraph; defaults to "bidirectional_rnn"
Returns:
A tuple (outputs, output_state_fw, output_state_bw) where: outputs is a length T list of outputs (one for each input), which are depth-concatenated forward and backward outputs. output_state_fw is the final state of the forward rnn. output_state_bw is the final state of the backward rnn.
Raises:
- TypeError: If cell_fw or cell_bw is not an instance of RNNCell.
- ValueError: If inputs is None or an empty list.
来源:https://tensorflow.google.cn/api_docs/python/tf/nn/static_bidirectional_rnn
import tensorflow as tf
import numpy as np
# 设置训练参数
learning_rate = 0.01
max_examples = 400000
batch_size = 128
display_step = 10 # 每间隔10次训练就展示一次训练情况
n_input = 100
n_steps = 300
fw_n_hidden = 256
bw_n_hidden = 128
n_classes = 10
x = tf.placeholder("float", [10000, n_steps, n_input])
y = tf.placeholder('float', [10000, n_classes])
weights = tf.Variable(tf.random_normal([(fw_n_hidden + bw_n_hidden), n_classes]))
biases = tf.Variable(tf.random_normal([n_classes]))
x = tf.transpose(x, [1, 0, 2])
print(x.shape) # (256, 10000, 100)
x = tf.reshape(x, [-1, n_input])
print(x.shape) # (2560000, 100)
x = tf.split(x, n_steps)
print(len(x), x[0].shape) # (10000, 512)
lstm_fw_cell = tf.contrib.rnn.BasicLSTMCell(fw_n_hidden, forget_bias=1.0) # 正向RNN,输出神经元数量为256
lstm_bw_cell = tf.contrib.rnn.BasicLSTMCell(bw_n_hidden, forget_bias=1.0) # 反向RNN,输出神经元数量为128
outputs, fw_state, bw_state = tf.contrib.rnn.static_bidirectional_rnn(lstm_fw_cell, lstm_bw_cell, x, dtype=tf.float32)
print(outputs[0].shape) # (10000, 384),384为正向RNN的输出神经元数量256和反向RNN的
print(len(outputs))#300,等于时间步的长度,一般取outputs[-1]也就是最后一步的输出进行运算输出神经元数量128之和
#lstm中隐状态c和h
print(fw_state.h.shape)#(10000, 256)
print(fw_state.c.shape)#(10000, 256)
print(bw_state.h.shape)#(10000, 128)
print(bw_state.c.shape)#(10000, 128)