CODEWITHSUNDEEP
tensorflowrecurrent-neural-network
Saeed
Saeed

Reputation: 357

What is the equivalent of tf.nn.rnn in new versions of TensorFlow?

I used to create the RNN network, in version 0.8 of TensorFlow, using: 

from tensorflow.python.ops import rnn

# Define a lstm cell with tensorflow
lstm_cell = rnn_cell.BasicLSTMCell(n_hidden, forget_bias=1.0)

# Get lstm cell output
outputs, states = rnn.rnn(cell=lstm_cell, inputs=x, dtype=tf.float32)

rnn.rnn() is not available anymore, and it sounds it has been moved to tf.contrib. What is the exact code to create RNN network out of a BasicLSTMCell?

Or, in the case that I have an stacked LSTM, 

lstm_cell = tf.contrib.rnn.BasicLSTMCell(hidden_size, forget_bias=0.0)
stacked_lstm = tf.contrib.rnn.MultiRNNCell([lstm_cell] * num_layers)
outputs, new_state =  tf.nn.rnn(stacked_lstm, inputs, initial_state=_initial_state)

What is the replacement for tf.nn.rnn in new versions of TensorFlow?

Upvotes: 12

Views: 6683

Answers (2)

ruoho ruotsi
ruoho ruotsi

Reputation: 1313

tf.nn.rnn is equivalent to tf.nn.static_rnn.

Note: before version 1.2 of TensorFlow, the namespace tf.nn.static_rnn did not exist, but only tf.contrib.rnn.static_rnn (which is now an alias for tf.nn.static_rnn).

Upvotes: 13

Franck Dernoncourt
Franck Dernoncourt

Reputation: 83187

You should use tf.nn.dynamic_rnn.

FYI: What is the upside of using tf.nn.rnn instead of tf.nn.dynamic_rnn in TensorFlow?

Upvotes: 2

Related Questions