Comment Comment by yablak If you're using TF 1.2.1 or above, you want to replace this line:cell = tf.contrib.rnn.MultiRNNCell([drop_cell] * num_layers)with:cell = tf.contrib.rnn.MultiRNNCell([create_cell() for _ in range(num_layers)])otherwise you'll be using the same weights for each LSTM layer and your model is less expressive. parent yablak 9 years ago
cell = tf.contrib.rnn.MultiRNNCell([drop_cell] * num_layers)
with:
cell = tf.contrib.rnn.MultiRNNCell([create_cell() for _ in range(num_layers)])
otherwise you'll be using the same weights for each LSTM layer and your model is less expressive.