multi-layer network

[zdarktknight]

This is a very good tutorial.

I am trying to extend it to multi-layer network by adding lstm_cell = tf.contrib.rnn.MultiRNNCell([lstm_cell] * 2, state_is_tuple=True)

But I have a valueError saying:

ValueError: Dimensions must be equal, but are 256 and 157 for 'rnn/while/rnn/multi_rnn_cell/cell_0/cell_0/lstm_cell/MatMul_1' (op: 'MatMul') with input shapes: [?,256], [157,512].

Could you please help me with it?

Thank you very much.