pixel-rnn-tensorflow
pixel-rnn-tensorflow copied to clipboard
add two output maps in diagonal BiLSTM
output_state_bw_with_last_zeros = tf.concat(1, [output_state_bw_except_last, dummy_zeros])
should be
output_state_bw_with_last_zeros = tf.concat(1, [dummy_zeros, output_state_bw_except_last])
yes, I think you are right! In https://github.com/igul222/pixel_rnn theano code just do like that.
yesyes, I think so!!!