Inputs are not sequence wth RNN and TensorFlow
I have a very simple lstm code with tensorflow and python where my code is
output = tf.nn.rnn(tf.nn.rnn_cell.BasicLSTMCell(10), input_flattened, initial_state=tf.placeholder("float", [None, 20]))
where my smoothed input is form [?, 5, 22501]
I am getting an error TypeError: inputs must be a sequence
on the state
lstm parameter and I am ripping my hair out trying to figure out why it is giving me this error. Any help would be greatly appreciated.
source to share
I think that when you use the tf.nn.rnn function it expects a list of tensors, not just one tensor. You have to unpack the input in the time direction so that it is a list of tensors of the form [?, 22501]. You can also use tf.nn.dynamic_rnn which I think can handle this unpacking for you.
source to share