What are the meanings of the variables which get from RNN

[zxzStar]

Hi, Aymeric Damien, thank you so much for this examples, it's great and helpful.
I use your RNN (recurrent_network.py) to train my dataset and get a good result. I use this way to get all variables:

def RNN(x, weights, biases):
    x = tf.transpose(x, [1, 0, 2])
    x = tf.reshape(x, [-1, n_input])
    x = tf.split(0, n_steps, x)
    lstm_cell = rnn_cell.LSTMCell(n_hidden, use_peepholes=False, forget_bias=1.0)
    outputs, states = rnn.rnn(lstm_cell, x, dtype=tf.float32)
    return tf.matmul(outputs[-1], weights['out']) + biases['out']
all_variables = tf.get_collection(tf.GraphKeys.VARIABLES)

all_variable = sess.run(all_variables)
print all_variable

Then I can get 14 arrays which are all variables of the RNN. But I don't know their meanings. Could you help me? Thanks a lot!

My network parameters:

n_input = 140 
n_hidden = 32 
n_classes = 2 

The dimensions of 14 variables are as follows:

[0] 32*2（hidden->output?）
[1] 2
[2] 172*128  (n_input+n_hidden)*(n_hidden*4)  
[3] 128
[4] 1
[5] 1
[6] 32*2
[7] 32*2
[8] 2
[9] 2
[10] 172*128
[11] 172*128
[12] 128
[13] 128

Thanks a lot!