How can I use RNN-RBM for univariate prediction problem?

I used

music = tf.concat([music, x_out],0)

And

loop_vars = [time_steps, iterations, U, u_t, x, music]
[_, _, _, _, _, music] = tf.while_loop(lambda count, num_iter, *args: count < num_iter, generate_recurrence,
loop_vars, shape_invariants=[time_steps.get_shape(), iterations.get_shape(),
U.get_shape(), u_t.get_shape(),x.get_shape(), tf.TensorShape([None, None])])

Originally posted by @Guije2018 in #12 (comment)

Hello Dan,

in file rbm_chords.py
songs = get_songs('Pop_Music_Midi') #These songs have already been converted from midi to msgpack

How to convert midi to msgpack please?

Many thanks to all,
Dorje

0%|                                                                         | 0/3 [00:00<?, ?it/s]
Traceback (most recent call last):
  File "rnn_rbm_generate.py", line 48, in <module>
    main(sys.argv[1])
  File "rnn_rbm_generate.py", line 43, in main
    generated_music = sess.run(generate(300), feed_dict={x: song_primer}) #Prime the network with song primer and generate an original song
  File "/Users/roberthecimovic/Documents/Code/Github/poly-rnn/rnn_rbm.py", line 109, in generate
    x.get_shape(), tf.TensorShape([None, 780])])
TypeError: while_loop() got an unexpected keyword argument 'shape_invariants'
(venv) FAIL: 1

I had this working and then I decided to do a clean install of python and am getting stuck on the last generation step, when before this wasn't a problem at all. No idea what I'm getting wrong.

I tried running your code with slight tweaks which are no way related to this part but this doesnt seem to be working .

`
Uarr = tf.scan(rnn_recurrence, x, initializer=u0)

    U = Uarr[int(np.floor(prime_length / midi_manipulation.num_timesteps)), :, :]

    time_steps = tf.constant(1, tf.int32)

    iterations = tf.constant(num)

    u_t = tf.zeros([1, n_visible], tf.float32)

    music = tf.zeros([1, n_visible], tf.float32)


    loop_vars = [time_steps, iterations, U, u_t, x, music]
    [_, _, _, _, _, music] = tf.while_loop(lambda count, num_iter, *args: count < num_iter, generate_recurrence,
                                           loop_vars, shape_invariants=[time_steps, iterations,
                                                             U, u_t,x, tf.TensorShape([1, 15600])])

    return music`

Traceback (most recent call last): File "rnn_rbm_generate.py", line 45, in <module> main(sys.argv[1]) File "rnn_rbm_generate.py", line 42, in main midi_manipulation.write_song(new_song_path, generated_music) File "/home/gloomyghost/rbm/Music_RNN_RBM/midi_manipulation.py", line 15, in write_song noteStateMatrixToMidi(song, name=path) File "/home/gloomyghost/rbm/Music_RNN_RBM/midi_manipulation.py", line 138, in noteStateMatrixToMidi midi.write_midifile("{}.mid".format(name), pattern) File "/usr/local/lib/python2.7/dist-packages/midi/fileio.py", line 150, in write_midifile midifile = open(midifile, 'wb') IOError: [Errno 2] No such file or directory: 'music_outputs/0_Every Time We Touch - Chorus.midi.mid'

how to install and import midi module in anaconda

When I run weight_initializations.py I get the error:
AttributeError: module 'tensorflow.python.ops.control_flow_ops' has no attribute 'While'
Has anyone found a solution to this problem? I'm currently using Python3.6 and Tensorflow1.8.

Hi there, I'm having an issue. I changed the initial midi files given in the git, and now when I try to generate music it bring

python rnn_rbm_generate.py parameter_checkpoints/epoch_149.ckpt

Traceback (most recent call last):
File "rnn_rbm_generate.py", line 49, in
main(sys)
File "rnn_rbm_generate.py", line 32, in main
song_primer = midi_manipulation.get_song(primer_song)
File "/media/aram/Mordigan/Happy_Music_Generator-master/Music_RNN_RBM/midi_manipulation.py", line 19, in get_song
song = np.array(midiToNoteStateMatrix(path))
File "/media/aram/Mordigan/Happy_Music_Generator-master/Music_RNN_RBM/midi_manipulation.py", line 37, in midiToNoteStateMatrix
pattern = midi.read_midifile(midifile)
File "/usr/local/lib/python2.7/dist-packages/midi/fileio.py", line 156, in read_midifile
midifile = open(midifile, 'rb')
IOError: [Errno 2] No such file or directory: 'Pop_Music_Midi/Every Time We Touch - Chorus.midi'

There should not be such a file, and I cant get how to fix this, any help?

def generate(num, x=x, size_bt=size_bt, u0=u0, n_visible=n_visible, prime_length=100):
    """
        This function handles generating music. This function is one of the outputs of the build_rnnrbm function
        Args:
            num (int): The number of timesteps to generate
            x (tf.placeholder): The data vector. We can use feed_dict to set this to the music primer. 
            size_bt (tf.float32): The batch size
            u0 (tf.Variable): The initial state of the RNN
            n_visible (int): The size of the data vectors
            prime_length (int): The number of timesteps into the primer song that we use befoe beginning to generate music
        Returns:
            The generated music, as a tf.Tensor

    """
    Uarr = tf.scan(rnn_recurrence, x, initializer=u0)
    U = Uarr[int(np.floor(prime_length/midi_manipulation.num_timesteps)), :, :]
    [_, _, _, _, _, music] = control_flow_ops.while_loop(lambda count, num_iter, *args: count < num_iter,
                                                     generate_recurrence, [tf.constant(1, tf.int32), tf.constant(num), U,
                                                     tf.zeros([1, n_visible], tf.float32), x,
                                                     tf.zeros([1, n_visible],  tf.float32)])
    return music

Error:
Traceback (most recent call last):
File "rnn_rbm_generate.py", line 45, in
main(sys.argv[1])
File "rnn_rbm_generate.py", line 40, in main
generated_music = sess.run(generate(300), feed_dict={x: song_primer}) #Prime the network with song primer and generate an original song
File "C:\Users\HP\Downloads\music-github\Music_RNN_RBM\rnn_rbm.py", line 96, in generate
tf.zeros([1, n_visible], tf.float32)])
File "C:\Users\HP\Anaconda3\lib\site-packages\tensorflow\python\ops\control_flow_ops.py", line 3291, in while_loop
return_same_structure)
File "C:\Users\HP\Anaconda3\lib\site-packages\tensorflow\python\ops\control_flow_ops.py", line 3004, in BuildLoop
pred, body, original_loop_vars, loop_vars, shape_invariants)
File "C:\Users\HP\Anaconda3\lib\site-packages\tensorflow\python\ops\control_flow_ops.py", line 2939, in _BuildLoop
body_result = body(*packed_vars_for_body)
File "C:\Users\HP\Downloads\music-github\Music_RNN_RBM\rnn_rbm.py", line 74, in generate_recurrence
music = tf.concat(0, [music, x_out])
File "C:\Users\HP\Anaconda3\lib\site-packages\tensorflow\python\ops\array_ops.py", line 1122, in concat
tensor_shape.scalar())
File "C:\Users\HP\Anaconda3\lib\site-packages\tensorflow\python\framework\tensor_shape.py", line 848, in assert_is_compatible_with
raise ValueError("Shapes %s and %s are incompatible" % (self, other))
ValueError: Shapes (2, 1, 780) and () are incompatible

python2 rnn_rbm_generate.py /Users/humankhoobsirat/Desktop/MusicAI/Music_RNN_RBM-master/parameter_checkpoints/pretrained.ckpt

Traceback (most recent call last):
  File "rnn_rbm_generate.py", line 12, in <module>
    import rnn_rbm
  File "/Users/humankhoobsirat/Desktop/MusicAI/Music_RNN_RBM-master/rnn_rbm.py", line 9, in <module>
    import midi_manipulation
  File "/Users/humankhoobsirat/Desktop/MusicAI/Music_RNN_RBM-master/midi_manipulation.py", line 1, in <module>
    import midi
  File "/usr/local/lib/python2.7/site-packages/midi/__init__.py", line 2, in <module>
    from .events import *
  File "/usr/local/lib/python2.7/site-packages/midi/events.py", line 27
    class AbstractEvent(object,metaclass=AutoRegister):
                                        ^
SyntaxError: invalid syntax

I am stuck in first part of your README.md

I have download importing things from pandas to python-midi by pip
What is that I was missed?