Path: blob/master/Sequence Models/Week 1/Jazz improvisation with LSTM/__pycache__/data_utils.cpython-36.pyc
14462 views
3
ڰqZ� � @ s� d dl T d dlT d dlmZ ed�\ZZee�\ZZ Z
Z e e
e��ZdZejd�Zejdef�Zejdef�Zdd� Zeee e
e d d
d fd d
�Zeeefdd�ZdS )� )�*)�to_categoricalzdata/original_metheny.mid�@ � �N c C sH t d�\} }t|�\}}}}tt|��}t||dd�\}}}||||fS )Nzdata/original_metheny.mid�< � )�get_musical_data�get_corpus_data�len�set�data_processing) �chords�abstract_grammars�corpus�tones�
tones_indices�
indices_tones�N_tones�X�Y� r �C/home/jovyan/work/Week 1/Jazz improvisation with LSTM/data_utils.py�load_music_utils
s
r �
i� g �?c s� t j� } d}
ttt�d �} td� �xDtd| �D �]4} t j� }
x"t| D ]}|
j|j d |� qJW t
| �\}}t |j � �}� fdd�|D �}d}x(tt|�d �D ]}||| d 7 }q�W ||d 7 }|j
d
d �j
d d �}t|�}t||
�}t|�}t|�}td
tdd� |D ��| f � x |D ]}| j|
|j |� �q"W x |
D ]}| j|
|j |� �qDW |
d7 }
q2W | jdtjdd�� tjj| �}|jdd� |j� td� |j� | S )a�
Generates music using a model trained to learn musical patterns of a jazz soloist. Creates an audio stream
to save the music and play it.
Arguments:
model -- Keras model Instance, output of djmodel()
corpus -- musical corpus, list of 193 tones as strings (ex: 'C,0.333,<P1,d-5>')
abstract_grammars -- list of grammars, on element can be: 'S,0.250,<m2,P-4> C,0.250,<P4,m-2> A,0.250,<P4,m-2>'
tones -- set of unique tones, ex: 'A,0.250,<M2,d-4>' is one element of the set.
tones_indices -- a python dictionary mapping unique tone (ex: A,0.250,< m2,P-4 >) into their corresponding indices (0-77)
indices_tones -- a python dictionary mapping indices (0-77) into their corresponding unique tone (ex: A,0.250,< m2,P-4 >)
Tx -- integer, number of time-steps used at training time
temperature -- scalar value, defines how conservative/creative the model is when generating music
Returns:
predicted_tones -- python list containing predicted tones
g � z2Predicting new values for different set of chords.r � c s g | ] }� | �qS r r )�.0�p)r r r �
<listcomp>? s z"generate_music.<locals>.<listcomp>zC,0.25 � z Az Cz Xz]Generated %s sounds using the predicted values for the set of chords ("%s") and after pruningc S s g | ]}t |tj�r|�qS r )�
isinstance�note�Note)r �kr r r r X s g @� )�numberzoutput/my_music.midi�wbz5Your generated music is saved in output/my_music.midi�����)�stream�Stream�intr r �print�range�Voice�insert�offset�predict_and_sample�list�squeeze�replace�
prune_grammar�unparse_grammar� prune_notes�clean_up_notes�tempo�
MetronomeMark�midi� translate�streamToMidiFile�open�write�close)�inference_modelr r r r r ZT_y� max_triesZ diversityZ
out_streamZ curr_offsetZ
num_chords�iZ curr_chords�j�_�indices�predZpredicted_tonesr$ Zsounds�m�mc�mfr )r r �generate_music sB
rK c C s2 | j |||g�}tj|dd�}t|dd�}||fS )a�
Predicts the next value of values using the inference model.
Arguments:
inference_model -- Keras model instance for inference time
x_initializer -- numpy array of shape (1, 1, 78), one-hot vector initializing the values generation
a_initializer -- numpy array of shape (1, n_a), initializing the hidden state of the LSTM_cell
c_initializer -- numpy array of shape (1, n_a), initializing the cell state of the LSTM_cel
Ty -- length of the sequence you'd like to generate.
Returns:
results -- numpy-array of shape (Ty, 78), matrix of one-hot vectors representing the values generated
indices -- numpy-array of shape (Ty, 1), matrix of indices representing the values generated
r )�axisr )� num_classesr( )�predict�np�argmaxr )rA �
x_initializer�
a_initializer�
c_initializerrG rF �resultsr r r r1 s s r1 N)r r r )� music_utils�
preprocess� keras.utilsr r r r r
r r r r r r r �n_arO �zerosrQ rR rS r rK r1 r r r r �<module> s
^