def midi_files_to_sequence_proto(midi_files, batch_size, input_size):
all_sequences = filter(lambda x: x != None, [
midi_file_to_sequence_proto(midi_file, batch_size, input_size)
for midi_file in midi_files
])
examples = []
#print("All our events are: ", all_events)
unit = melody_pipelines.MelodyExtractor(min_bars=1,
min_unique_pitches=1,
gap_bars=1)
melody_extractor = melody_pipelines.MelodyExtractor(
min_bars=7,
min_unique_pitches=5,
gap_bars=1.0,
ignore_polyphonic_notes=False)
quantizer = pipelines_common.Quantizer(steps_per_quarter=4)
one_hot_encoding = magenta.music.OneHotEventSequenceEncoderDecoder(
magenta.music.MelodyOneHotEncoding(0, 128))
for note_sequence in all_sequences:
#quantized_sequence = quantize_note_sequence(
# note_sequence, steps_per_quarter=1)
#quantized_sequence = mm.quantize_note_sequence(note_sequence, steps_per_quarter=1)
#outputs = unit.transform(quantized_sequence)
#single_quant = quantizer.transform(note_sequence)
#print("single quant seq is: ", single_quant)
# quantized = quantizer.transform(note_sequence)
#print("single quant is: ", single_quant)
single_quant = note_sequence
melody = melody_extractor.transform(single_quant)
print("outputs/melody is ", outputs)
#melody = outputs
# self.config.min_note,
# self.config.max_note,
# self.config.transpose_to_key)
melody.squash(0, 128, 0)
one_hot = one_hot_encoding.encode(melody)
examples.append(one_hot)
#seq_example =
#seq_example = make_sequence_example(inputs, labels)
#examples.append(seq_example)
# melody = melodies_lib.Melody(events.notes)
# melody.squash(0,128,0)
# examples.append(self.med.encode(melody))
print("done with a batch!!!")
return get_padded_batch(examples, batch_size, input_size)
def testMelodyRNNPipeline(self):
note_sequence = magenta.common.testing_lib.parse_test_proto(
music_pb2.NoteSequence, """
time_signatures: {
numerator: 4
denominator: 4}
tempos: {
qpm: 120}""")
magenta.music.testing_lib.add_track_to_sequence(
note_sequence, 0, [(12, 100, 0.00, 2.0), (11, 55, 2.1, 5.0),
(40, 45, 5.1, 8.0), (55, 120, 8.1, 11.0),
(53, 99, 11.1, 14.1)])
quantizer = note_sequence_pipelines.Quantizer(steps_per_quarter=4)
melody_extractor = melody_pipelines.MelodyExtractor(
min_bars=7,
min_unique_pitches=5,
gap_bars=1.0,
ignore_polyphonic_notes=False)
one_hot_encoding = magenta.music.OneHotEventSequenceEncoderDecoder(
magenta.music.MelodyOneHotEncoding(self.config.min_note,
self.config.max_note))
quantized = quantizer.transform(note_sequence)[0]
melody = melody_extractor.transform(quantized)[0]
melody.squash(self.config.min_note, self.config.max_note,
self.config.transpose_to_key)
one_hot = one_hot_encoding.encode(melody)
expected_result = {'training_melodies': [one_hot], 'eval_melodies': []}
pipeline_inst = melody_rnn_pipeline.get_pipeline(self.config,
eval_ratio=0.0)
result = pipeline_inst.transform(note_sequence)
self.assertEqual(expected_result, result)
def testMelodyExtractor(self):
note_sequence = common_testing_lib.parse_test_proto(
music_pb2.NoteSequence, """
time_signatures: {
numerator: 4
denominator: 4}
tempos: {
qpm: 60}""")
testing_lib.add_track_to_sequence(note_sequence, 0, [(12, 100, 2, 4),
(11, 1, 6, 7)])
testing_lib.add_track_to_sequence(note_sequence, 1, [(12, 127, 2, 4),
(14, 50, 6, 8)])
quantized_sequence = sequences_lib.quantize_note_sequence(
note_sequence, steps_per_quarter=1)
expected_events = [[
NO_EVENT, NO_EVENT, 12, NO_EVENT, NOTE_OFF, NO_EVENT, 11
], [
NO_EVENT, NO_EVENT, 12, NO_EVENT, NOTE_OFF, NO_EVENT, 14, NO_EVENT
]]
expected_melodies = []
for events_list in expected_events:
melody = melodies_lib.Melody(events_list,
steps_per_quarter=1,
steps_per_bar=4)
expected_melodies.append(melody)
unit = melody_pipelines.MelodyExtractor(min_bars=1,
min_unique_pitches=1,
gap_bars=1)
self._unit_transform_test(unit, quantized_sequence, expected_melodies)
def get_pipeline(config):
"""Returns the Pipeline instance which creates the RNN dataset.
Args:
config: A MelodyRnnConfig object.
Returns:
A pipeline.Pipeline instance.
"""
quantizer = pipelines_common.Quantizer(steps_per_quarter=4)
melody_extractor = melody_pipelines.MelodyExtractor(
min_bars=7,
max_steps=512,
min_unique_pitches=5,
gap_bars=1.0,
ignore_polyphonic_notes=False)
encoder_pipeline = EncoderPipeline(config)
partitioner = pipelines_common.RandomPartition(
tf.train.SequenceExample, ['eval_melodies', 'training_melodies'],
[FLAGS.eval_ratio])
dag = {
quantizer: dag_pipeline.Input(music_pb2.NoteSequence),
melody_extractor: quantizer,
encoder_pipeline: melody_extractor,
partitioner: encoder_pipeline,
dag_pipeline.Output(): partitioner
}
return dag_pipeline.DAGPipeline(dag)
def get_pipeline(config, eval_ratio):
"""Returns the Pipeline instance which creates the RNN dataset.
Args:
config: A MelodyRnnConfig object.
eval_ratio: Fraction of input to set aside for evaluation set.
Returns:
A pipeline.Pipeline instance.
"""
partitioner = pipelines_common.RandomPartition(
music_pb2.NoteSequence,
['eval_melodies', 'training_melodies'],
[eval_ratio])
dag = {partitioner: dag_pipeline.DagInput(music_pb2.NoteSequence)}
for mode in ['eval', 'training']:
time_change_splitter = note_sequence_pipelines.TimeChangeSplitter(
name='TimeChangeSplitter_' + mode)
quantizer = note_sequence_pipelines.Quantizer(
steps_per_quarter=config.steps_per_quarter, name='Quantizer_' + mode)
melody_extractor = melody_pipelines.MelodyExtractor(
min_bars=7, max_steps=512, min_unique_pitches=5,
gap_bars=1.0, ignore_polyphonic_notes=True,
name='MelodyExtractor_' + mode)
encoder_pipeline = EncoderPipeline(config, name='EncoderPipeline_' + mode)
dag[time_change_splitter] = partitioner[mode + '_melodies']
dag[quantizer] = time_change_splitter
dag[melody_extractor] = quantizer
dag[encoder_pipeline] = melody_extractor
dag[dag_pipeline.DagOutput(mode + '_melodies')] = encoder_pipeline
return dag_pipeline.DAGPipeline(dag)
def get_pipeline(config, eval_ratio=0.0):
partitioner = pipelines_common.RandomPartition(
music_pb2.NoteSequence, ['eval_melodies', 'training_melodies'],
[eval_ratio])
dag = {partitioner: dag_pipeline.DagInput(music_pb2.NoteSequence)}
for mode in ['eval', 'training']:
time_change_splitter = note_sequence_pipelines.TimeChangeSplitter(
name='TimeChangeSplitter_' + mode)
repeat_sequence = RepeatSequence(min_duration=16,
name='RepeatSequence_' + mode)
transposition_pipeline = note_sequence_pipelines.TranspositionPipeline(
(0, ), name='TranspositionPipeline_' + mode)
quantizer = note_sequence_pipelines.Quantizer(
steps_per_quarter=config.steps_per_quarter,
name='Quantizer_' + mode)
melody_extractor = melody_pipelines.MelodyExtractor(
min_bars=7,
max_steps=512,
min_unique_pitches=5,
gap_bars=1.0,
ignore_polyphonic_notes=True,
name='MelodyExtractor_' + mode)
encoder_pipeline = EncoderPipeline(config,
name='EncoderPipeline_' + mode)
dag[time_change_splitter] = partitioner[mode + '_melodies']
dag[repeat_sequence] = time_change_splitter
dag[quantizer] = repeat_sequence
dag[transposition_pipeline] = quantizer
dag[melody_extractor] = transposition_pipeline
dag[encoder_pipeline] = melody_extractor
dag[dag_pipeline.DagOutput(mode + '_melodies')] = encoder_pipeline
return dag_pipeline.DAGPipeline(dag)
def testMelodyRNNPipeline(self):
FLAGS.eval_ratio = 0.0
note_sequence = magenta.common.testing_lib.parse_test_proto(
music_pb2.NoteSequence,
"""
time_signatures: {
numerator: 4
denominator: 4}
tempos: {
qpm: 120}""")
magenta.music.testing_lib.add_track_to_sequence(
note_sequence, 0,
[(12, 100, 0.00, 2.0), (11, 55, 2.1, 5.0), (40, 45, 5.1, 8.0),
(55, 120, 8.1, 11.0), (53, 99, 11.1, 14.1)])
quantizer = pipelines_common.Quantizer(steps_per_quarter=4)
melody_extractor = melody_pipelines.MelodyExtractor(
min_bars=7, min_unique_pitches=5, gap_bars=1.0,
ignore_polyphonic_notes=False)
one_hot_encoder = magenta.music.OneHotMelodyEncoderDecoder(0, 127, 0)
quantized = quantizer.transform(note_sequence)[0]
print quantized.tracks
melody = melody_extractor.transform(quantized)[0]
one_hot = one_hot_encoder.squash_and_encode(melody)
print one_hot
expected_result = {'training_melodies': [one_hot], 'eval_melodies': []}
pipeline_inst = melody_rnn_create_dataset.get_pipeline(one_hot_encoder)
result = pipeline_inst.transform(note_sequence)
self.assertEqual(expected_result, result)
def testMelodyExtractor(self):
quantized_sequence = sequences_lib.QuantizedSequence()
quantized_sequence.steps_per_quarter = 1
testing_lib.add_quantized_track_to_sequence(quantized_sequence, 0,
[(12, 100, 2, 4),
(11, 1, 6, 7)])
testing_lib.add_quantized_track_to_sequence(quantized_sequence, 1,
[(12, 127, 2, 4),
(14, 50, 6, 8)])
expected_events = [[
NO_EVENT, NO_EVENT, 12, NO_EVENT, NOTE_OFF, NO_EVENT, 11
], [
NO_EVENT, NO_EVENT, 12, NO_EVENT, NOTE_OFF, NO_EVENT, 14, NO_EVENT
]]
expected_melodies = []
for events_list in expected_events:
melody = melodies_lib.Melody(events_list,
steps_per_quarter=1,
steps_per_bar=4)
expected_melodies.append(melody)
unit = melody_pipelines.MelodyExtractor(min_bars=1,
min_unique_pitches=1,
gap_bars=1)
self._unit_transform_test(unit, quantized_sequence, expected_melodies)
