def __init__(self, dataset_dir='../../datasets'):
filename = 'Nottingham.zip'
source = 'http://www-etud.iro.umontreal.ca/~boulanni/Nottingham.zip'
super(Nottingham, self).__init__(filename=filename, source=source, dataset_dir=dataset_dir)
# now the file path has been installed to self.dataset_locations directory
# grab the appropriate filenames
train_filenames = os.path.join(self.dataset_location, 'Nottingham', 'train', '*.mid')
valid_filenames = os.path.join(self.dataset_location, 'Nottingham', 'valid', '*.mid')
test_filenames = os.path.join(self.dataset_location, 'Nottingham', 'test', '*.mid')
# glob the files
train_files = glob.glob(train_filenames)
valid_files = glob.glob(valid_filenames)
test_files = glob.glob(test_filenames)
# grab the datasets from midireading the files
train_datasets = [midiread(f, r=(21, 109), dt=0.3).piano_roll.astype(theano.config.floatX) for f in train_files]
valid_datasets = [midiread(f, r=(21, 109), dt=0.3).piano_roll.astype(theano.config.floatX) for f in valid_files]
test_datasets = [midiread(f, r=(21, 109), dt=0.3).piano_roll.astype(theano.config.floatX) for f in test_files]
# get the data shapes
self.train_shapes = [train.shape for train in train_datasets]
self.valid_shapes = [valid.shape for valid in valid_datasets]
self.test_shapes = [test.shape for test in test_datasets]
# put them into shared variables
log.debug('Putting Nottingham into theano shared variables')
self.train = dataset_shared(numpy.concatenate(train_datasets), name='nottingham_train', borrow=True)
self.valid = dataset_shared(numpy.concatenate(valid_datasets), name='nottingham_valid', borrow=True)
self.test = dataset_shared(numpy.concatenate(test_datasets), name='nottingham_test', borrow=True)
def __init__(self, path='datasets/Nottingham',
source='http://www-etud.iro.umontreal.ca/~boulanni/Nottingham.zip',
train_filter='.*train.*',
valid_filter='.*valid.*',
test_filter='.*test.*', ):
super(Nottingham, self).__init__(path=path, source=source,
train_filter=train_filter,
valid_filter=valid_filter,
test_filter=test_filter)
# grab the datasets from midireading the files
train_datasets = [
midiread(f, r=(21, 109), dt=0.3).piano_roll.astype(config.floatX)
for f in find_files(self.path, train_filter)
]
valid_datasets = [
midiread(f, r=(21, 109), dt=0.3).piano_roll.astype(config.floatX)
for f in find_files(self.path, valid_filter)
]
test_datasets = [
midiread(f, r=(21, 109), dt=0.3).piano_roll.astype(config.floatX)
for f in find_files(self.path, test_filter)
]
self.train_inputs = numpy.concatenate(train_datasets)
self.train_targets = None
self.valid_inputs = numpy.concatenate(valid_datasets)
self.valid_targets = None
self.test_inputs = numpy.concatenate(test_datasets)
self.test_targets = None
def __init__(self, path='datasets/MuseData',
source='http://www-etud.iro.umontreal.ca/~boulanni/MuseData.zip',
train_filter='.*train.*',
valid_filter='.*valid.*',
test_filter='.*test.*', ):
super(MuseData, self).__init__(path=path, source=source,
train_filter=train_filter,
valid_filter=valid_filter,
test_filter=test_filter)
# grab the datasets from midireading the files
train_datasets = [
midiread(f, r=(21, 109), dt=0.3).piano_roll.astype(theano.config.floatX)
for f in find_files(self.path, train_filter)
]
valid_datasets = [
midiread(f, r=(21, 109), dt=0.3).piano_roll.astype(theano.config.floatX)
for f in find_files(self.path, valid_filter)
]
test_datasets = [
midiread(f, r=(21, 109), dt=0.3).piano_roll.astype(theano.config.floatX)
for f in find_files(self.path, test_filter)
]
self.train_inputs = numpy.concatenate(train_datasets)
self.train_targets = None
self.valid_inputs = numpy.concatenate(valid_datasets)
self.valid_targets = None
self.test_inputs = numpy.concatenate(test_datasets)
self.test_targets = None
def __init__(self, dataset_dir='../../datasets'):
"""
Parameters
----------
dataset_dir : str
The `dataset_dir` parameter to a ``FileDataset``.
"""
log.debug("Loading MuseData midi dataset...")
filename = 'MuseData.zip'
source = 'http://www-etud.iro.umontreal.ca/~boulanni/MuseData.zip'
super(MuseData, self).__init__(filename=filename,
source=source,
dataset_dir=dataset_dir)
# now the file path has been installed to self.dataset_locations directory
# grab the appropriate filenames
train_filenames = os.path.join(self.dataset_location, 'MuseData',
'train', '*.mid')
valid_filenames = os.path.join(self.dataset_location, 'MuseData',
'valid', '*.mid')
test_filenames = os.path.join(self.dataset_location, 'MuseData',
'test', '*.mid')
# glob the files
train_files = glob.glob(train_filenames)
valid_files = glob.glob(valid_filenames)
test_files = glob.glob(test_filenames)
# grab the datasets from midireading the files
train_datasets = [
midiread(f, r=(21, 109),
dt=0.3).piano_roll.astype(theano.config.floatX)
for f in train_files
]
valid_datasets = [
midiread(f, r=(21, 109),
dt=0.3).piano_roll.astype(theano.config.floatX)
for f in valid_files
]
test_datasets = [
midiread(f, r=(21, 109),
dt=0.3).piano_roll.astype(theano.config.floatX)
for f in test_files
]
# get the data shapes
self.train_shapes = [train.shape for train in train_datasets]
self.valid_shapes = [valid.shape for valid in valid_datasets]
self.test_shapes = [test.shape for test in test_datasets]
# put them into shared variables
log.debug('Putting MuseData into theano shared variables')
self.train = dataset_shared(numpy.concatenate(train_datasets),
name='muse_train',
borrow=True)
self.valid = dataset_shared(numpy.concatenate(valid_datasets),
name='muse_valid',
borrow=True)
self.test = dataset_shared(numpy.concatenate(test_datasets),
name='muse_test',
borrow=True)