def test_function_main(self, mock_dataset, mock_process, mock_write,
mock_update):
"""main should call the right functions"""
sys.argv = [
'', "dir_1", '-o', 'dir_2', '-n', 'nersc_', '-w', '700', '-s',
'700', '-r', '50'
]
with mock.patch.object(
Archive, "get_unprocessed_files") as mock_get_unprocessed:
mock_get_unprocessed.side_effect = __class__.side_effect_function(
Archive)
main()
mock_get_unprocessed.assert_called_once()
mock_process.assert_called_once_with('foo', 'fake_file')
mock_write.assert_called_once()
mock_update.assert_called_once()
def generate(self):
# How to pass arguments to build_dataset.py
parser = self.parser
# args = parser.parse_args([])
# build_dataset.main(args)
# This gives us the full A matrix
files_to_read = ['data/2017CHR_CSV_Analytic_Data.csv']
cols_per_file_to_read = [
'[ 6, 11, 16, 21, 26, 31, 36, 41, 46, 51, 56, 61, 66, 71, 76, 81,\
87, 93, 99, 104, 109, 114, 119, 124, 129, 134, 139, 144, 149, \
154, 159, 164, 169, 174, 179, 184, 189, 194, 199, 204, 209, 214, \
219, 224, 234, 239, 244, 249, 254, 259, 265, 270, 275, 280, 285, \
290, 295, 302, 307, 311, 316, 321, 326, 331, 336, 341, 346, 351]'
]
num_cols_per_file = [2]
A = []
for file_num in range(len(files_to_read)):
args = parser.parse_args([
'--filename', files_to_read[file_num], '--columns_to_read',
cols_per_file_to_read[file_num], '--is_b_vec', 'False'
])
A_new = build_dataset.main(args)
num_cols = num_cols_per_file[file_num]
if file_num == 0:
A = A_new
else:
for i in range(num_cols):
A.append(A_new[i])
# pdb.set_trace()
# This gives us the full b-vector
args = parser.parse_args([
'--filename', 'data/2017CHR_CSV_Analytic_Data.csv',
'--columns_to_read', '[229]', '--is_b_vec', 'False'
]) #see cols_per_file_to_read
y = build_dataset.main(args)
# pdb.set_trace()
return A, y
def main():
train, test = build_dataset.main()
train, test = build_dataset.normalize_datasets(train, test)
train = over_under_sampling(train)
logger.info('Applying Logistic Regression')
logistic_model(train, test)
logger.info('Applying Random Forest')
random_forest_model(train, test)
logger.info('Applying Gaussian Naive Bayes')
naive_bayes_model(train, test)
def main():
"""
Execute generic classification methods on DNA methylation data
:return: metrics of each classifier
"""
train, test = build_dataset.main()
#train, test = build_dataset.normalize_datasets(train, test)
#logger.info('Applying Logistic Regression')
#logistic_model(train, test)
logger.info('Applying Random Forest')
random_forest_model(train, test)
def predict():
'''
For rendering results on HTML GUI
'''
list = [
"BRAC: Breast Invasive Carcinoma", "LUAD: Lung Adenocarcinoma",
"BLAC Urothelial Bladder Carcinoma", "PRAD: Prostate Adenocarcinoma",
"LUSC: Lung Squamous Cell Carcinoma", "THCA: Thyroid Cancer",
"HNSC: Head-Neck Squamous Cell Carcinoma"
]
train, test = build_dataset.main()
k = random_forest_model(train, test)
output = random.choice(list)
return render_template('index.html',
prediction_text='OUTPUT = {}'.format(output))
import numpy as np
import os
from build_dataset import main, reconstruct_full_sequence, min_max_from_folder
from glob import glob
dataset = "d:/data/thesis_model2/MIDI_tests/session*.mid"
files = glob(dataset)
out_dir = "d:/data/thesis_model2/MIDI_tests/trims/"
if not os.path.exists(out_dir):
os.mkdir(out_dir)
dataset = {
"NR_BARS": 4,
"BAR_LEN": 96
}
main(files, out_dir, dataset)
# check first seq. is the same as the one from loading the file
min_pitch, max_pitch = min_max_from_folder(out_dir)
first_sequence = np.load(glob(os.path.join(out_dir, '*.npy'))[0])
reconstructed_seq = reconstruct_full_sequence(first_sequence, min_pitch, max_pitch)
orig_first_sequence = pypianoroll.Multitrack(files[0]).tracks[0]
orig_first_sequence.binarize()
orig_first_sequence = orig_first_sequence.pianoroll[:first_sequence.shape[0]]
assert False not in np.unique(orig_first_sequence == reconstructed_seq)
print('ALL GOOD')