def task_features():
"""Step 2: Generate features (=fMRI regressors) from raw_features (csv file with 3 columns onset-amplitude-duration) by convolution with an hrf kernel."""
input_data_type = 'raw-features'
output_data_type = 'features'
extension = '.csv'
source = 'fMRI'
for language in languages:
for model in models:
output_parent_folder = get_output_parent_folder(
source, output_data_type, language, model)
input_parent_folder = get_output_parent_folder(
source, input_data_type, language, model)
dependencies = [
get_path2output(input_parent_folder, input_data_type, language,
model, run_name, extension)
for run_name in run_names
]
targets = [
get_path2output(output_parent_folder, output_data_type,
language, model, run_name, extension)
for run_name in run_names
]
yield {
'name':
model,
'file_dep': ['features.py'] + dependencies,
'targets':
targets,
'actions': [
'python features.py --tr {} --language {} --model {} '.
format(tr, language, model) + optional + optional_parallel
],
}
def task_design_matrices():
"""Step 3: Generate design matrices from features in a given language."""
input_data_type = 'features'
output_data_type = 'design-matrices'
extension = '.csv'
source = 'fMRI'
for language in languages:
for models in aggregated_models:
output_parent_folder = get_output_parent_folder(
source, output_data_type, language, models)
dependencies = []
for model in models.split('+'):
input_parent_folder = get_output_parent_folder(
source, input_data_type, language, model)
dependencies += [
get_path2output(input_parent_folder, input_data_type,
language, model, run_name, extension)
for run_name in run_names
]
targets = [
get_path2output(output_parent_folder, output_data_type,
language, models, run_name, extension)
for run_name in run_names
]
yield {
'name':
models,
'file_dep': ['design-matrices.py'] + dependencies,
'targets':
targets,
'actions': [
'python design-matrices.py --language {} --models {} '.
format(language, ' '.join(models.split('+'))) + optional
],
}
def task_raw_features():
"""Step 1: Generate raw features from raw data (text, wave) model predictions."""
extension = '.csv'
output_data_type = 'raw-features'
source = 'fMRI'
for language in languages:
for model in models:
if model.split('_')[0] in ['rms', 'f0', 'mfcc']:
input_data_type = 'wave'
extension_input = '.wav'
else:
input_data_type = 'text'
extension_input = '.txt'
input_parent_folder = join(
paths.path2data,
'{0}/{1}/{2}'.format(input_data_type, language,
params.get_category(model)))
dependencies = [
join(
input_parent_folder, '{0}_{1}_{2}_{3}'.format(
input_data_type, language, params.get_category(model),
run_name) + extension_input) for run_name in run_names
]
output_parent_folder = get_output_parent_folder(
source, output_data_type, language, model)
targets = [
get_path2output(output_parent_folder, output_data_type,
language, model, run_name, extension)
for run_name in run_names
]
yield {
'name':
model,
'file_dep': ['raw_features.py'] + dependencies,
'targets':
targets,
'actions': [
'python raw_features.py --language {} --model_name {} --model_category {} '
.format(language, model, params.get_category(model)) +
optional + optional_parallel
],
}
action='store_true',
help="Precise if we overwrite existing files")
parser.add_argument("--parallel",
default=False,
action='store_true',
help="Precise if we want to run code in parallel")
args = parser.parse_args()
input_data_type = 'raw-features'
output_data_type = 'features'
extension = '.csv'
source = 'fMRI'
model = args.model
output_parent_folder = get_output_parent_folder(source, output_data_type,
args.language, model)
check_folder(
output_parent_folder
) # check if the output_parent_folder exists and create it if not
raw_features = get_data(args.language,
input_data_type,
model=model,
source='fMRI')
if not args.parallel:
for i, run in enumerate(raw_features):
compute_features(run, output_parent_folder, output_data_type,
args.language, model, extension, args.tr,
args.overwrite)
else: