def test_default_values(self):
print("Testing init_eeg_acquisition with default values.")
client, server = init_eeg_acquisition(
self.parameters, self.save, server=True)
client.start_acquisition()
time.sleep(0.1)
client.stop_acquisition()
client.cleanup()
server.stop()
server.join()
self.assertTrue('raw_data.csv' in client._processor._filename)
self.assertEqual(client.device_info.name, 'DSI')
self.assertEqual(client.device_info.fs, 300)
def test_can_use_lsl(self):
print("Testing init_eeg_acquisition with LSL device")
params = self.parameters
params['acq_device'] = 'LSL'
client, server = init_eeg_acquisition(params, self.save, server=True)
with client:
time.sleep(0.1)
client.cleanup()
server.stop()
server.join()
self.assertEqual(client.device_info.name, 'LSL')
self.assertEqual(client.device_info.fs, 256)
def test_allows_customization(self):
print("Testing init_eeg_acquisition with custom values.")
f = 'foo.csv'
params = self.parameters
params['raw_data_name'] = f
params['acq_port'] = 9000
client, server = init_eeg_acquisition(params, self.save, server=True)
with client:
time.sleep(0.1)
client.cleanup()
server.stop()
server.join()
self.assertTrue(f in client._processor._filename)
self.assertEqual(client.device_info.name, 'DSI')
self.assertEqual(client.device_info.fs, 300)
def execute_task(task_type: dict, parameters: dict, save_folder: str) -> bool:
"""Excecute Task.
Executes the desired task by setting up the display window and
data acquisition, then passing on to the start_task funtion
which will initialize experiment.
Input:
parameters (dict): parameter dictionary
task_type (dict): type and mode of experiment
save_folder (str): path to save folder
"""
exp_type = ExperimentType(task_type['exp_type'])
fake = parameters['fake_data']
signal_model = None
language_model = None
filename = None
# Init EEG Model, if needed. Calibration Tasks Don't require probabilistic
# modules to be loaded.
if not fake and exp_type not in ExperimentType.calibration_tasks():
# Try loading in our signal_model and starting a langmodel(if enabled)
try:
signal_model, filename = load_signal_model()
except Exception as e:
logging.debug('Cannot load signal model. Exiting')
raise e
# if Language Model enabled init lm
if parameters['languagemodelenabled']:
language_model = init_language_model(parameters)
# Initialize Display Window
# We have to wait until after the prompt to load the signal model before
# displaying the window, otherwise in fullscreen mode this throws an error
display = init_display_window(parameters)
print_message(display, "Initializing...")
# Initialize DAQ
daq, server = init_eeg_acquisition(parameters, save_folder, server=fake)
# Start Task
try:
start_task(display,
daq,
exp_type,
parameters,
save_folder,
language_model=language_model,
signal_model=signal_model,
fake=fake,
auc_filename=filename)
# If exception, close all display and acquisition objects
except Exception as e:
_clean_up_session(display, daq, server)
raise e
return _clean_up_session(display, daq, server)
def execute_task(task_type: dict, parameters: dict, save_folder: str) -> bool:
"""Excecute Task.
Executes the desired task by setting up the display window and
data acquistion, then passing on to the start_task funtion
which will initialize experiment.
Input:
parameters (dict): parameter dictionary
task_type (dict): type and mode of experiment
save_folder (str): path to save folder
"""
fake = parameters['fake_data']
# Init EEG Model, if needed. Calibration Tasks Don't require probalistic
# modules to be loaded.
if task_type['exp_type'] > 1:
# Try loading in our classifier and starting a langmodel(if enabled)
try:
# EEG Model, Load in pre-trained classifier
if fake:
classifier = None
else:
classifier = load_classifier()
except Exception as e:
print("Cannot load EEG classifier. Exiting")
raise e
# if Language Model enabled and data not fake, init lm
if parameters['languagemodelenabled'] == 'true' \
and not fake:
try:
lmodel = init_language_model(parameters)
except:
print("Cannot init language model. Setting to None.")
lmodel = None
else:
lmodel = None
else:
classifier = None
lmodel = None
# Initialize DAQ
daq, server = init_eeg_acquisition(parameters, save_folder, server=fake)
# Initialize Display Window
display = init_display_window(parameters)
# Start Task
try:
start_task(display,
daq,
task_type,
parameters,
save_folder,
lmodel=lmodel,
classifier=classifier,
fake=fake)
# If exception, close all display and acquisition objects
except Exception as e:
_clean_up_session(display, daq, server)
raise e
return _clean_up_session(display, daq, server)