def mce_def(info): # noqa
mce_def = MCE()
parent = FileSource()
parent.mne_info = info
parent.output = np.random.rand(info['nchan'], 1)
mce_def.parent = parent
return mce_def
def inv_model_def(info): # noqa
inv_model_def = InverseModel()
parent = FileSource()
parent.mne_info = info
parent.output = np.random.rand(info['nchan'], 1)
inv_model_def.parent = parent
return inv_model_def
def beamformer_default(info): # noqa
beamformer_default = Beamformer()
parent = FileSource()
parent.mne_info = info
parent.output = np.random.rand(info['nchan'], 1)
beamformer_default.parent = parent
return beamformer_default
def env_extractor(info, data_path): # noqa
env_extractor = EnvelopeExtractor()
env_extractor.mne_info = info
N_SEN = len(info['ch_names'])
env_extractor.input = np.random.rand(N_SEN)
parent = FileSource(data_path)
parent.output = np.random.rand(info['nchan'], 1)
parent.mne_info = info
env_extractor.parent = parent
return env_extractor
def lin_filter(info, data_path): # noqa
lin_filter = LinearFilter()
lin_filter.mne_info = info
N_SEN = len(info['ch_names'])
lin_filter.input = np.random.rand(N_SEN)
parent = FileSource(data_path)
parent.output = np.random.rand(info['nchan'], 1)
parent.mne_info = info
lin_filter.parent = parent
return lin_filter
def coh_computer(info, data_path): # noqa
coh_computer = Coherence()
coh_computer.mne_info = info
N_SEN = len(info['ch_names'])
coh_computer.input = np.random.rand(N_SEN)
parent = FileSource(data_path)
parent.output = np.random.rand(info['nchan'], 1)
parent.mne_info = info
coh_computer.parent = parent
return coh_computer
def ica_rejector(info, data_path): # noqa
ica_rejector = ICARejection()
ica_rejector.mne_info = info
N_SEN = len(info['ch_names'])
ica_rejector.input = np.random.rand(N_SEN)
parent = FileSource(data_path)
parent.output = np.random.rand(info['nchan'], 1)
parent.mne_info = info
ica_rejector.parent = parent
return ica_rejector
def lsl_streamer(info, data_path): # noqa
lsl_streamer = LSLStreamOutput()
lsl_streamer.mne_info = info
N_SEN = len(info['ch_names'])
lsl_streamer.input = np.random.rand(N_SEN)
parent = FileSource(data_path)
parent.output = np.random.rand(info['nchan'], 1)
parent.mne_info = info
lsl_streamer.parent = parent
return lsl_streamer
def signal_viewer(info, data_path): # noqa
signal_viewer = SignalViewer()
signal_viewer.mne_info = info
N_SEN = len(info['ch_names'])
signal_viewer.input = np.random.rand(N_SEN)
parent = FileSource(data_path)
parent.output = np.random.rand(info['nchan'], 1)
parent.mne_info = info
signal_viewer.parent = parent
return signal_viewer
def beamformer(info, fwd_model_path, data_path): # noqa
is_adaptive = True
beamformer = Beamformer(fwd_path=fwd_model_path, is_adaptive=is_adaptive)
beamformer.mne_info = info
N_SEN = len(info['ch_names'])
beamformer.input = np.random.rand(N_SEN)
parent = FileSource(data_path)
parent.output = np.random.rand(info['nchan'], 1)
parent.mne_info = info
beamformer.parent = parent
return beamformer
def mne_gcs(info, fwd_model_path, data_path): # noqa
snr = 1
mne_gcs = MneGcs(snr=snr, fwd_path=fwd_model_path, seed=0)
mne_gcs.mne_info = info
N_SEN = len(info['ch_names'])
mne_gcs.input = np.random.rand(N_SEN)
parent = FileSource(data_path)
parent.output = np.random.rand(info['nchan'], 1)
parent.mne_info = info
mne_gcs.parent = parent
return mne_gcs
def file_outputter(info, data_path, tmp_path): # noqa
output_path = op.join(tmp_path, 'output.h5')
file_outputter = FileOutput(output_path)
file_outputter.mne_info = info
N_SEN = len(info['ch_names'])
file_outputter.input = np.random.rand(N_SEN)
parent = FileSource(data_path)
parent.output = np.random.rand(info['nchan'], 1)
parent.mne_info = info
file_outputter.parent = parent
return file_outputter
def mce(info, fwd_model_path): # noqa
n_comp = 10
print(fwd_model_path)
mce = MCE(fwd_model_path, n_comp)
mce.mne_info = info
N_SEN = len(info['ch_names'])
mce.input = np.random.rand(N_SEN)
parent = FileSource()
parent.output = np.random.rand(info['nchan'], 1)
parent.mne_info = info
mce.parent = parent
return mce
def inv_model(info, fwd_model_path): # noqa
snr = 1
method = 'MNE'
inv_model = InverseModel(
snr=snr, forward_model_path=fwd_model_path, method=method)
inv_model.mne_info = info
N_SEN = len(info['ch_names'])
inv_model.input = np.random.rand(N_SEN)
parent = FileSource()
parent.output = np.random.rand(info['nchan'], 1)
parent.mne_info = info
inv_model.parent = parent
return inv_model
def preprocessor(info, data_path): # noqa
preprocessor = Preprocessing()
preprocessor.mne_info = info
N_SEN = len(info['ch_names'])
preprocessor.input = np.random.rand(N_SEN)
parent = FileSource(data_path)
parent.output = np.random.rand(info['nchan'], 1)
parent.mne_info = info
preprocessor.parent = parent
app = QCoreApplication(sys.argv)
parent.updater = AsyncUpdater(app, parent)
return preprocessor
linear_filter.input_node = source linear_filter.initialize() linear_filter.update() # this linear filter should at least remove DC. Thus, new means should be somewhat close to zero means = np.abs(np.mean(linear_filter.output, axis=TIME_AXIS)) mean_max = np.mean(np.max(linear_filter.output, axis=TIME_AXIS)) assert(np.all(means < 0.1 * mean_max)) linear_filter.lower_cutoff = None linear_filter.initialize() linear_filter.update() assert(linear_filter.output is source.output) # Envelope extractor envelope_extractor = EnvelopeExtractor() envelope_extractor.input_node = linear_filter envelope_extractor.initialize() envelope_extractor.update() # TODO: come up with an actual way to test this stuff assert(envelope_extractor.output is not None) from cognigraph.nodes.sources import FileSource source = FileSource(r"C:\Users\evgenii\Downloads\brainvision\Bulavenkova_A_2017-10-24_15-33-18_Rest.vmrk") source.update()