def test_if_pipeline_works_at_all(self):
pipeline = Pipeline()
pipeline.source = self.input_node
for processor_node in self.processor_nodes:
pipeline.add_processor(processor_node=processor_node)
pipeline.add_output(self.output_node)
pipeline.run()
def pipeline():
source = ConcreteSource()
processor = ConcreteProcessor()
output = ConcreteOutput()
pipeline = Pipeline()
pipeline.source = source
source.add_child(processor)
processor.add_child(output)
pipeline._processors.append(processor)
pipeline._outputs.append(output)
return pipeline
import mne
from pyqtgraph import QtCore, QtGui
import pyqtgraph
pyqtgraph.setConfigOption('useOpenGL',True)
from cognigraph.pipeline import Pipeline
from cognigraph.nodes import sources, processors, outputs
from cognigraph import TIME_AXIS
from cognigraph.gui.window import GUIWindow
import logging
app = QtGui.QApplication(sys.argv)
pipeline = Pipeline()
cur_dir = '/home/dmalt/Code/python/cogni_submodules'
test_data_path = cur_dir + '/tests/data/'
print(test_data_path)
# sim_data_fname = 'raw_sim.fif'
data_fname = 'Koleno.fif'
data_path = op.join(test_data_path, data_fname)
source = sources.FifSource(file_path=data_path)
pipeline.source = source
# Processors
%gui qt from PyQt5 import QtCore from cognigraph.pipeline import Pipeline from cognigraph.nodes import sources, processors, outputs from cognigraph.gui.window import GUIWindow pipeline = Pipeline() file_path = r"C:\Users\evgenii\Downloads\brainvision\Bulavenkova_A_2017-10-24_15-33-18_Rest.vmrk" pipeline.source = sources.FileSource(file_path=file_path) # linear_filter = processors.LinearFilter(lower_cutoff=0.1, upper_cutoff=40) # pipeline.add_processor(linear_filter) # pipeline.add_processor(processors.MNE(method='MNE')) # pipeline.add_processor(processors.EnvelopeExtractor()) # pipeline.add_output(outputs.BrainViewer()) # pipeline.add_output(outputs.LSLStreamOutput()) # pipeline.initialize_all_nodes() window = GUIWindow(pipeline=pipeline) window.init_ui() window.setWindowFlags(QtCore.Qt.WindowStaysOnTopHint) window.show()
def assemble_pipeline(file_path=None,
fwd_path=None,
subject=None,
subjects_dir=None,
inverse_method='mne'):
pipeline = Pipeline()
source = sources.FileSource(file_path=file_path)
source.loop_the_file = True
source.MAX_SAMPLES_IN_CHUNK = 10000
pipeline.source = source
# ----------------------------- processors ----------------------------- #
preprocessing = processors.Preprocessing(collect_for_x_seconds=120)
pipeline.add_processor(preprocessing)
linear_filter = processors.LinearFilter(lower_cutoff=8.0,
upper_cutoff=12.0)
pipeline.add_processor(linear_filter)
if inverse_method == 'mne':
inverse_model = processors.InverseModel(method='MNE',
snr=1.0,
forward_model_path=fwd_path)
# inverse_model = processors.MneGcs(snr=1.0, seed=1000,
# forward_model_path=fwd_path)
pipeline.add_processor(inverse_model)
envelope_extractor = processors.EnvelopeExtractor(0.99)
pipeline.add_processor(envelope_extractor)
elif inverse_method == 'beamformer':
inverse_model = processors.Beamformer(
forward_model_path=fwd_path,
is_adaptive=True,
output_type='activation',
forgetting_factor_per_second=0.95)
pipeline.add_processor(inverse_model)
envelope_extractor = processors.EnvelopeExtractor(0.99)
pipeline.add_processor(envelope_extractor)
elif inverse_method == 'mce':
inverse_model = processors.MCE(forward_model_path=fwd_path, snr=1.0)
pipeline.add_processor(inverse_model)
envelope_extractor = processors.EnvelopeExtractor(0.995)
pipeline.add_processor(envelope_extractor)
# ---------------------------------------------------------------------- #
# ------------------------------ outputs ------------------------------ #
global_mode = outputs.BrainViewer.LIMITS_MODES.GLOBAL
brain_viewer = outputs.BrainViewer(limits_mode=global_mode,
buffer_length=6,
surfaces_dir=None)
pipeline.add_output(brain_viewer, parent=envelope_extractor)
# roi_average = processors.AtlasViewer(SUBJECT, subjects_dir)
# roi_average.parent = inverse_model
# pipeline.add_processor(roi_average)
# aec = processors.AmplitudeEnvelopeCorrelations(
# method=None,
# seed=1000
# # method='temporal_orthogonalization',
# # method='geometric_correction',
# # seed=0
# )
# pipeline.add_processor(aec)
# aec.parent = inverse_model
# # coh = processors.Coherence(
# # method='coh', seed=0)
# aec_env = processors.EnvelopeExtractor(0.995)
# pipeline.add_processor(aec_env)
# seed_viewer = outputs.BrainViewer(
# limits_mode=global_mode, buffer_length=6,
# surfaces_dir=op.join(subjects_dir, SUBJECT))
# pipeline.add_output(seed_viewer, parent=aec_env)
# pipeline.add_output(outputs.LSLStreamOutput())
# signal_viewer = outputs.SignalViewer()
# signal_viewer_src = outputs.SignalViewer()
# pipeline.add_output(signal_viewer, parent=linear_filter)
# pipeline.add_output(signal_viewer_src, parent=roi_average)
# con_viewer = outputs.ConnectivityViewer(
# surfaces_dir=op.join(subjects_dir, SUBJECT))
# pipeline.add_output(con_viewer, parent=aec)
# --------------------------------------------------------------------- #
return pipeline
import sys from pyqtgraph import QtCore, QtGui from cognigraph.pipeline import Pipeline from cognigraph.nodes import sources, processors, outputs from cognigraph import TIME_AXIS from cognigraph.gui.window import GUIWindow app = QtGui.QApplication(sys.argv) # BOBE pipeline = Pipeline() file_path = r"/home/evgenii/Downloads/brainvision/Bulavenkova_A_2017-10-24_15-33-18_Rest.vhdr" source = sources.FileSource(file_path=file_path) pipeline.source = source inverse = processors.InverseModel() pipeline.add_processor(inverse) three_dee = outputs.ThreeDeeBrain() pipeline.add_output(three_dee) pipeline.initialize_all_nodes() three_dee.widget.show() source.output = source.data.take(indices=(0,), axis=TIME_AXIS) inverse.update()
%gui qt from pyqtgraph import QtCore from cognigraph.pipeline import Pipeline from cognigraph.nodes import sources, processors, outputs from cognigraph.gui.window import GUIWindow pipeline = Pipeline() file_path = r"C:\Users\evgenii\Downloads\brainvision\Bulavenkova_A_2017-10-24_15-33-18_Rest.vmrk" pipeline.source = sources.BrainvisionSource(file_path=file_path) # linear_filter = processors.LinearFilter(lower_cutoff=0.1, upper_cutoff=40) # pipeline.add_processor(linear_filter) # pipeline.add_processor(processors.InverseModel(method='MNE')) # pipeline.add_processor(processors.EnvelopeExtractor()) # pipeline.add_output(outputs.ThreeDeeBrain()) # pipeline.add_output(outputs.LSLStreamOutput()) # pipeline.initialize_all_nodes() window = GUIWindow(pipeline=pipeline) window.init_ui() window.setWindowFlags(QtCore.Qt.WindowStaysOnTopHint) window.show()
from cognigraph.pipeline import Pipeline
from cognigraph.nodes import sources, processors, outputs
# from cognigraph import TIME_AXIS
from cognigraph.gui.window import GUIWindow
import logging
import mne
mne.set_log_level('WARNING')
app = QtWidgets.QApplication(sys.argv)
logging.basicConfig(level=logging.INFO,
format='%(asctime)s:%(name)-17s:%(levelname)s:%(message)s')
logger = logging.getLogger(__name__)
pipeline = Pipeline()
cur_dir = '/home/dmalt/Code/python/cogni_submodules'
test_data_path = cur_dir + '/tests/data/'
print(test_data_path)
sim_data_fname = 'raw_sim_nobads.fif'
# sim_data_fname = 'DF_2018-03-02_11-34-38.edf'
# sim_data_fname = 'Koleno.fif'
fwd_fname = 'dmalt_custom_mr-fwd.fif'
# fwd_fname = 'sample_1005-eeg-ico-4-fwd.fif'
# fwd_fname = 'sample_1005-eeg-oct-6-fwd.fif'
# fwd_fname = 'DF_2018-03-02_11-34-38-fwd.fif'
surf_dir = '/home/dmalt/mne_data/MNE-sample-data/subjects/sample/surf'
fwd_path = op.join(test_data_path, fwd_fname)
import sys from pyqtgraph import QtCore, QtGui from cognigraph.pipeline import Pipeline from cognigraph.nodes import sources, processors, outputs from cognigraph import TIME_AXIS from cognigraph.gui.window import GUIWindow app = QtGui.QApplication(sys.argv) pipeline = Pipeline() # file_path = r"/home/dmalt/Code/python/real_eyes/Koleno.eeg" file_path = r"/home/dmalt/Data/cognigraph/data/Koleno.eeg" # vhdr_file_path = r"/home/dmalt/Code/python/real_eyes/Koleno.vhdr" # vhdr_file_path = r"/home/dmalt/Data/cognigraph/data/Koleno.vhdr" # vhdr_file_path = r"/home/dmalt/Data/cognigraph/data/Koleno.fif" fwd_path = r'/home/dmalt/mne_data/MNE-sample-data/MEG/sample/dmalt_custom-fwd.fif' source = sources.FileSource(file_path=file_path) pipeline.source = source # pipeline.source = sources.LSLStreamSource(stream_name='cognigraph-mock-stream') # Processors preprocessing = processors.Preprocessing(collect_for_x_seconds=120) pipeline.add_processor(preprocessing) linear_filter = processors.LinearFilter(lower_cutoff=8.0, upper_cutoff=12.0) pipeline.add_processor(linear_filter) inverse_model = processors.InverseModel(method='MNE',