def test_dataset(self):
dataset = TUHEEGAbnormalDataset(self.PATH)
preprocessing = Pipeline([
CommonChannelSet(),
LowestFrequency(),
BandPassFrequency(0.1, 47),
ToDataframe(),
DynamicWindow(4),
JoinedPreprocessor(inputs=[[
BinarizedSpearmanCorrelation(),
CorrelationToAdjacency()
],
Bandpower()],
output=GraphWithFeatures())
])
dataset = dataset.set_pipeline(preprocessing).load()
0, os.path.abspath(os.path.join(os.path.dirname(__file__), '../..')))
from pyeeglab import TUHEEGAbnormalDataset, Pipeline, CommonChannelSet, \
LowestFrequency, ToDataframe, DynamicWindow, BinarizedSpearmanCorrelation, \
CorrelationToAdjacency, Bandpower, GraphWithFeatures, ForkedPreprocessor
dataset = TUHEEGAbnormalDataset('../../data/tuh_eeg_abnormal/')
preprocessing = Pipeline([
CommonChannelSet(),
LowestFrequency(),
ToDataframe(),
DynamicWindow(8),
ForkedPreprocessor(inputs=[[
BinarizedSpearmanCorrelation(),
CorrelationToAdjacency()
], Bandpower()],
output=GraphWithFeatures())
],
to_numpy=False)
dataset = dataset.set_pipeline(preprocessing).load()
data, labels = dataset['data'], dataset['labels']
data = [[[nx_to_node_features(G, ['features'])[0],
add_eye(nx_to_adj(G)[0])] for G in graphs] for graphs in data]
data = [[e for G in graphs for e in G] for graphs in data]
classes = len(set(labels))
graphs = len(data[0])
from pyeeglab import TUHEEGAbnormalDataset, SinglePickleCache, Pipeline, CommonChannelSet, \
LowestFrequency, ToDataframe, DynamicWindow, BinarizedSpearmanCorrelation, \
CorrelationToAdjacency, Bandpower, GraphWithFeatures, JoinedPreprocessor
dataset = TUHEEGAbnormalDataset('../../data/tuh_eeg_abnormal/v2.0.0/edf')
dataset.set_cache_manager(SinglePickleCache('../../export'))
preprocessing = Pipeline([
CommonChannelSet(),
LowestFrequency(),
ToDataframe(),
DynamicWindow(8),
JoinedPreprocessor(
inputs=[
[BinarizedSpearmanCorrelation(), CorrelationToAdjacency()],
Bandpower()
],
output=GraphWithFeatures()
)
])
dataset = dataset.set_pipeline(preprocessing).load()
data, labels = dataset['data'], dataset['labels']
data = [
[
[
nx_to_node_features(G, ['features'])[0],
add_eye(nx_to_adj(G)[0])
]
for G in graphs