def test_sleep_physionet_age(physionet_tmpdir, monkeypatch, download_is_error):
"""Test Sleep Physionet URL handling."""
# check download_is_error patching
with pytest.raises(AssertionError, match='Test should not download'):
age.fetch_data(subjects=[0], recording=[1], path=physionet_tmpdir)
# then patch
my_func = _FakeFetch()
monkeypatch.setattr(pooch, 'retrieve', my_func)
paths = age.fetch_data(subjects=[0], recording=[1], path=physionet_tmpdir)
assert_array_equal(_keep_basename_only(paths),
[['SC4001E0-PSG.edf', 'SC4001EC-Hypnogram.edf']])
paths = age.fetch_data(subjects=[0, 1],
recording=[1],
path=physionet_tmpdir)
assert_array_equal(_keep_basename_only(paths),
[['SC4001E0-PSG.edf', 'SC4001EC-Hypnogram.edf'],
['SC4011E0-PSG.edf', 'SC4011EH-Hypnogram.edf']])
paths = age.fetch_data(subjects=[0],
recording=[1, 2],
path=physionet_tmpdir)
assert_array_equal(_keep_basename_only(paths),
[['SC4001E0-PSG.edf', 'SC4001EC-Hypnogram.edf'],
['SC4002E0-PSG.edf', 'SC4002EC-Hypnogram.edf']])
EXPECTED_CALLS = ({
'name': 'SC4001E0-PSG.edf',
'hash': 'adabd3b01fc7bb75c523a974f38ee3ae4e57b40f'
}, {
'name': 'SC4001EC-Hypnogram.edf',
'hash': '21c998eadc8b1e3ea6727d3585186b8f76e7e70b'
}, {
'name': 'SC4001E0-PSG.edf',
'hash': 'adabd3b01fc7bb75c523a974f38ee3ae4e57b40f'
}, {
'name': 'SC4001EC-Hypnogram.edf',
'hash': '21c998eadc8b1e3ea6727d3585186b8f76e7e70b'
}, {
'name': 'SC4011E0-PSG.edf',
'hash': '4d17451f7847355bcab17584de05e7e1df58c660'
}, {
'name': 'SC4011EH-Hypnogram.edf',
'hash': 'd582a3cbe2db481a362af890bc5a2f5ca7c878dc'
}, {
'name': 'SC4001E0-PSG.edf',
'hash': 'adabd3b01fc7bb75c523a974f38ee3ae4e57b40f'
}, {
'name': 'SC4001EC-Hypnogram.edf',
'hash': '21c998eadc8b1e3ea6727d3585186b8f76e7e70b'
}, {
'name': 'SC4002E0-PSG.edf',
'hash': 'c6b6d7a8605cc7e7602b6028ee77f6fbf5f7581d'
}, {
'name': 'SC4002EC-Hypnogram.edf',
'hash': '386230188a3552b1fc90bba0fb7476ceaca174b6'
})
base_path = age.data_path(path=physionet_tmpdir)
_check_mocked_function_calls(my_func, EXPECTED_CALLS, base_path)
def test_sleep_physionet_age(physionet_tmpdir, mocker):
"""Test Sleep Physionet URL handling."""
my_func = mocker.patch('mne.datasets.sleep_physionet._utils._fetch_file',
side_effect=_fake_fetch_file)
params = {'path': physionet_tmpdir, 'update_path': False}
with pytest.raises(ValueError, match='Only subjects 0 to 19 are'):
paths = age.fetch_data(subjects=[20], recording=[1], **params)
paths = age.fetch_data(subjects=[0], recording=[1], **params)
assert_array_equal(_keep_basename_only(paths),
[['SC4001E0-PSG.edf', 'SC4001EC-Hypnogram.edf']])
paths = age.fetch_data(subjects=[0, 1], recording=[1], **params)
assert_array_equal(_keep_basename_only(paths),
[['SC4001E0-PSG.edf', 'SC4001EC-Hypnogram.edf'],
['SC4011E0-PSG.edf', 'SC4011EH-Hypnogram.edf']])
paths = age.fetch_data(subjects=[0], recording=[1, 2], **params)
assert_array_equal(_keep_basename_only(paths),
[['SC4001E0-PSG.edf', 'SC4001EC-Hypnogram.edf'],
['SC4002E0-PSG.edf', 'SC4002EC-Hypnogram.edf']])
EXPECTED_CALLS = ({
'name': 'SC4001E0-PSG.edf',
'hash': 'adabd3b01fc7bb75c523a974f38ee3ae4e57b40f'
}, {
'name': 'SC4001EC-Hypnogram.edf',
'hash': '21c998eadc8b1e3ea6727d3585186b8f76e7e70b'
}, {
'name': 'SC4001E0-PSG.edf',
'hash': 'adabd3b01fc7bb75c523a974f38ee3ae4e57b40f'
}, {
'name': 'SC4001EC-Hypnogram.edf',
'hash': '21c998eadc8b1e3ea6727d3585186b8f76e7e70b'
}, {
'name': 'SC4011E0-PSG.edf',
'hash': '4d17451f7847355bcab17584de05e7e1df58c660'
}, {
'name': 'SC4011EH-Hypnogram.edf',
'hash': 'd582a3cbe2db481a362af890bc5a2f5ca7c878dc'
}, {
'name': 'SC4001E0-PSG.edf',
'hash': 'adabd3b01fc7bb75c523a974f38ee3ae4e57b40f'
}, {
'name': 'SC4001EC-Hypnogram.edf',
'hash': '21c998eadc8b1e3ea6727d3585186b8f76e7e70b'
}, {
'name': 'SC4002E0-PSG.edf',
'hash': 'c6b6d7a8605cc7e7602b6028ee77f6fbf5f7581d'
}, {
'name': 'SC4002EC-Hypnogram.edf',
'hash': '386230188a3552b1fc90bba0fb7476ceaca174b6'
})
base_path = age.data_path(path=physionet_tmpdir)
_check_mocked_function_calls(mocked_func=my_func,
call_fname_hash_pairs=EXPECTED_CALLS,
base_path=base_path)
def test_sleep_physionet_age(physionet_tmpdir, mocker):
"""Test Sleep Physionet URL handling."""
my_func = mocker.patch('mne.datasets.sleep_physionet._utils._fetch_file',
side_effect=_fake_fetch_file)
params = {'path': physionet_tmpdir, 'update_path': False}
with pytest.raises(ValueError, match='Only subjects 0 to 19 are'):
paths = age.fetch_data(subjects=[20], recording=[1], **params)
paths = age.fetch_data(subjects=[0], recording=[1], **params)
assert_array_equal(_keep_basename_only(paths),
[['SC4001E0-PSG.edf', 'SC4001EC-Hypnogram.edf']])
paths = age.fetch_data(subjects=[0, 1], recording=[1], **params)
assert_array_equal(_keep_basename_only(paths),
[['SC4001E0-PSG.edf', 'SC4001EC-Hypnogram.edf'],
['SC4011E0-PSG.edf', 'SC4011EH-Hypnogram.edf']])
paths = age.fetch_data(subjects=[0], recording=[1, 2], **params)
assert_array_equal(_keep_basename_only(paths),
[['SC4001E0-PSG.edf', 'SC4001EC-Hypnogram.edf'],
['SC4002E0-PSG.edf', 'SC4002EC-Hypnogram.edf']])
EXPECTED_CALLS = (
{'name': 'SC4001E0-PSG.edf',
'hash': 'adabd3b01fc7bb75c523a974f38ee3ae4e57b40f'},
{'name': 'SC4001EC-Hypnogram.edf',
'hash': '21c998eadc8b1e3ea6727d3585186b8f76e7e70b'},
{'name': 'SC4001E0-PSG.edf',
'hash': 'adabd3b01fc7bb75c523a974f38ee3ae4e57b40f'},
{'name': 'SC4001EC-Hypnogram.edf',
'hash': '21c998eadc8b1e3ea6727d3585186b8f76e7e70b'},
{'name': 'SC4011E0-PSG.edf',
'hash': '4d17451f7847355bcab17584de05e7e1df58c660'},
{'name': 'SC4011EH-Hypnogram.edf',
'hash': 'd582a3cbe2db481a362af890bc5a2f5ca7c878dc'},
{'name': 'SC4001E0-PSG.edf',
'hash': 'adabd3b01fc7bb75c523a974f38ee3ae4e57b40f'},
{'name': 'SC4001EC-Hypnogram.edf',
'hash': '21c998eadc8b1e3ea6727d3585186b8f76e7e70b'},
{'name': 'SC4002E0-PSG.edf',
'hash': 'c6b6d7a8605cc7e7602b6028ee77f6fbf5f7581d'},
{'name': 'SC4002EC-Hypnogram.edf',
'hash': '386230188a3552b1fc90bba0fb7476ceaca174b6'})
base_path = age.data_path(path=physionet_tmpdir)
_check_mocked_function_calls(mocked_func=my_func,
call_fname_hash_pairs=EXPECTED_CALLS,
base_path=base_path)
def __init__(self,
subject_ids=None,
recording_ids=None,
preload=False,
load_eeg_only=True,
crop_wake_mins=30,
crop=None):
if subject_ids is None:
subject_ids = range(83)
if recording_ids is None:
recording_ids = [1, 2]
paths = fetch_data(subject_ids,
recording=recording_ids,
on_missing='warn')
all_base_ds = list()
for p in paths:
raw, desc = self._load_raw(p[0],
p[1],
preload=preload,
load_eeg_only=load_eeg_only,
crop_wake_mins=crop_wake_mins,
crop=crop)
base_ds = BaseDataset(raw, desc)
all_base_ds.append(base_ds)
super().__init__(all_base_ds)
def test_sleep_physionet_age_missing_subjects(physionet_tmpdir, subject,
download_is_error):
"""Test handling of missing subjects in Sleep Physionet age fetcher."""
with pytest.raises(ValueError,
match='This dataset contains subjects 0 to 82'):
age.fetch_data(subjects=[subject],
recording=[1],
on_missing='raise',
path=physionet_tmpdir)
with pytest.warns(RuntimeWarning,
match='This dataset contains subjects 0 to 82'):
age.fetch_data(subjects=[subject],
recording=[1],
on_missing='warn',
path=physionet_tmpdir)
paths = age.fetch_data(subjects=[subject],
recording=[1],
on_missing='ignore',
path=physionet_tmpdir)
assert paths == []
def test_sleep_physionet_age_missing_recordings(physionet_tmpdir, subject,
recording, download_is_error):
"""Test handling of missing recordings in Sleep Physionet age fetcher."""
with pytest.raises(ValueError,
match=f'Requested recording {recording} for subject'):
age.fetch_data(subjects=[subject],
recording=[recording],
on_missing='raise',
path=physionet_tmpdir)
with pytest.warns(RuntimeWarning,
match=f'Requested recording {recording} for subject'):
age.fetch_data(subjects=[subject],
recording=[recording],
on_missing='warn',
path=physionet_tmpdir)
paths = age.fetch_data(subjects=[subject],
recording=[recording],
on_missing='ignore',
path=physionet_tmpdir)
assert paths == []
def fetch_and_extract(self, data_folder, subject_ids):
prepared_filenames = []
orig_filenames = fetch_data(subject_ids)
for psg_file, _hypnogram_file in orig_filenames:
target_filename = Path(psg_file).name.replace(".edf", ".npy")
target_filename = data_folder / target_filename
prepared_filenames.append(target_filename)
if target_filename.exists():
continue
print("Loading PSG file with MNE...")
raw_edf = mne.io.read_raw_edf(psg_file)
raw_edf.set_channel_types(self.mapping)
eeg_array = raw_edf.get_data(picks="eeg")
eeg_array = np.ascontiguousarray(eeg_array.T, dtype=np.float32)
scale = eeg_array.std(axis=0).mean()
eeg_array /= scale
print(f"Saving scale EEG channel data to {target_filename}...")
np.save(target_filename, eeg_array, allow_pickle=False)
return prepared_filenames
from sklearn.pipeline import make_pipeline
from sklearn.preprocessing import FunctionTransformer
TIME_OF_W_TO_KEEP_SEC = 1800
nb_subjects = 20
# CAN BE UP TO 20
nights = [1]
# [1,2]
subjects = range(0,nb_subjects)
# Get path to local copies of data from Physionet files are PSG.edf and Hypnogram.edf
files = fetch_data(subjects=subjects, recording=nights)
mapping = {'EOG horizontal': 'eog',
'Resp oro-nasal': 'misc',
'EMG submental': 'misc',
'Temp rectal': 'misc',
'Event marker': 'misc'}
sleep_stages = {
'Sleep stage W' : 0,
'Sleep stage 1' : 1,
'Sleep stage 2' : 2,
'Sleep stage 3' : 3,
'Sleep stage 4' : 3,
'Sleep stage R' : 4,
'Sleep stage ?' : 5
#
# * ``-PSG.edf`` containing the polysomnography. The :term:`raw` data from the
# EEG helmet,
# * ``-Hypnogram.edf`` containing the :term:`annotations` recorded by an
# expert.
#
# Combining these two in a :class:`mne.io.Raw` object then we can extract
# :term:`events` based on the descriptions of the annotations to obtain the
# :term:`epochs`.
#
# Read the PSG data and Hypnograms to create a raw object
# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
ALICE, BOB = 0, 1
[alice_files, bob_files] = fetch_data(subjects=[ALICE, BOB], recording=[1])
mapping = {
'EOG horizontal': 'eog',
'Resp oro-nasal': 'resp',
'EMG submental': 'emg',
'Temp rectal': 'misc',
'Event marker': 'misc'
}
raw_train = mne.io.read_raw_edf(alice_files[0])
annot_train = mne.read_annotations(alice_files[1])
raw_train.set_annotations(annot_train, emit_warning=False)
raw_train.set_channel_types(mapping)
#
# * ``-PSG.edf`` containing the polysomnography. The :term:`raw` data from the
# EEG helmet,
# * ``-Hypnogram.edf`` containing the :term:`annotations` recorded by an
# expert.
#
# Combining these two in a :class:`mne.io.Raw` object then we can extract
# :term:`events` based on the descriptions of the annotations to obtain the
# :term:`epochs`.
#
# Read the PSG data and Hypnograms to create a raw object
# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
ALICE, BOB = 0, 1
[alice_files, bob_files] = fetch_data(subjects=[ALICE, BOB], recording=[1])
mapping = {'EOG horizontal': 'eog',
'Resp oro-nasal': 'misc',
'EMG submental': 'misc',
'Temp rectal': 'misc',
'Event marker': 'misc'}
raw_train = mne.io.read_raw_edf(alice_files[0])
annot_train = mne.read_annotations(alice_files[1])
raw_train.set_annotations(annot_train, emit_warning=False)
raw_train.set_channel_types(mapping)
# plot some data
raw_train.plot(duration=60, scalings='auto')
if __name__ == '__main__':
parser = argparse.ArgumentParser(
description='Automatic Sleep Scoring with NPA.')
args = parser.parse_args()
os.makedirs(data_dir + '/epochs/', exist_ok=True)
os.makedirs(data_dir + '/FOOOFs/', exist_ok=True)
print('Arguments for this experiment:', args)
start_all = time.time()
all_epochs = []
datafiles = fetch_data(subjects=list(range(20)))
# fg = FOOOFGroup(peak_width_limits=[1, 12.0])
for eeg_filename, label_filename in datafiles:
eeg = mne.io.read_raw_edf(eeg_filename, preload=True, verbose=0)
labels = mne.read_annotations(label_filename)
subject_id = int(eeg_filename.split('\\')[-1][3:5])
session_id = int(eeg_filename.split('\\')[-1][5])
eeg.set_annotations(labels, emit_warning=False)
eeg.set_channel_types(mapping)
chunk_length = 2.
tmax = chunk_length - 1. / eeg.info['sfreq']
import mne
from mne.datasets.sleep_physionet.age import fetch_data
from mne.time_frequency import psd_welch
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score
from sklearn.metrics import confusion_matrix
from sklearn.metrics import classification_report
from sklearn.pipeline import make_pipeline
from sklearn.preprocessing import FunctionTransformer
GROUPA, GROUPB = 0, 1
#加载两个组的实验数据
[groupa_files, groupb_files] = fetch_data(subjects=[GROUPA, GROUPB], recording=[1])
#通道名称映射
mapping = {'EOG horizontal': 'eog',
'Resp oro-nasal': 'misc',
'EMG submental': 'misc',
'Temp rectal': 'misc',
'Event marker': 'misc'}
#读取GROUPA组的edf文件和其对应的注释文件
raw_train = mne.io.read_raw_edf(groupa_files[0])
annot_train = mne.read_annotations(groupa_files[1])
raw_train.set_annotations(annot_train, emit_warning=False)
raw_train.set_channel_types(mapping)
# 绘制0s开始，时间窗口长度为60s的连续通道数据波形图
raw_train.plot(duration=60, scalings='auto')
#
# License: BSD Style.
import os
import numpy as np
import mne
from mne.datasets.sleep_physionet.age import fetch_data
from joblib import Memory
##############################################################################
# Define code to get epochs for all subjects
# subjects = [0, 1]
subjects = range(20)
n_groups = 5 # keep 5 subjects out
files = fetch_data(subjects=subjects, recording=[1])
mapping = {
'EOG horizontal': 'eog',
'Resp oro-nasal': 'misc',
'EMG submental': 'misc',
'Temp rectal': 'misc',
'Event marker': 'misc'
}
mem = Memory(mne.get_config('MNE_DATA'))
@mem.cache
def get_epochs_data(raw_fname, annot_fname):
print("Extracting Epochs from: %s" % os.path.basename(raw_fname))
