braindecode 0.1.4.7__tar.gz → 0.1.5__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/Braindecode.egg-info/PKG-INFO +10 -6
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/Braindecode.egg-info/SOURCES.txt +4 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/Braindecode.egg-info/requires.txt +3 -2
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/PKG-INFO +10 -6
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/README.rst +9 -5
- Braindecode-0.1.5/braindecode/datasets/__init__.py +3 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/braindecode/datasets/bbci.py +75 -13
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/braindecode/datasets/multiple.py +4 -1
- Braindecode-0.1.5/braindecode/datautil/__init__.py +3 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/braindecode/datautil/iterators.py +146 -58
- Braindecode-0.1.5/braindecode/datautil/signal_target.py +38 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/braindecode/datautil/signalproc.py +74 -3
- Braindecode-0.1.5/braindecode/datautil/splitters.py +74 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/braindecode/datautil/trial_segment.py +117 -25
- Braindecode-0.1.5/braindecode/datautil/util.py +37 -0
- Braindecode-0.1.5/braindecode/experiments/__init__.py +3 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/braindecode/experiments/experiment.py +187 -1
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/braindecode/experiments/monitors.py +132 -10
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/braindecode/experiments/stopcriteria.py +51 -2
- Braindecode-0.1.5/braindecode/mne_ext/__init__.py +3 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/braindecode/mne_ext/signalproc.py +37 -5
- Braindecode-0.1.5/braindecode/models/__init__.py +3 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/braindecode/models/deep4.py +12 -0
- Braindecode-0.1.5/braindecode/models/eegnet.py +111 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/braindecode/models/shallow_fbcsp.py +12 -0
- Braindecode-0.1.5/braindecode/models/util.py +39 -0
- Braindecode-0.1.5/braindecode/torch_ext/__init__.py +3 -0
- Braindecode-0.1.5/braindecode/torch_ext/constraints.py +27 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/braindecode/torch_ext/functions.py +1 -1
- Braindecode-0.1.5/braindecode/torch_ext/init.py +23 -0
- Braindecode-0.1.5/braindecode/torch_ext/losses.py +31 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/braindecode/torch_ext/modules.py +12 -2
- Braindecode-0.1.5/braindecode/torch_ext/util.py +42 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/braindecode/util.py +0 -42
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/docs/Makefile +2 -2
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/docs/conf.py +54 -5
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/docs/index.rst +15 -1
- Braindecode-0.1.5/docs/notebooks/BBCI_Data.ipynb +933 -0
- Braindecode-0.1.5/docs/notebooks/BBCI_Data_Start_Stop.ipynb +842 -0
- Braindecode-0.1.5/docs/notebooks/BBCI_Data_Trialwise.ipynb +820 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/docs/notebooks/Cropped_Decoding.ipynb +137 -103
- Braindecode-0.1.5/docs/notebooks/Experiment_Class.ipynb +807 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/docs/notebooks/TrialWise_Decoding.ipynb +216 -141
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/docs/source/braindecode.datasets.rst +5 -7
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/docs/source/braindecode.datautil.rst +5 -7
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/docs/source/braindecode.experiments.rst +5 -7
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/docs/source/braindecode.mne_ext.rst +5 -7
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/docs/source/braindecode.models.rst +18 -4
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/docs/source/braindecode.rst +5 -7
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/docs/source/braindecode.torch_ext.rst +13 -7
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/setup.py +3 -2
- Braindecode-0.1.4.7/braindecode/datasets/__init__.py +0 -0
- Braindecode-0.1.4.7/braindecode/datautil/__init__.py +0 -0
- Braindecode-0.1.4.7/braindecode/datautil/signal_target.py +0 -20
- Braindecode-0.1.4.7/braindecode/datautil/splitters.py +0 -33
- Braindecode-0.1.4.7/braindecode/datautil/util.py +0 -2
- Braindecode-0.1.4.7/braindecode/experiments/__init__.py +0 -0
- Braindecode-0.1.4.7/braindecode/mne_ext/__init__.py +0 -0
- Braindecode-0.1.4.7/braindecode/models/__init__.py +0 -0
- Braindecode-0.1.4.7/braindecode/torch_ext/__init__.py +0 -0
- Braindecode-0.1.4.7/braindecode/torch_ext/constraints.py +0 -13
- Braindecode-0.1.4.7/braindecode/torch_ext/losses.py +0 -11
- Braindecode-0.1.4.7/braindecode/torch_ext/util.py +0 -33
- Braindecode-0.1.4.7/docs/notebooks/BBCI_Data.ipynb +0 -683
- Braindecode-0.1.4.7/docs/notebooks/BBCI_Data_Start_Stop.ipynb +0 -877
- Braindecode-0.1.4.7/docs/notebooks/Experiment_Class.ipynb +0 -819
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/Braindecode.egg-info/dependency_links.txt +0 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/Braindecode.egg-info/not-zip-safe +0 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/Braindecode.egg-info/top_level.txt +0 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/LICENSE.txt +0 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/MANIFEST.in +0 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/braindecode/__init__.py +0 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/docs/source/modules.rst +0 -0
- {Braindecode-0.1.4.7 → Braindecode-0.1.5}/setup.cfg +0 -0
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 1.1
|
|
2
2
|
Name: Braindecode
|
|
3
|
-
Version: 0.1.
|
|
3
|
+
Version: 0.1.5
|
|
4
4
|
Summary: A deep learning toolbox to decode raw time-domain EEG.
|
|
5
5
|
Home-page: https://github.com/robintibor/braindecode
|
|
6
6
|
Author: Robin Tibor Schirrmeister
|
|
@@ -12,14 +12,17 @@ Description: Note: The old braindecode repository has been moved to
|
|
|
12
12
|
Braindecode
|
|
13
13
|
===========
|
|
14
14
|
|
|
15
|
-
|
|
15
|
+
**Under construction**
|
|
16
|
+
|
|
17
|
+
A deep learning toolbox to decode raw time-domain EEG.
|
|
18
|
+
|
|
16
19
|
|
|
17
20
|
Installation
|
|
18
21
|
============
|
|
19
22
|
|
|
20
|
-
|
|
23
|
+
1. Install pytorch from http://pytorch.org/ (you don't need to install torchvision).
|
|
21
24
|
|
|
22
|
-
|
|
25
|
+
2. Install newest version of python-mne:
|
|
23
26
|
|
|
24
27
|
.. code-block:: bash
|
|
25
28
|
|
|
@@ -28,17 +31,18 @@ Description: Note: The old braindecode repository has been moved to
|
|
|
28
31
|
python setup.py install
|
|
29
32
|
|
|
30
33
|
|
|
31
|
-
|
|
34
|
+
3. Install braindecode via pip:
|
|
32
35
|
|
|
33
36
|
.. code-block:: bash
|
|
34
37
|
|
|
35
38
|
pip install braindecode
|
|
36
39
|
|
|
37
40
|
|
|
41
|
+
|
|
38
42
|
Documentation
|
|
39
43
|
=============
|
|
40
44
|
|
|
41
|
-
Documentation is online under
|
|
45
|
+
Documentation is online under https://robintibor.github.io/braindecode/
|
|
42
46
|
|
|
43
47
|
Keywords: eeg deep-learning brain-state-decoding
|
|
44
48
|
Platform: UNKNOWN
|
|
@@ -28,10 +28,13 @@ braindecode/mne_ext/__init__.py
|
|
|
28
28
|
braindecode/mne_ext/signalproc.py
|
|
29
29
|
braindecode/models/__init__.py
|
|
30
30
|
braindecode/models/deep4.py
|
|
31
|
+
braindecode/models/eegnet.py
|
|
31
32
|
braindecode/models/shallow_fbcsp.py
|
|
33
|
+
braindecode/models/util.py
|
|
32
34
|
braindecode/torch_ext/__init__.py
|
|
33
35
|
braindecode/torch_ext/constraints.py
|
|
34
36
|
braindecode/torch_ext/functions.py
|
|
37
|
+
braindecode/torch_ext/init.py
|
|
35
38
|
braindecode/torch_ext/losses.py
|
|
36
39
|
braindecode/torch_ext/modules.py
|
|
37
40
|
braindecode/torch_ext/util.py
|
|
@@ -40,6 +43,7 @@ docs/conf.py
|
|
|
40
43
|
docs/index.rst
|
|
41
44
|
docs/notebooks/BBCI_Data.ipynb
|
|
42
45
|
docs/notebooks/BBCI_Data_Start_Stop.ipynb
|
|
46
|
+
docs/notebooks/BBCI_Data_Trialwise.ipynb
|
|
43
47
|
docs/notebooks/Cropped_Decoding.ipynb
|
|
44
48
|
docs/notebooks/Experiment_Class.ipynb
|
|
45
49
|
docs/notebooks/TrialWise_Decoding.ipynb
|
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 1.1
|
|
2
2
|
Name: Braindecode
|
|
3
|
-
Version: 0.1.
|
|
3
|
+
Version: 0.1.5
|
|
4
4
|
Summary: A deep learning toolbox to decode raw time-domain EEG.
|
|
5
5
|
Home-page: https://github.com/robintibor/braindecode
|
|
6
6
|
Author: Robin Tibor Schirrmeister
|
|
@@ -12,14 +12,17 @@ Description: Note: The old braindecode repository has been moved to
|
|
|
12
12
|
Braindecode
|
|
13
13
|
===========
|
|
14
14
|
|
|
15
|
-
|
|
15
|
+
**Under construction**
|
|
16
|
+
|
|
17
|
+
A deep learning toolbox to decode raw time-domain EEG.
|
|
18
|
+
|
|
16
19
|
|
|
17
20
|
Installation
|
|
18
21
|
============
|
|
19
22
|
|
|
20
|
-
|
|
23
|
+
1. Install pytorch from http://pytorch.org/ (you don't need to install torchvision).
|
|
21
24
|
|
|
22
|
-
|
|
25
|
+
2. Install newest version of python-mne:
|
|
23
26
|
|
|
24
27
|
.. code-block:: bash
|
|
25
28
|
|
|
@@ -28,17 +31,18 @@ Description: Note: The old braindecode repository has been moved to
|
|
|
28
31
|
python setup.py install
|
|
29
32
|
|
|
30
33
|
|
|
31
|
-
|
|
34
|
+
3. Install braindecode via pip:
|
|
32
35
|
|
|
33
36
|
.. code-block:: bash
|
|
34
37
|
|
|
35
38
|
pip install braindecode
|
|
36
39
|
|
|
37
40
|
|
|
41
|
+
|
|
38
42
|
Documentation
|
|
39
43
|
=============
|
|
40
44
|
|
|
41
|
-
Documentation is online under
|
|
45
|
+
Documentation is online under https://robintibor.github.io/braindecode/
|
|
42
46
|
|
|
43
47
|
Keywords: eeg deep-learning brain-state-decoding
|
|
44
48
|
Platform: UNKNOWN
|
|
@@ -4,14 +4,17 @@ https://github.com/robintibor/braindevel.
|
|
|
4
4
|
Braindecode
|
|
5
5
|
===========
|
|
6
6
|
|
|
7
|
-
|
|
7
|
+
**Under construction**
|
|
8
|
+
|
|
9
|
+
A deep learning toolbox to decode raw time-domain EEG.
|
|
10
|
+
|
|
8
11
|
|
|
9
12
|
Installation
|
|
10
13
|
============
|
|
11
14
|
|
|
12
|
-
|
|
15
|
+
1. Install pytorch from http://pytorch.org/ (you don't need to install torchvision).
|
|
13
16
|
|
|
14
|
-
|
|
17
|
+
2. Install newest version of python-mne:
|
|
15
18
|
|
|
16
19
|
.. code-block:: bash
|
|
17
20
|
|
|
@@ -20,14 +23,15 @@ Second install newest version of python-mne:
|
|
|
20
23
|
python setup.py install
|
|
21
24
|
|
|
22
25
|
|
|
23
|
-
|
|
26
|
+
3. Install braindecode via pip:
|
|
24
27
|
|
|
25
28
|
.. code-block:: bash
|
|
26
29
|
|
|
27
30
|
pip install braindecode
|
|
28
31
|
|
|
29
32
|
|
|
33
|
+
|
|
30
34
|
Documentation
|
|
31
35
|
=============
|
|
32
36
|
|
|
33
|
-
Documentation is online under
|
|
37
|
+
Documentation is online under https://robintibor.github.io/braindecode/
|
|
@@ -11,18 +11,34 @@ log = logging.getLogger(__name__)
|
|
|
11
11
|
|
|
12
12
|
|
|
13
13
|
class BBCIDataset(object):
|
|
14
|
+
"""
|
|
15
|
+
Loader class for files created by saving BBCI files in matlab (make
|
|
16
|
+
sure to save with '-v7.3' in matlab, see
|
|
17
|
+
https://de.mathworks.com/help/matlab/import_export/mat-file-versions.html#buk6i87
|
|
18
|
+
)
|
|
19
|
+
|
|
20
|
+
Parameters
|
|
21
|
+
----------
|
|
22
|
+
filename: str
|
|
23
|
+
load_sensor_names: list of str, optional
|
|
24
|
+
Also speeds up loading if you only load some sensors.
|
|
25
|
+
None means load all sensors.
|
|
26
|
+
check_class_names: bool, optional
|
|
27
|
+
check if the class names are part of some known class names at
|
|
28
|
+
Translational NeuroTechnology Lab, AG Ball, Freiburg, Germany.
|
|
29
|
+
"""
|
|
14
30
|
def __init__(self, filename, load_sensor_names=None, check_class_names=False):
|
|
15
31
|
self.__dict__.update(locals())
|
|
16
32
|
del self.self
|
|
17
33
|
|
|
18
34
|
def load(self):
|
|
19
|
-
cnt = self.
|
|
20
|
-
cnt = self.
|
|
35
|
+
cnt = self._load_continuous_signal()
|
|
36
|
+
cnt = self._add_markers(cnt)
|
|
21
37
|
return cnt
|
|
22
38
|
|
|
23
|
-
def
|
|
24
|
-
wanted_chan_inds, wanted_sensor_names = self.
|
|
25
|
-
fs = self.
|
|
39
|
+
def _load_continuous_signal(self):
|
|
40
|
+
wanted_chan_inds, wanted_sensor_names = self._determine_sensors()
|
|
41
|
+
fs = self._determine_samplingrate()
|
|
26
42
|
with h5py.File(self.filename, 'r') as h5file:
|
|
27
43
|
samples = int(h5file['nfo']['T'][0, 0])
|
|
28
44
|
cnt_signal_shape = (samples, len(wanted_chan_inds))
|
|
@@ -48,7 +64,7 @@ class BBCIDataset(object):
|
|
|
48
64
|
cnt = mne.io.RawArray(continuous_signal.T, info)
|
|
49
65
|
return cnt
|
|
50
66
|
|
|
51
|
-
def
|
|
67
|
+
def _determine_sensors(self):
|
|
52
68
|
all_sensor_names = self.get_all_sensors(self.filename, pattern=None)
|
|
53
69
|
if self.load_sensor_names is None:
|
|
54
70
|
|
|
@@ -71,11 +87,11 @@ class BBCIDataset(object):
|
|
|
71
87
|
len(eeg_sensor_names) == 16), (
|
|
72
88
|
"Recheck this code if you have different sensors...")
|
|
73
89
|
self.load_sensor_names = eeg_sensor_names
|
|
74
|
-
chan_inds = self.
|
|
75
|
-
|
|
90
|
+
chan_inds = self._determine_chan_inds(all_sensor_names,
|
|
91
|
+
self.load_sensor_names)
|
|
76
92
|
return chan_inds, self.load_sensor_names
|
|
77
93
|
|
|
78
|
-
def
|
|
94
|
+
def _determine_samplingrate(self):
|
|
79
95
|
with h5py.File(self.filename, 'r') as h5file:
|
|
80
96
|
fs = h5file['dat']['fs'][0, 0]
|
|
81
97
|
assert isinstance(fs, int) or fs.is_integer()
|
|
@@ -83,7 +99,7 @@ class BBCIDataset(object):
|
|
|
83
99
|
return fs
|
|
84
100
|
|
|
85
101
|
@staticmethod
|
|
86
|
-
def
|
|
102
|
+
def _determine_chan_inds(all_sensor_names, sensor_names):
|
|
87
103
|
assert sensor_names is not None
|
|
88
104
|
chan_inds = [all_sensor_names.index(s) for s in sensor_names]
|
|
89
105
|
assert len(chan_inds) == len(sensor_names), ("All"
|
|
@@ -94,6 +110,20 @@ class BBCIDataset(object):
|
|
|
94
110
|
|
|
95
111
|
@staticmethod
|
|
96
112
|
def get_all_sensors(filename, pattern=None):
|
|
113
|
+
"""
|
|
114
|
+
Get all sensors that exist in the given file.
|
|
115
|
+
|
|
116
|
+
Parameters
|
|
117
|
+
----------
|
|
118
|
+
filename: str
|
|
119
|
+
pattern: str, optional
|
|
120
|
+
Only return those sensor names that match the given pattern.
|
|
121
|
+
|
|
122
|
+
Returns
|
|
123
|
+
-------
|
|
124
|
+
sensor_names: list of str
|
|
125
|
+
Sensor names that match the pattern or all sensor names in the file.
|
|
126
|
+
"""
|
|
97
127
|
with h5py.File(filename, 'r') as h5file:
|
|
98
128
|
clab_set = h5file['dat']['clab'][:].squeeze()
|
|
99
129
|
all_sensor_names = [''.join(chr(c) for c in h5file[obj_ref]) for
|
|
@@ -104,7 +134,7 @@ class BBCIDataset(object):
|
|
|
104
134
|
all_sensor_names)
|
|
105
135
|
return all_sensor_names
|
|
106
136
|
|
|
107
|
-
def
|
|
137
|
+
def _add_markers(self, cnt):
|
|
108
138
|
with h5py.File(self.filename, 'r') as h5file:
|
|
109
139
|
event_times_in_ms = h5file['mrk']['time'][:].squeeze()
|
|
110
140
|
event_classes = h5file['mrk']['event']['desc'][:].squeeze()
|
|
@@ -115,7 +145,7 @@ class BBCIDataset(object):
|
|
|
115
145
|
for obj_ref in class_name_set]
|
|
116
146
|
|
|
117
147
|
if self.check_class_names:
|
|
118
|
-
|
|
148
|
+
_check_class_names(
|
|
119
149
|
all_class_names, event_times_in_ms, event_classes,)
|
|
120
150
|
|
|
121
151
|
event_times_in_samples = event_times_in_ms * cnt.info['sfreq'] / 1000.0
|
|
@@ -151,7 +181,20 @@ class BBCIDataset(object):
|
|
|
151
181
|
return cnt
|
|
152
182
|
|
|
153
183
|
|
|
154
|
-
def
|
|
184
|
+
def _check_class_names(all_class_names, event_times_in_ms, event_classes, ):
|
|
185
|
+
"""
|
|
186
|
+
Checks if the class names are part of some known class names used in
|
|
187
|
+
translational neurotechnology lab, AG Ball, Freiburg.
|
|
188
|
+
|
|
189
|
+
Logs warning in case class names are not known.
|
|
190
|
+
|
|
191
|
+
Parameters
|
|
192
|
+
----------
|
|
193
|
+
all_class_names: list of str
|
|
194
|
+
event_times_in_ms: list of number
|
|
195
|
+
event_classes: list of number
|
|
196
|
+
|
|
197
|
+
"""
|
|
155
198
|
if all_class_names == ['Right Hand', 'Left Hand', 'Rest', 'Feet']:
|
|
156
199
|
pass
|
|
157
200
|
elif ((all_class_names == ['1', '10', '11', '111', '12', '13',
|
|
@@ -176,6 +219,8 @@ def check_class_names(all_class_names, event_times_in_ms, event_classes,):
|
|
|
176
219
|
event_classes[left_mask] = 2
|
|
177
220
|
event_classes[rest_mask] = 3
|
|
178
221
|
event_classes[feet_mask] = 4
|
|
222
|
+
log.warn("Swapped class names {:s}... might cause problems...".format(
|
|
223
|
+
all_class_names))
|
|
179
224
|
elif all_class_names == ['Right Hand Start', 'Left Hand Start',
|
|
180
225
|
'Rest Start', 'Feet Start',
|
|
181
226
|
'Right Hand End',
|
|
@@ -320,7 +365,24 @@ def check_class_names(all_class_names, event_times_in_ms, event_classes,):
|
|
|
320
365
|
log.warn("Unknown class names {:s}".format(
|
|
321
366
|
all_class_names))
|
|
322
367
|
|
|
368
|
+
|
|
323
369
|
def load_bbci_sets_from_folder(folder, runs='all'):
|
|
370
|
+
"""
|
|
371
|
+
Load bbci datasets from files in given folder.
|
|
372
|
+
|
|
373
|
+
Parameters
|
|
374
|
+
----------
|
|
375
|
+
folder: str
|
|
376
|
+
Folder with .BBCI.mat files inside
|
|
377
|
+
runs: list of int
|
|
378
|
+
If you only want to load specific runs.
|
|
379
|
+
Assumes filenames with such kind of part: S001R02 for Run 2.
|
|
380
|
+
Tries to match this regex: ``'S[0-9]{3,3}R[0-9]{2,2}_'``.
|
|
381
|
+
|
|
382
|
+
Returns
|
|
383
|
+
-------
|
|
384
|
+
|
|
385
|
+
"""
|
|
324
386
|
bbci_mat_files = sorted(glob(os.path.join(folder, '*.BBCI.mat')))
|
|
325
387
|
if runs != 'all':
|
|
326
388
|
file_run_numbers = [int(re.search('S[0-9]{3,3}R[0-9]{2,2}_', f).group()[5:7])
|
|
@@ -9,11 +9,14 @@ log = logging.getLogger(__name__)
|
|
|
9
9
|
class MultipleSetLoader(object):
|
|
10
10
|
"""
|
|
11
11
|
Class to load multiple sets.
|
|
12
|
-
|
|
12
|
+
|
|
13
|
+
Resamples individual sets down to lowest common sampling frequency,
|
|
14
|
+
if necessary.
|
|
13
15
|
|
|
14
16
|
Parameters
|
|
15
17
|
----------
|
|
16
18
|
set_loaders: object with load method: () -> MNE Raw object
|
|
19
|
+
The individual set loaders.
|
|
17
20
|
"""
|
|
18
21
|
|
|
19
22
|
def __init__(self, set_loaders):
|
|
@@ -4,7 +4,8 @@ from numpy.random import RandomState
|
|
|
4
4
|
|
|
5
5
|
def get_balanced_batches(n_trials, rng, shuffle, n_batches=None,
|
|
6
6
|
batch_size=None):
|
|
7
|
-
"""Create indices for batches balanced in size
|
|
7
|
+
"""Create indices for batches balanced in size
|
|
8
|
+
(batches will have maximum size difference of 1).
|
|
8
9
|
Supply either batch size or number of batches. Resulting batches
|
|
9
10
|
will not have the given batch size but rather the next largest batch size
|
|
10
11
|
that allows to split the set into balanced batches (maximum size difference 1).
|
|
@@ -13,14 +14,12 @@ def get_balanced_batches(n_trials, rng, shuffle, n_batches=None,
|
|
|
13
14
|
----------
|
|
14
15
|
n_trials : int
|
|
15
16
|
Size of set.
|
|
16
|
-
rng :
|
|
17
|
+
rng : RandomState
|
|
17
18
|
|
|
18
|
-
shuffle :
|
|
19
|
+
shuffle : bool
|
|
19
20
|
Whether to shuffle indices before splitting set.
|
|
20
|
-
n_batches :
|
|
21
|
-
|
|
22
|
-
batch_size :
|
|
23
|
-
(Default value = None)
|
|
21
|
+
n_batches : int, optional
|
|
22
|
+
batch_size : int, optional
|
|
24
23
|
|
|
25
24
|
Returns
|
|
26
25
|
-------
|
|
@@ -89,28 +88,89 @@ class BalancedBatchSizeIterator(object):
|
|
|
89
88
|
self.rng = RandomState(328774)
|
|
90
89
|
|
|
91
90
|
|
|
91
|
+
class ClassBalancedBatchSizeIterator(object):
|
|
92
|
+
"""
|
|
93
|
+
Create batches of balanced size, that are also balanced per class, i.e.
|
|
94
|
+
each class should be sampled roughly with the same frequency during
|
|
95
|
+
training.
|
|
96
|
+
|
|
97
|
+
Parameters
|
|
98
|
+
----------
|
|
99
|
+
batch_size: int
|
|
100
|
+
Resulting batches will not necessarily have the given batch size
|
|
101
|
+
but rather the next largest batch size that allows to split the set into
|
|
102
|
+
balanced batches (maximum size difference 1).
|
|
103
|
+
"""
|
|
104
|
+
|
|
105
|
+
def __init__(self, batch_size):
|
|
106
|
+
self.batch_size = batch_size
|
|
107
|
+
self.rng = RandomState(328774)
|
|
108
|
+
|
|
109
|
+
def get_batches(self, dataset, shuffle):
|
|
110
|
+
n_trials = dataset.X.shape[0]
|
|
111
|
+
batches = get_balanced_batches(n_trials,
|
|
112
|
+
batch_size=self.batch_size,
|
|
113
|
+
rng=self.rng,
|
|
114
|
+
shuffle=shuffle)
|
|
115
|
+
if shuffle:
|
|
116
|
+
n_classes = np.max(dataset.y) + 1
|
|
117
|
+
class_probabilities = [np.mean(dataset.y == i_class)
|
|
118
|
+
for i_class in range(n_classes)]
|
|
119
|
+
class_probabilities = np.array(class_probabilities)
|
|
120
|
+
# choose trials in inverse probability of class
|
|
121
|
+
trial_probabilities = [1.0 / class_probabilities[y]
|
|
122
|
+
for y in dataset.y]
|
|
123
|
+
trial_probabilities = np.array(trial_probabilities) / np.sum(
|
|
124
|
+
trial_probabilities)
|
|
125
|
+
i_trial_to_balanced = self.rng.choice(n_trials, n_trials,
|
|
126
|
+
p=trial_probabilities)
|
|
127
|
+
|
|
128
|
+
for batch_inds in batches:
|
|
129
|
+
if shuffle:
|
|
130
|
+
batch_inds = [i_trial_to_balanced[i_trial] for i_trial in
|
|
131
|
+
batch_inds]
|
|
132
|
+
batch_X = dataset.X[batch_inds]
|
|
133
|
+
batch_y = dataset.y[batch_inds]
|
|
134
|
+
|
|
135
|
+
# add empty fourth dimension if necessary
|
|
136
|
+
if batch_X.ndim == 3:
|
|
137
|
+
batch_X = batch_X[:, :, :, None]
|
|
138
|
+
yield (batch_X, batch_y)
|
|
139
|
+
|
|
140
|
+
def reset_rng(self):
|
|
141
|
+
self.rng = RandomState((4, 7, 2017))
|
|
142
|
+
|
|
143
|
+
|
|
92
144
|
class CropsFromTrialsIterator(object):
|
|
93
|
-
|
|
94
|
-
|
|
95
|
-
|
|
96
|
-
|
|
97
|
-
|
|
98
|
-
|
|
99
|
-
|
|
100
|
-
|
|
101
|
-
|
|
102
|
-
|
|
103
|
-
|
|
104
|
-
|
|
105
|
-
|
|
106
|
-
|
|
107
|
-
|
|
108
|
-
|
|
109
|
-
|
|
145
|
+
"""
|
|
146
|
+
Iterator sampling crops out the trials so that each sample
|
|
147
|
+
(after receptive size of the ConvNet) in each trial is predicted.
|
|
148
|
+
|
|
149
|
+
Predicting the given input batches can lead to some samples
|
|
150
|
+
being predicted multiple times, if the receptive field size
|
|
151
|
+
(input_time_length - n_preds_per_input + 1) is not a divisor
|
|
152
|
+
of the trial length. :func:`compute_preds_per_trial_for_set`
|
|
153
|
+
can help with removing the overlapped predictions again for evaluation.
|
|
154
|
+
|
|
155
|
+
Parameters
|
|
156
|
+
----------
|
|
157
|
+
batch_size: int
|
|
158
|
+
input_time_length: int
|
|
159
|
+
Input time length of the ConvNet, determines size of batches in
|
|
160
|
+
3rd dimension.
|
|
161
|
+
n_preds_per_input: int
|
|
162
|
+
Number of predictions ConvNet makes per one input. Can be computed
|
|
163
|
+
by making a forward pass with the given input time length, the
|
|
164
|
+
output length in 3rd dimension is n_preds_per_input.
|
|
165
|
+
|
|
166
|
+
See Also
|
|
167
|
+
--------
|
|
168
|
+
braindecode.experiments.monitors.compute_preds_per_trial_for_set : Assigns predictions to trials, removes overlaps.
|
|
169
|
+
"""
|
|
170
|
+
def __init__(self, batch_size, input_time_length, n_preds_per_input):
|
|
110
171
|
self.batch_size = batch_size
|
|
111
172
|
self.input_time_length = input_time_length
|
|
112
173
|
self.n_preds_per_input = n_preds_per_input
|
|
113
|
-
self.check_preds_smaller_trial_len = check_preds_smaller_trial_len
|
|
114
174
|
self.rng = RandomState((2017, 6, 28))
|
|
115
175
|
|
|
116
176
|
def reset_rng(self):
|
|
@@ -123,21 +183,25 @@ class CropsFromTrialsIterator(object):
|
|
|
123
183
|
i_trial_starts = [n_receptive_field - 1] * len(dataset.X)
|
|
124
184
|
i_trial_stops = [trial.shape[1] for trial in dataset.X]
|
|
125
185
|
|
|
126
|
-
|
|
127
|
-
|
|
128
|
-
|
|
129
|
-
|
|
186
|
+
# Check whether input lengths ok
|
|
187
|
+
input_lens = i_trial_stops
|
|
188
|
+
for i_trial, input_len in enumerate(input_lens):
|
|
189
|
+
assert input_len >= self.input_time_length, (
|
|
190
|
+
"Input length {:d} of trial {:d} is smaller than the "
|
|
191
|
+
"input time length {:d}".format(input_len, i_trial,
|
|
192
|
+
self.input_time_length))
|
|
193
|
+
start_stop_blocks_per_trial = _compute_start_stop_block_inds(
|
|
130
194
|
i_trial_starts, i_trial_stops, self.input_time_length,
|
|
131
|
-
self.n_preds_per_input,
|
|
195
|
+
self.n_preds_per_input, check_preds_smaller_trial_len=True)
|
|
132
196
|
for i_trial, trial_blocks in enumerate(start_stop_blocks_per_trial):
|
|
133
197
|
assert trial_blocks[0][0] == 0
|
|
134
198
|
assert trial_blocks[-1][1] == i_trial_stops[i_trial]
|
|
135
199
|
|
|
136
|
-
return self.
|
|
200
|
+
return self._yield_block_batches(dataset.X, dataset.y,
|
|
137
201
|
start_stop_blocks_per_trial,
|
|
138
202
|
shuffle=shuffle)
|
|
139
203
|
|
|
140
|
-
def
|
|
204
|
+
def _yield_block_batches(self, X, y, start_stop_blocks_per_trial, shuffle):
|
|
141
205
|
# add trial nr to start stop blocks and flatten at same time
|
|
142
206
|
i_trial_start_stop_block = [(i_trial, start, stop)
|
|
143
207
|
for i_trial, block in
|
|
@@ -146,8 +210,6 @@ class CropsFromTrialsIterator(object):
|
|
|
146
210
|
i_trial_start_stop_block = np.array(i_trial_start_stop_block)
|
|
147
211
|
if i_trial_start_stop_block.ndim == 1:
|
|
148
212
|
i_trial_start_stop_block = i_trial_start_stop_block[None,:]
|
|
149
|
-
#if shuffle:
|
|
150
|
-
# self.rng.shuffle(i_trial_start_stop_block)
|
|
151
213
|
|
|
152
214
|
blocks_per_batch = get_balanced_batches(len(i_trial_start_stop_block),
|
|
153
215
|
batch_size=self.batch_size,
|
|
@@ -155,29 +217,39 @@ class CropsFromTrialsIterator(object):
|
|
|
155
217
|
shuffle=shuffle)
|
|
156
218
|
for i_blocks in blocks_per_batch:
|
|
157
219
|
start_stop_blocks = i_trial_start_stop_block[i_blocks]
|
|
158
|
-
batch =
|
|
159
|
-
X, y, start_stop_blocks
|
|
220
|
+
batch = _create_batch_from_i_trial_start_stop_blocks(
|
|
221
|
+
X, y, start_stop_blocks)
|
|
160
222
|
yield batch
|
|
161
223
|
|
|
162
224
|
|
|
163
|
-
def
|
|
164
|
-
|
|
165
|
-
|
|
166
|
-
|
|
167
|
-
|
|
168
|
-
|
|
169
|
-
|
|
170
|
-
|
|
171
|
-
|
|
172
|
-
|
|
173
|
-
|
|
225
|
+
def _compute_start_stop_block_inds(i_trial_starts, i_trial_stops,
|
|
226
|
+
input_time_length, n_preds_per_input,
|
|
227
|
+
check_preds_smaller_trial_len):
|
|
228
|
+
"""
|
|
229
|
+
Compute start stop block inds for all trials
|
|
230
|
+
Parameters
|
|
231
|
+
----------
|
|
232
|
+
i_trial_starts: 1darray/list of int
|
|
233
|
+
Indices of first samples to predict(!).
|
|
234
|
+
i_trial_stops: 1daray/list of int
|
|
235
|
+
Indices one past last sample to predict.
|
|
236
|
+
input_time_length: int
|
|
237
|
+
n_preds_per_input: int
|
|
238
|
+
check_preds_smaller_trial_len: bool
|
|
239
|
+
Check whether predictions fit inside trial
|
|
240
|
+
Returns
|
|
241
|
+
-------
|
|
242
|
+
start_stop_blocks_per_trial: list of list of (int, int)
|
|
243
|
+
Per trial, a list of 2-tuples indicating start and stop index
|
|
244
|
+
of the inputs needed to predict entire trial.
|
|
245
|
+
"""
|
|
174
246
|
# create start stop indices for all batches still 2d trial -> start stop
|
|
175
247
|
start_stop_blocks_per_trial = []
|
|
176
248
|
for i_trial in range(len(i_trial_starts)):
|
|
177
|
-
|
|
178
|
-
|
|
179
|
-
start_stop_blocks =
|
|
180
|
-
|
|
249
|
+
i_trial_start = i_trial_starts[i_trial]
|
|
250
|
+
i_trial_stop = i_trial_stops[i_trial]
|
|
251
|
+
start_stop_blocks = _get_start_stop_blocks_for_trial(
|
|
252
|
+
i_trial_start, i_trial_stop, input_time_length,
|
|
181
253
|
n_preds_per_input)
|
|
182
254
|
|
|
183
255
|
if check_preds_smaller_trial_len:
|
|
@@ -196,22 +268,38 @@ def compute_start_stop_block_inds(i_trial_starts, i_trial_stops,
|
|
|
196
268
|
return start_stop_blocks_per_trial
|
|
197
269
|
|
|
198
270
|
|
|
199
|
-
def
|
|
200
|
-
|
|
271
|
+
def _get_start_stop_blocks_for_trial(i_trial_start, i_trial_stop,
|
|
272
|
+
input_time_length, n_preds_per_input):
|
|
273
|
+
|
|
274
|
+
"""
|
|
275
|
+
Compute start stop block inds for one trial
|
|
276
|
+
Parameters
|
|
277
|
+
----------
|
|
278
|
+
i_trial_start: int
|
|
279
|
+
Index of first sample to predict(!).
|
|
280
|
+
i_trial_stops: 1daray/list of int
|
|
281
|
+
Index one past last sample to predict.
|
|
282
|
+
input_time_length: int
|
|
283
|
+
n_preds_per_input: int
|
|
284
|
+
Returns
|
|
285
|
+
-------
|
|
286
|
+
start_stop_blocks: list of (int, int)
|
|
287
|
+
A list of 2-tuples indicating start and stop index
|
|
288
|
+
of the inputs needed to predict entire trial.
|
|
289
|
+
"""
|
|
201
290
|
start_stop_blocks = []
|
|
202
|
-
i_window_stop =
|
|
291
|
+
i_window_stop = i_trial_start # now when we add sample preds in loop,
|
|
203
292
|
# first sample of trial corresponds to first prediction
|
|
204
|
-
while i_window_stop <
|
|
293
|
+
while i_window_stop < i_trial_stop:
|
|
205
294
|
i_window_stop += n_preds_per_input
|
|
206
|
-
i_adjusted_stop = min(i_window_stop,
|
|
295
|
+
i_adjusted_stop = min(i_window_stop, i_trial_stop)
|
|
207
296
|
i_window_start = i_adjusted_stop - input_time_length
|
|
208
297
|
start_stop_blocks.append((i_window_start, i_adjusted_stop))
|
|
209
298
|
|
|
210
299
|
return start_stop_blocks
|
|
211
300
|
|
|
212
301
|
|
|
213
|
-
def
|
|
214
|
-
n_preds_per_input):
|
|
302
|
+
def _create_batch_from_i_trial_start_stop_blocks(X, y, i_trial_start_stop_block):
|
|
215
303
|
Xs = []
|
|
216
304
|
ys = []
|
|
217
305
|
for i_trial, start, stop in i_trial_start_stop_block:
|