alessiozanga / pyeeglab Goto Github PK

Hi,
I'm trying to create the TUH Artifact Dataset as follows:

dataset = TUHEEGArtifactDataset('~/Documents/eeg/tuh_artifact_eeg/edf')
dataset.set_cache_manager(PickleCache('~/Documents/eeg/export'))

preprocessing = Pipeline([
    CommonChannelSet(),
    LowestFrequency(),
    ToDataframe(),
    MinMaxCentralizedNormalization(),
    ToNumpy()
])
dataset = dataset.set_pipeline(preprocessing).load()
data, labels = dataset['data'], dataset['labels']

But it fails with the following message:
ValueError: tmin (600.9454) must be less than tmax (600.9)

I'm using the v2.0.0 of the dataset: https://www.isip.piconepress.com/projects/tuh_eeg/downloads/tuh_eeg_artifact/v2.0.0/edf/

Hi,
Thanks for the nice library. Just a small comment, the first example in the README is:

from pyeeglab import *
dataset = TUHEEGAbnormalDataset()
pipeline = Pipeline([
    CommonChannelSet(),
    LowestFrequency(),
    ToDataframe(),
    MinMaxCentralizedNormalization(),
    DynamicWindow(8),
    ToNumpy()
])
dataset = dataset.set_pipeline(preprocessing).load()
data, labels = dataset['data'], dataset['labels']

but it should be:

from pyeeglab import *
dataset = TUHEEGAbnormalDataset()
preprocessing = Pipeline([
    CommonChannelSet(),
    LowestFrequency(),
    ToDataframe(),
    MinMaxCentralizedNormalization(),
    DynamicWindow(8),
    ToNumpy()
])
dataset = dataset.set_pipeline(preprocessing).load()
data, labels = dataset['data'], dataset['labels']

Hi,

Is there any plan to add support for TUH EEG Epilepsy dataset?

https://www.isip.piconepress.com/projects/tuh_eeg/downloads/tuh_eeg_epilepsy/

Thanks.

Hi! I'm trying to use your dataset on Mac OS X, and sched_getaffinity is not included there. Is there any hotfix for this? I'm considering using simple Pool(2) instead of len you provided, is that correct?

I have error when try run example

`
RuntimeError:
An attempt has been made to start a new process before the
current process has finished its bootstrapping phase.

    This probably means that you are not using fork to start your
    child processes and you have forgotten to use the proper idiom
    in the main module:

        if __name__ == '__main__':
            freeze_support()
            ...

    The "freeze_support()" line can be omitted if the program
    is not going to be frozen to produce an executable.

Process finished with exit code -1
`

Hi Alessio,

Thanks for open sourcing the library. I am trying to process it tuh_eeg_seizure dataset. But the process stuck at Dumping cache file . Any ideas what's causing this? if it is normal how long shall I wait for it to finish? Here's the code snippet I used:

path = "./tuh_eeg_seizure/v1.5.2/edf/train"
tuh_artifact_dataset = TUHEEGSeizureDataset(path)

preprocessing = Pipeline([
    CommonChannelSet(),
    LowestFrequency(),
    ToDataframe(),
    MinMaxCentralizedNormalization(),
    DynamicWindow(8),
    ToNumpy()
])

Shouldn't the second element of the list be 03_tcp_ar_a, as specified in the directory tree of the artifact dataset? https://www.isip.piconepress.com/projects/tuh_eeg/downloads/tuh_eeg_artifact/v1.0.0/edf/

Hi Alessio,

I am trying processing TUH EEG Seizure dataset (v1.5.2) on a machine with 8 cpus and 64 GB of RAM but the process is taking really long (so far process is running for like 10 hours) and seems stuck at DEBUG: Environment variables:....... . Any idea how to resolve this and process it quickly!

Thanks in advance.

Hi,

There are train and dev splits available for TUH EEG Seizure dataset but it is unclear how train and dev splits are pre-processed from the following:

Could you please explain? Does it mean that the current implementation of tuh_eeg_seizure_dataset.py jointly process train and dev sets?

I tried using the path ./data/tuh_eeg_seizure/v1.5.2/edf/train to process training set and ./data/tuh_eeg_seizure/v1.5.2/edf/dev for processing development set independently but there is channel mismatch error during processing of development set.

Traceback (most recent call last): File "/opt/conda/lib/python3.7/multiprocessing/pool.py", line 121, in worker result = (True, func(*args, **kwds)) File "/opt/conda/lib/python3.7/multiprocessing/pool.py", line 47, in starmapstar return list(itertools.starmap(args[0], args[1])) File "/opt/conda/lib/python3.7/site-packages/PyEEGLab-0.9.4-py3.7.egg/pyeeglab/pipeline/pipeline.py", line 39, in _trigger_pipeline data = preprocessor.run(data, **kwargs) File "/opt/conda/lib/python3.7/site-packages/PyEEGLab-0.9.4-py3.7.egg/pyeeglab/preprocess/signal/channel_selector.py", line 29, in run data.set_channels(channels) File "/opt/conda/lib/python3.7/site-packages/PyEEGLab-0.9.4-py3.7.egg/pyeeglab/io/raw.py", line 63, in set_channels self.reader = self.open().drop_channels(channels) File "/opt/conda/lib/python3.7/site-packages/mne/channels/channels.py", line 845, in drop_channels return self._pick_drop_channels(idx) File "/opt/conda/lib/python3.7/site-packages/mne/channels/channels.py", line 859, in _pick_drop_channels pick_info(self.info, idx, copy=False) File "<decorator-gen-8>", line 21, in pick_info File "/opt/conda/lib/python3.7/site-packages/mne/io/pick.py", line 475, in pick_info raise ValueError('No channels match the selection.') ValueError: No channels match the selection. """