You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It also took very long time for me. The time consuming part is looping through EEG signals and extracting features for each single sliding window. I think two ways this can be speed up is
parallelize the preprocessing,
the time slots with seizure are small so this is a highly imabalanced problem. At the end of the day you probably don't need >30 minutes of non-seizure data to learn how to detect the <10s of seizure epochs. So cropping the non-seizure times in a way that it still provides enough diversity of non-seizure patterns would be another option.
Unfortunately I don't have time to improve the code. If someone else wants to develop some improvement feel free to do so or reach out to me at [email protected]
Hi. The preprocessing.py file takes very long to run on my desktop.
Can someone suggest some configuration (on AWS) where this can be run faster?
The text was updated successfully, but these errors were encountered: