timbrer / ntxhackathon22 Goto Github PK

Hello,

I had a look at the data we collected for the idea to detect at least the eye-blinks.
It can be found at data/blinking_10sbreak_thenevery5secondsblink.csv/cleanedRecording.csv on the mpe/analysis_test branch.
Cleaned it up (extracted just the first four channels, which had real data, removed everything behind timestamp 198, converted the number-format).

But I can't really 'see' any real signal at the expected time-points when I visualize it as plot. Not even with some sliding average.

Any idea,what I could do wrong?