Extracted Phasic Driver is not non-negative

[rivasd]

Hi! Thank you for taking the time to port this to Python :)

I am however a bit confused, since I thought Ledalab implemented an algorithm that constrained the phasic driver to be non-negative (since negative sudomotor nerver activations make no sense). The current output, that I can get with leda2.analysis.phasicDriverRaw is not non-negative and very dense, I assume because it perfectly reconstructs the data when summed with the tonic component

You can see what I mean with the green curve here.

Any pointers on how to better specify what analysis I want from Ledapy ?

[MarcoFiletti]

Hi, yes phasicDriverRaw is the raw (uncorrected) data.

To get the phasic driver, use leda2.analysis.phasicData, that one will be non-negative