ChristophRaab / rrslvq

Code release of Reactive Robust Learning Vector Quantization

Home Page:https://www.sciencedirect.com/science/article/abs/pii/S0925231220305063

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Distinguishing abrupt and gradual drift

szamani20 opened this issue · comments

I run an implementation of your work based on skmultiflow implementation here:
https://scikit-multiflow.readthedocs.io/en/stable/api/generated/skmultiflow.drift_detection.KSWIN.html

But I couldn't find any way to distinguish between abrupt and gradual drift that you discuss in the paper. Could you please explain whether this feature is implemented or not? Thanks.

commented

Currently, the ConceptDriftStream Class of skmultiflow simulates the abrupt drift. Please see the documentation for more details.
Further, the skmultiflow package has some gradual drift streams like the Hyperplane Generator to simulate this kind of drift.

To use KSWIN within this setting, you can choose appropriate parameters, as discussed in the paper. For example, a small window size if the rate of drift is assumed to be high. However, there is currently no option to specify kwsin to a certain drift in terms of just selecting the drift it has to deal with.