Let's say I obtained a timeserie (in blue) (with some missing data) that is (as far as i understood) : - following a general trend between specific points - more or less cyclic


I have drawn the red curve myself. I need to find a way to obtain it in realtime... I mean, that even when I am in peak in the middle of the graph, it should show me the red line.

I want to use machine learning (tf, sklearn or statsmodel) to learn from few curves and identify the new one in realtime...

Have anyone face the same kind of problem ?

Notes :

  1. Realtime mean that i d'ont have the full input (blue). ie I have it until may 17 10:00, i have to find the output (red) at may 17 10:00.

  2. There no model for getting red from blue... In fact, peaks have to be ignore to get the red curve.

  3. The zeros come from the fact that there is a big change in the blue curve, and the red curve's "origin" will change and go up

  • $\begingroup$ you need to look at signal processing and more specifically at smoothing. there are good python notebooks out there. Not sure it will really be able to get the new peak 'instantly'. $\endgroup$ – lcrmorin Feb 13 at 17:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.