0

I need to split this cyclical data that has a variable period. I need to detect each sharp rise to split each into its own series, then overlay all they cycles on top of each other for statistical analysis. I was going to do this by just iterating across a python array (and may still), but wanted to give pandas a shot. Seems like this would be a fairly standard operation on cyclical data (not sure about the dynamic period though).

Smart people on stack overflow say not to iterate across a pandas object in this question. Is it any worse than iterating across a standard python array. Any tools or vocabulary that I should know to point me in the right direction?

enter image description here

ericnutsch
  • 275
  • 1
  • 2
  • 10

1 Answers1

0

Turns out this is called Dynamic Time Warping and is not trivial.

Pandas aside, dtw-python is a place to start.

ericnutsch
  • 275
  • 1
  • 2
  • 10