0
$\begingroup$

I'm currently trying to use dynamic time warping to group a large number of time-series I have. Unfortunately, each series has around 400000 entries, and Python can't handle creating a 400000 x 400000 matrix. Is there a another way to use DTW that doesn't use as much space?

$\endgroup$

1 Answer 1

0
$\begingroup$

If you just need the distance (not the warping path) then DTW is just O(n) not O(n^2). Does that solve your problem? Useful tutorial https://www.cs.unm.edu/~mueen/DTW.pdf

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.