Does DTW implementation work with multivariate time series?

slaypni / fastdtw

A Python implementation of FastDTW

MIT License

774 stars 122 forks source link

Does DTW implementation work with multivariate time series? #16

Open hadware opened 6 years ago

hadware commented 6 years ago

The question is pretty self explanatory. If it does, maybe it'd be good to mention it in the README.md.

nabsabraham commented 5 years ago

@hadware did you figure out if it does?

hadware commented 5 years ago

It's been a long time since i've been working on a project that required this, but I think it does.

fmeccanici commented 4 years ago

Yes it does this. The numpy array should have the following structure ( I'm using it for Cartesian space trajectories):

    x                      y                  z

[[ 0.41493465 -0.39322514 1.14685164] [ 0.41546143 -0.3939345 1.14602891] [ 0.41591845 -0.39457089 1.14531319] ... [ 0.60821661 0.07143943 0.7978909 ] [ 0.60813073 0.0714391 0.79794052] [ 0.60804489 0.07143876 0.79798989]]

I had to transpose my numpy multidimensional array to acquire this structure.