I have N time series of DIFFERENT lengths with i number of data points in each observation.

I would like to compute the similarity of each time series and generate M number of clusters.

I have tried Jaquard simliarity, and followed the link below

Time Series Similarity : Differing Lengths with R

but am looking for a similar solution in Python here (which I have found https://github.com/slaypni/fastdtw)

The other problem is the NUMBER of features in the dataset (it is not just X & Y). X = Time, but I have 5 vectors for Y.

import numpy as np
from scipy.spatial.distance import euclidean

from fastdtw import fastdtw
x = x
y = y
distance, path = fastdtw(x, y, dist=euclidean)

Works fine for uneven dimension size, but if I add multiple dimensions it kicks out an error. Should I do some transposing?

Finally, if I can calculate the distance between each pair, what is the best way to cluster them?


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.