Multidimensional Subsequence DTW

[JamberFX]

I am trying to use dtw_subsequence_path to get the distance and path between two variable length 9-dimensional time series. I have the time series stored in numpy arrays of shapes (9, 528) and (9, 1332) respectively. I use the code below to compare the two sequences but the results (also shown below) make it appear as though only a few datapoint of the two sequences were compared. I need the shorter of the two sequences to be used as the query to find the best match in the longer of the two sequences (subsequence match). How can I do this with tslearn?

path, dist = dtw_subsequence_path(shortvec, longvec)

print(path)
[(0, 5), (1, 5), (2, 6), (3, 6), (4, 6), (5, 6), (6, 6), (7, 7), (8, 8)]
print(dist) 
29.942063162380773