sktime: [BUG] `DtwPythonDist` output has randomly permuted rows/columns
Hi,
I tried using DtwPythonDist() to calculate distances between a dataset of time series. It worked well, but I noticed that I get different results every time I run it.
Here’s my code. I run it in my local machine using PyCharm.
dtwpy = DtwPythonDist()
dist_mat = dtwpy.transform(data)
I used it with two different datasets: the dataset with all equal lengths, and the dataset with unequal lengths. The DtwPythonDist() behaves the same. If I run it twice, I would get two totally different distance matrices.
Am I missing something? What should I do?
Thank you.
About this issue
- Original URL
- State: closed
- Created 3 years ago
- Comments: 17
Hi, I’m really sorry for the late reply. I just had the time to try it again with sktime version 0.29.0. It works now; the results don’t permute anymore. Thank you for your help.