Open aghababa opened 3 years ago
Why do I get negative values from distance with some values for gamma? For example, on a trajectory dataset, I had to choose gamma=1e-15 in order to get positive values. But this is nothing than dtw!
The dataset I used is available here: https://archive.ics.uci.edu/ml/datasets/GPS+Trajectories
In other datasets that I examined, gamma needs to be chosen carefully and small enough to works. Am I right?
Negative distances seem to be a possible output of the algorithm, a feature which is addressed in a later work that is also implemented at: https://github.com/google-research/soft-dtw-divergences
Why do I get negative values from distance with some values for gamma? For example, on a trajectory dataset, I had to choose gamma=1e-15 in order to get positive values. But this is nothing than dtw!
The dataset I used is available here: https://archive.ics.uci.edu/ml/datasets/GPS+Trajectories
In other datasets that I examined, gamma needs to be chosen carefully and small enough to works. Am I right?