4 ms·
Where's the problem? Just adjust your distance metric accordingly. Or if you have a kNN kernel already available which takes inputs in Euclidean space, just m
by cperciva 17y ago
Where's the problem? Just adjust your distance metric accordingly.
Or if you have a kNN kernel already available which takes inputs in Euclidean space, just map a dimension t into two dimensions, sin(t) and cos(t).
- Bappz 17y agoOf course, thanks.