You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Generic multiscale approach is convonving with some kernel and subsequent resampling (for the appropriate scale). The problems of it is computational complexity (more than O(NlogN), unless we reuse subscales relation), which is better done with webgl or SIMD etc, which is too heavy for the simple thing.
That would allow for dy/gl-waveform#9, because difference of pot2 scales is too prominent; and for future NN works.
The text was updated successfully, but these errors were encountered:
Generic multiscale approach is convonving with some kernel and subsequent resampling (for the appropriate scale). The problems of it is computational complexity (more than O(NlogN), unless we reuse subscales relation), which is better done with webgl or SIMD etc, which is too heavy for the simple thing.
That would allow for dy/gl-waveform#9, because difference of pot2 scales is too prominent; and for future NN works.
The text was updated successfully, but these errors were encountered: