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
The paper is interested in having multisamples sampled from an isotropic gaussian with mean equal to the point we're interested in, and variance as specified in the paper, so I just sample according to that gaussian directly instead of bothering with the helix which they used for speedup purposes (which I'm not certain gets us a ton of speedup tbh since we're just sampling from a gaussian which I don't believe is that expensive, tho I may be wrong on that one)
It seems that you did not sample 7 points on the helix for training as described in the paper, is it because it has not been completed yet?
The text was updated successfully, but these errors were encountered: