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 implementation in the answers for question 56 doesn't make use of the fact that a 2d gaussian is separable and can be implemented as two 1d gaussians - which is an order of magnitude faster on my machine. Also, the implementation is incorrect for mu != 0.0.
Thanks. You're right for the corrected version and for the separable version even though I would prefer to have first the non separable version and the separable as an additionla remark. Could you make a PR (modifying the ktx files)?
The implementation in the answers for question 56 doesn't make use of the fact that a 2d gaussian is separable and can be implemented as two 1d gaussians - which is an order of magnitude faster on my machine. Also, the implementation is incorrect for
mu != 0.0
.Test for
mu != 0.0
Raises:
Plot the functions:
Time the functions:
Output:
The text was updated successfully, but these errors were encountered: