Closed j2kun closed 2 years ago
TruncatedGaussian.GetMeanAndVariance
I found that, but then it leads to this 300-line section. It appears to be representing the truncated Gaussian in a generalized form, and it's so formula dense I'm not sure (a) how this exactly relates to the truncated Gaussian, nor (b) what parts of that correspond to the additive correction formulas in the paper, or if that's even the right way to think about it...
That code does correspond to the formulas in the paper. The reason it is complex is because it needs to work around the limited precision of floating-point numbers in C#.
Thanks!
Hi there, I'm new to this and just learning by reading some papers and comparing the code. I see equations 4.4-4.5 from Herbrich 05 and I'm trying to locate similar methods, but I don't see it in TruncatedGaussian.cs.
Could you kindly point me to where to look for this? Or is it disguised somehow else?