Théo Galy-Fajou
Théo Galy-Fajou
@aviatesk solution should be applicable for any package right?
The whole idea is to be free enough to have `contourf` and `contour` are different `contour` having different range. It can be more interesting (to show distributions pdf for example),...
Sorry, I made a mistake: trying JuliaGaussianProcesses/KernelFunctions.jl#208 with this branch does not seem to solve the problem. Also precompiling this patch returns a surprising warning: ```julia ┌ Warning: Error requiring...
~~I officially don't know how to import rrule correctly via Requires...~~
I am not sure the failing checks are related to this PR...
Right, so you just followed their given API. I will make an issue in the Julia repo then.
I now see that this has been changed #795 But somehow this does not appear on the stable docs :  Seen on https://julianlsolvers.github.io/Optim.jl/stable/#user/config/#general-options Additionally, the mentioned piece of code...
Thanks for the advice! Technically, I implemented the algorithm 1, 2, 3 from this paper (same authors etc), "Bayesian inference for logistic models using Pólya-Gamma latent variables' https://arxiv.org/abs/1205.0310, just at...
Thanks for the monster review! I tried to address your comments and integrate them. The inference tests should tell us where things go wrong :)
MGF are not necessarily defined on the whole Real/Complex domain, what should be returned when getting a value out of bound? NaN or ArgumentError ?