David Widmann
David Widmann
Shouldn't this have been a breaking major release? It completely broke my project (that due to Catalyst still has to use MTK v8) when running `Pkg.update()`. _Originally posted by @devmotion...
The PR backports #2719 to a MTK v8 on a new `backport_8` branch. I can't use MTK v9 yet due to the incompatibility with Catalyst, so #2719 is not sufficient...
In the following benchmarks the samplers in Distributions are as fast as the implementations in this package: ```julia using Distributions, PoissonRandom, StatsFuns using Plots function n_count(rng, λ, n) tmp =...
Fixes https://github.com/SciML/StochasticDelayDiffEq.jl/issues/45.
I had to spend a lot of time on debugging issues with the Whedon bot in https://github.com/JuliaCon/proceedings-review/issues/109, until I noticed that apparently the remaining culprit was the acknowledgement environment defined...
This PR adds some irrationals that seem to be "missing" and would be useful e.g. in SpecialFunctions. E.g., currently `pi` (in base), `2 * pi`, `4 * pi`, `pi /...
The definitions of `*` and `/` for pd matrices are too optimistic: In general, the resulting matrix is not pd (or psd) anymore, so we should not return a pd...
Fixes #176.
The PR fixes the first point in #31 and starts the sequence at zero (see #31 and [Owen's paper](https://arxiv.org/abs/2008.08051) referenced therein for the motivation). With this change the sequences generated...