David Widmann
David Widmann
This PR defines - `dim` - `whiten`, `whiten!` - `unwhiten`, `unwhiten!` - `quad`, `quad!` - `invquad`, `invquad!` - `X_A_Xt`, `Xt_A_X`, `X_invA_Xt`, `Xt_invA_X` for `Cholesky`. Ultimately, I think we should forward...
Hej! I like this package, and in principle I think it could be a nice building block for different GP packages such as https://github.com/PieterjanRobbe/GaussianRandomFields.jl or https://github.com/STOR-i/GaussianProcesses.jl. But I'm not sure...
To me it seems a bit unintuitive that `AbstractFFT` assumes that plans have a field `pinv` which is populated when calling `inv`. When computing the inverse via `inv` I don't...
This is basically the same issue/suggestion as https://github.com/JuliaMath/FFTW.jl/issues/77. In my opinion, it would make sense to apply in-place plans with `lmul!` already in the generic implementation of `fft!`, `bfft!`, and...
It is confusing that `Proposal`s such as `StaticProposal` accept arrays of proposals and one can also specify arrays of `Proposal`s in the `MetropolisHastings` sampler. Would it be sufficient to only...
Implements the suggestions in https://github.com/FluxML/Zygote.jl/pull/1307#discussion_r976398636 and unifies both `pairwise` adjoints.
This PR fixes the `promote_rule` definitions by using basically the same implementation as ForwardDiff (https://github.com/JuliaDiff/ForwardDiff.jl/blob/6a6443b754b0fcfb4d671c9a3d01776df801f498/src/dual.jl#L440-L452). Currently, abstract types (such as `Real`, `AbstractFloat` etc.) are not handled properly as `promote_rule` should...
This PR adds a Bernoulli distribution with the logit of the success rate as parameter. This is motivated by use cases e.g. in Turing (https://github.com/TuringLang/Turing.jl/issues/1890) where it is more convenient...
Just noticed that `ThreadSafeVarInfo` makes heavy use of `threadid`: https://github.com/TuringLang/DynamicPPL.jl/blob/715526ffa70292436e479e18d762e7ebf31c9181/src/threadsafe.jl#L22-L29 That should be changed since it is broken in Julia >= 1.8 if users don't use `@threads :static for ......
This PR adds `AbstractPDMat(::AbstractMatrix)`, `AbstractPDMat(::Cholesky)` and `AbstractPDMat(::CholTypeSparse)`. The main motivation is that this allows to specialize on the matrix type, which allows downstream packages such as Distributions (e.g. in https://github.com/JuliaStats/Distributions.jl/blob/9f9c5caf6cd99c5205a11ca1c8d6b6471fc9ce29/src/multivariate/mvnormal.jl#L201-L203)...