Mohamed Tarek

Return CGResult from cg and allow absolute tol

8

This fixes #236. I also allowed passing absolute tolerance as opposed to just a relative one. Since it makes sense for the residual to be

[Do not merge] Fix 246 - QR ortho fallback for LOBPCG

36

Fixes #246 by implementing QR orthogonalization as a fallback when CholQR fails. CC: @mfherbst, @haampie, @lobpcg

CGResult that saves info

2

It would be nice to have a `CGResult` struct that saves things like the last residual, number of iterations needed to converge, and options. If this is approved in principle,...

Enable the use of functions as preconditioners

2

Often the preconditioner can be defined as a function instead of a type with `ldiv!` method. For example, see https://github.com/JuliaMath/IterativeSolvers.jl/issues/258. It would be nice to be able to wrap the...

Checklist ticking off implemented algorithms with references?

3

This is a great package! It would be nice to have a checklist with references to help contributors or generally curious people much like this issue https://github.com/JuliaMath/IterativeSolvers.jl/issues/1.

ReverseDiff support in an extension

7

It would be cool to make ReverseDiff work with this using the `ReverseDiff.@grad` macro. Examples: https://github.com/JuliaDiff/ReverseDiff.jl/blob/master/test/MacrosTests.jl#L184

`using Flux, cuDNN` freezes, but `using Flux, CUDA, cuDNN` works

1

I am trying to load `Flux` and then `cuDNN` but it freezes at `using cuDNN`. However, loading `Flux`, then `CUDA`, then `cuDNN` works. Any idea why?

logdet differentiation fails

3

The following example doesn't work on latest Enzyme release and Julia 1.10. ```julia using LinearAlgebra import Enzyme f(x) = logdet([exp(x[1]);;]) f(rand(1)) Enzyme.gradient(Enzyme.Forward, f, rand(1)) ``` The part of the error...

KSDist is broken

6

```julia julia> using Distributions julia> logpdf(KSDist(1), 0.5) ERROR: MethodError: no method matching pdf(::KSDist, ::Float64) Closest candidates are: pdf(::Chernoff, ::Real) at C:\Users\user\.julia\packages\Distributions\wRw5p\src\univariate\continuous\chernoff.jl:168 pdf(::Kolmogorov, ::Real) at C:\Users\user\.julia\packages\Distributions\wRw5p\src\univariate\continuous\kolmogorov.jl:81 pdf(::DiscreteNonParametric{T,P,Ts,Ps} where Ps

When creating a dataset, it's often convenient to be able to upload a script that was used to generate the dataset. Generating derived datasets can be a complicated process and...