Asbjørn Nilsen Riseth
Asbjørn Nilsen Riseth
*This was originally at https://github.com/JuliaNLSolvers/ConstrainedOptim.jl/issues/4* Here's a list of things that I believe we should focus on next. If anyone is interested in helping out here that would be very...
Tracing is a bit messy. Could do with an improvement.
Ref https://github.com/JuliaNLSolvers/Optim.jl/pull/505#issuecomment-357463873 Needs https://github.com/JuliaLang/julia/issues/20791
If someone creates a new `Optimizer` that can support manifolds, it is useful to have tests that check the correct handling of this. I propose that we cook up -...
The non-allocating `GradientCache` constructor has a line that calls `eltype(df)`, but I can't see `df` anywhere in the arguments list? https://github.com/JuliaDiffEq/DiffEqDiffTools.jl/blob/master/src/gradients.jl#L57 https://github.com/JuliaDiffEq/DiffEqDiffTools.jl/blob/master/README.md#allocating-cache-constructor @dextorious / @ChrisRackauckas
We should create documentation, or at the very least explain the input and output structure for the linesearch algorithms.
In HagerZhang and MoreThuente we throw errors if the step direction is not a descent direction (that is, d\phi(0) \geq 0). No tests are made in BackTracking and StrongWolfe, and...
I'd like to see a flexible logging / tracing functionality here. Currently there is little info provided when things go wrong inside the linesearches. Finiteness tests such as #101 should...
- [ ] Test linesearch behaviour when alpha = NaN, Inf or negative - [ ] Add counter tests (we can create `OnceDifferentiable` objects using NLSolversBase)
Morethuente, hagerzhang and strongwolfe all evaluate `df.g!` or `df.fg!`, and should thus report return any step that will give `g(x+alpha*p)` within tolerance from Optim, NLsolve etc.