Optimization.jl
Optimization.jl copied to clipboard
SciML
Mathematical Optimization in Julia. Local, global, gradient-based and derivative-free. Linear, Quadratic, Convex, Mixed-Integer, and Nonlinear Optimization in one simple, fast, and differentiable inte...
The following code hangs for me: ``` using Optimization; using ModelingToolkit n = 100 u = zeros(2n) obj(u, _) = sum(u.^2) function cons(u, _) vcat(u[1:n] - u[n+1:2n], u[1:n] + u[n+1:2n])...
The reason for the `p` in the optimization problem definition is to allow for ChainRules support. Specifically: https://arxiv.org/abs/1804.05098 https://web.stanford.edu/~boyd/papers/pdf/diff_cvxpy.pdf https://link.springer.com/article/10.1007/s12532-012-0043-2 https://link.springer.com/article/10.1007/BF02055196 https://arxiv.org/abs/1810.13400 https://link.springer.com/article/10.1007/s12532-012-0043-2 equation (3) might be the clearest.
Optimization.jl has currently been fairly fast and loose with its handling of algorithm traits and high level error throwing, but we should fix that up as this package is now...
This actually causes some serious load time issues in some cases (system images, hint hint) and thus it would be good to remove the use of Requires.jl here.
Currently, the final state of the Adam and BFGS is dropped when solve stops. However, it can be beneficial to resume training, and then it would help to resume Adam...
Following @Vaibhavdixit02 example [here](https://discourse.julialang.org/t/ac-optimal-power-flow-in-various-nonlinear-optimization-frameworks/78486/31?u=ccoffrin), tried changing, ``` import ForwardDiff optprob = Optimization.OptimizationFunction(opf_objective, Optimization.AutoForwardDiff(); cons=opf_constraints) ``` to ``` import ModelingToolkit optprob = Optimization.OptimizationFunction(opf_objective, Optimization.AutoModelingToolkit(true,true); cons=opf_constraints) ``` I got the following error....
Is it possible to use jacobian coloring from SparseDiffTools in places like https://github.com/SciML/Optimization.jl/blob/24ad25d3e75816d454349156a7ac6bb094223ad2/src/function/zygote.jl#L44-L45
https://scimlbase.sciml.ai/dev/interfaces/Algorithms/#Abstract-SciML-Algorithms I think we're missing `AbstractOptimizationAlgorithm` and need to do an `AbstractGalacticOptimAlgorithm
https://galacticoptim.sciml.ai/stable/API/optimization_function/#Defining-Optimization-Functions-Via-AD ```jl julia> GalacticOptim.AutoReverseDiff(compile=true) ERROR: MethodError: no method matching GalacticOptim.AutoReverseDiff(; compile=true) Closest candidates are: GalacticOptim.AutoReverseDiff() at C:\Users\arno\.julia\packages\GalacticOptim\lVdNA\src\function\reversediff.jl:1 got unsupported keyword argument "compile" Stacktrace: [1] top-level scope @ REPL[172]:1 ```