add docstring

src/optim.jl

@@ -1,11 +1,29 @@
 using Setfield
 
+"""
+    hyperoptim(f, candidates, algorithm = Optim.NelderMead(), opts = Optim.Options(); R = 30, η = 3, inner = RandomSampler(), threads = true, verbose = false, Rmin = 0.1, kwargs...)
+
+Multi-start optimization with Optim and Hyperband. The interface is similar to that of `Optim.optimize`, but the initial guess is replaced by a vector of candidate vectors. The resources are used to set the allowed time for optimization by `Optim.optimize`.
+
+# Arguments:
+- `f` a function from `Rⁿ -> R`, i.e., it takes a vector of parameters.
+- `candidates`: A vector of vectors
+- `algorithm`: One of the Optim algorithms
+- `opts`: `Optim.Options`
+- `inner`: The inner sampler
+- `threads`: Use threading to evaluate the cost function
+- `verbose`: DESCRIPTION
+- `Rmin`: The minimum amount of resources to use.
+- `kwargs`: Are sent to `Optim.optimize`.
+
+See `?hyperband` for the remaining arguments.
+"""
 function hyperoptim(f, candidates, algorithm = Optim.NelderMead(), opts = Optim.Options(); R=30, η=3, inner = RandomSampler(), threads=true, verbose = false, Rmin = 0.1, kwargs...)
 
     fun = function(r, pars)
         r = max(r, Rmin)
         @set! opts.time_limit = r
-        res = Optim.optimize(f, pars, algorithm, opts)
+        res = Optim.optimize(f, pars, algorithm, opts; kwargs...)
         verbose && println("Resources: $(round(r, digits=3)), value: $(round(minimum(res), digits=6))")
         Optim.minimum(res), Optim.minimizer(res)
     end