Lbfgsb.jl

This is wrapper of the famous lbfgsb fortran library of Julia language inspired by this wrapper

version

0.0.1 (just usable)

Currently only for Linux with gfortran installed

File fotran routine is written by the authors of the following papers,

R. H. Byrd, P. Lu and J. Nocedal. A Limited Memory Algorithm for Bound Constrained Optimization, (1995), SIAM Journal on Scientific and Statistical Computing , 16, 5, pp. 1190-1208.
C. Zhu, R. H. Byrd and J. Nocedal. L-BFGS-B: Algorithm 778: L-BFGS-B, FORTRAN routines for large scale bound constrained optimization (1997), ACM Transactions on Mathematical Software, Vol 23, Num. 4, pp. 550 - 560.
J.L. Morales and J. Nocedal. L-BFGS-B: Remark on Algorithm 778: L-BFGS-B, FORTRAN routines for large scale bound constrained optimization (2011), to appear in ACM Transactions on Mathematical Software.

I am new to Julia language so the code will be cleaned gradually

First, run Pkg.clone("https://github.com/yuhonglin/Lbfgsb.jl"), then run Pkg.build("Lbfgsb")
Currently only provide a function called lbfgsb, see its option below,

ogFunc : Objective and gradient function. It accept current x and a gradient array g. It should return the objective function value and put the gradient into into g.
x : Initial value of x.
lb : lower bounds of each dimension of x. Set to -Inf if equals to [].
ub : upper bounds of each dimension of x. Set to Inf if equals to [].
btype : boundary types, see below,

btype	boundary type of corresponding dimension
0	unbounded
1	only lower bound
2	both lower and upper bound
3	only upper bound

Other parameters : see the paper above for reference (mostly does not need to modify)
iprint : printing level of the fortran routine, set to -1 if you does not want to print anything

BSD-3