This post will tell you how to install CUTEst using a different tool that makes it much easier. Also, I’ll install CUTEst.jl, the CUTEst interface for Julia.
Edit: Now, CUTEst.jl install CUTEst by itself. Check this post. Also, for Linux, I’ve created this CUTEst installer, which should be easier to use. February, 11, 2017.
Edit: Some corrections were made on February, 15, 2016.
Edit: Some corrections were made on November, 11, 2015.
This post will tell you how to install CUTEst using a different tool that makes
it much easier. Also, I’ll install CUTEst.jl, the CUTEst interface for Julia.
Edit:Now, CUTEst.jl install CUTEst by itself. Check this
post.
Also, for Linux, I’ve created this CUTEst
installer, which should be
easier to use. February, 11, 2017.
Edit:Some corrections were made on February, 15, 2016.
Edit:Some corrections were made on November, 11, 2015.
By now you probably know CUTEst,
the repository for testing and comparing nonlinear programming algorithms.
It’s widely used in the community for some time (considering CUTE and CUTEr,
the previous versions).
If not, this is a good change to test it, using Julia to play around.
This is a not a post to convince you to use Julia, but I have to say that it is
much easier to use CUTEst on Julia than on MatLab.
So, if you are starting on it, I suggest you take a look.
We will use Homebrew to install CUTEst, for two reasons:
It’s much easier (when you learn it)
Julia requires shared libraries, that the original installation did not
provide.
Homebrew is a kind of package manager (such as apt-get, pip, etc.).
For linux, there are many things that we don’t need from Homebrew, because you
normally already have a package manager. However, Homebrew is widely used by OSX
users, so it has a lot of packages.
The linux version is Linuxbrew.
The installation is quite simple:
Install brew
Install CUTEst
Install CUTEst.jl
I just made these steps and record my terminal, so you can check Asciinema, or the embedded version on the
bottom of the page. Be warned, though, that I was “cold running” them, so some
parts are very slow.
To install brew, I recommend you check the page. For the impatient,
To install CUTEst, read the tap cutest.
Again, for the impatient
brew tap optimizers/cutest
brew install cutest
brew install mastsif
for f in archdefs mastsif sifdecode cutest;do\echo"source $(brew --prefix$f)/$f.bashrc">>\$HOME/.bashrc;\done
echo'export LD_LIBRARY_PATH="$HOME/.linuxbrew/lib:$LD_LIBRARY_PATH"'>>$HOME/.bashrc
source$HOME/.bashrc
This should get CUTEst installed.
Notice the LD_LIBRARY_PATH variable, which points to where the CUTEst library
will be.
Test it with
brew test sifdecode
brew test cutest
That’s it. You have CUTEst installed to use with Fortran or C.
A can’t provide a simple example, because they aren’t simple (enough).
I’ll now go to Julia, and I recommend you try it.
To install Julia, go to their page, then downloads, then download the
static version of the stable release (or do what you want, I’m not your boss).
Then, in julia, to install CUTEst.jl,
issue the commands
If you’re familiar with CUTEst, you can use the classic functions cfn and ufn too, in the default way (as called from C) or a more Julian way.
This would become too long to explain now, so I’ll make a post in a few days (or
months).
If you need it, please contact me.
This concludes the new installation of CUTEst.
Warning: Due to current limitations we cannot open two problems at the same
time in CUTEst without the possibility of a segmentation fault.
So, if you need to run cutest for a list of problems, I suggest you use a bash
script to loop over each problem and call your Julia code passing the problem as
an input argument.
Julia has native support for calling C and FORTRAN functions. There are also add on packages which provide interfaces to C++, R and Python. We’ll have a brief look at the support for C and R here. Further details on these and the other supported languages can be found on github.
Why would you want to call other languages from within Julia? Here are a couple of reasons:
to access functionality which is not implemented in Julia;
to exploit some efficiency associated with another language.
The second reason should apply relatively seldom because, as we saw some time ago, Julia provides performance which rivals native C or FORTRAN code.
C
C functions are called via ccall(), where the name of the C function and the library it lives in are passed as a tuple in the first argument, followed by the return type of the function and the types of the function arguments, and finally the arguments themselves. It’s a bit klunky, but it works!
This function will not be vectorised by default (just try call csqrt() on a vector!), but it’s a simple matter to produce a vectorised version using the @vectorize_1arg macro.
julia> @vectorize_1arg Real csqrt;
julia> methods(csqrt)
# 4 methods for generic function "csqrt":
csqrt{T<:Real}(::AbstractArray{T<:Real,1}) at operators.jl:359
csqrt{T<:Real}(::AbstractArray{T<:Real,2}) at operators.jl:360
csqrt{T<:Real}(::AbstractArray{T<:Real,N}) at operators.jl:362
csqrt(x) at none:6
Note that a few extra specialised methods have been introduced and now calling csqrt() on a vector works perfectly.
I’ll freely admit that I don’t dabble in C too often these days. R, on the other hand, is a daily workhorse. So being able to import R functionality into Julia is very appealing. The first thing that we need to do is load up a few packages, the most important of which is RCall. There’s great documentation for the package here.
julia> using RCall
julia> using DataArrays, DataFrames
We immediately have access to R’s builtin data sets and we can display them using rprint().
julia> rprint(:HairEyeColor)
, , Sex = Male
Eye
Hair Brown Blue Hazel Green
Black 32 11 10 3
Brown 53 50 25 15
Red 10 10 7 7
Blond 3 30 5 8
, , Sex = Female
Eye
Hair Brown Blue Hazel Green
Black 36 9 5 2
Brown 66 34 29 14
Red 16 7 7 7
Blond 4 64 5 8
We can also copy those data across from R to Julia.
However, for some complex objects there is no simple way to translate between R and Julia, and in these cases rcopy() fails. We can see in the case below that the object of class lm returned by lm() does not diffuse intact across the R-Julia membrane.
julia> "fit <- lm(bwt ~ ., data = MASS::birthwt)" |> rcopy
ERROR: `rcopy` has no method matching rcopy(::LangSxp)
in rcopy at no file
in map_to! at abstractarray.jl:1311
in map_to! at abstractarray.jl:1320
in map at abstractarray.jl:1331
in rcopy at /home/colliera/.julia/v0.3/RCall/src/sexp.jl:131
in rcopy at /home/colliera/.julia/v0.3/RCall/src/iface.jl:35
in |> at operators.jl:178
But the call to lm() was successful and we can still look at the results.
You can use R to generate plots with either the base functionality or that provided by libraries like ggplot2 or lattice.
julia> reval("plot(1:10)"); # Will pop up a graphics window...
julia> reval("library(ggplot2)");
julia> rprint("ggplot(MASS::birthwt, aes(x = age, y = bwt)) + geom_point() + theme_classic()")
julia> reval("dev.off()") # ... and close the window.
Watch the videos below for some other perspectives on multi-language programming with Julia. Also check out the complete code for today (including examples with C++, FORTRAN and Python) on github.
