[julia-users] Re: CoinOptServices

Please open an issue on the CoinOptServices repository with as much information as you can to help reproduce the problem. On Tuesday, October 25, 2016 at 11:04:56 AM UTC-4, Frank Kampas wrote: > > When I run Pkg.build("CoinOptServices") using Julia 5.0 and Windows 10, I > get load errors and a

[julia-users] Re: CoinOptServices

I'm not aware of any broken functionality in CoinOptServices, but it hasn't been updated to get rid of deprecation warnings on 0.5. We'd welcome a PR. On Tuesday, October 25, 2016 at 9:01:29 AM UTC-4, Frank Kampas wrote: > > Is anybody working on fixing CoinOptServices? >

[julia-users] Re: ANN: JuMP 0.14 released

per function to convert the value of an optimization results, but also > simple real values. > """ > my_value(value::ForwardDiff.GradientNumber) = ForwardDiff.value(value) > my_value(value::Real) = value > my_value(val_vector::Vector) = [my_value(value) for value in val

[julia-users] ANN: JuMP 0.14 released

The JuMP team is happy to announce the release of JuMP 0.14. The release should clear most, if not all, deprecation warnings on Julia 0.5 and is compatible with ForwardDiff 0.2. The full release notes are here ,

[julia-users] Re: MUMPS server workaround?

On recent versions of Ubuntu you should be able to just install the "coinor-libipopt-dev" package and Pkg.build will pick up the system-wide Ipopt. On Wednesday, August 3, 2016 at 9:52:05 PM UTC-6, Sarah Koehler wrote: > > I have been trying to add Ipopt to Julia on a new computer (Ubuntu >

[julia-users] Re: JuliaOpt for nonlinear constrained optimization with around 500 variables

Hi Jorge, Gurobi is a state of the art solver. If your problem falls into a class supported by gurobi (e.g., convex quadratic), then you can rely on the results. Using Julia to specify the problem has a minimal effect on the solution time, specifically only in some preprocessing and generating

[julia-users] Re: ANN: JuMP 0.13 released (or, JuMP dumps camelCase)

eep up the good work! > > Uwe > > On Saturday, April 30, 2016 at 2:33:07 AM UTC+2, Miles Lubin wrote: >> >> The JuMP team is happy to announce the release of JuMP 0.13. >> >> This is the most visible JuMP release in quite a while since *all of >> JuMP's

[julia-users] Re: ANN: Optim.jl v0.4.5 released

Yes, the motivation was to allow passing options specific to the BFGS algorithm by using BFGS(foo=bar). On Tuesday, May 3, 2016 at 6:27:42 AM UTC-4, Kristoffer Carlsson wrote: > > Dispatching on Type{xx} is quite inconvenient and loocks the API a lot. By > using an instance you still have the

[julia-users] ANN: JuMP 0.13 released (or, JuMP dumps camelCase)

The JuMP team is happy to announce the release of JuMP 0.13. This is the most visible JuMP release in quite a while since *all of JuMP's macros and most exported methods have been renamed* to avoid the camelCase convention. The original naming scheme was put in place before the 0.1 release of

[julia-users] Re: Why does QR beat LU for sparse, square matrices?

Hey Jonas, I don't have an answer to this, but if you're looking for state-of-the-art performance for sparse linear algebra, I'd recommend using a proprietary library like PARDISO . Miles On Monday, April 25, 2016 at 4:49:40 PM UTC-4, Jonas Kersulis

[julia-users] Re: ANN: JuMP 0.12 released

rentiation and > otherwise forward mode differentation. > > > > On Wednesday, March 9, 2016 at 8:27:06 AM UTC-5, Miles Lubin wrote: >> >> On Wednesday, March 9, 2016 at 12:52:38 AM UTC-5, Evan Fields wrote: >>> >>> Great to hear. Two minor questions w

Re: [julia-users] ANN: A potential new Discourse-based Julia forum

I suspect julia-opt will wait and see how the julia-users transition goes. julia-opt has much lower volume than julia-users. On Wednesday, March 9, 2016 at 9:12:11 AM UTC-5, Johan Sigfrids wrote: > > What about other lists like julia-stats and julia-opt? Would they also > move to Discourse or

[julia-users] Re: ANN: JuMP 0.12 released

On Wednesday, March 9, 2016 at 12:52:38 AM UTC-5, Evan Fields wrote: > > Great to hear. Two minor questions which aren't clear (to me) from the > documentation: > - Once a user defined function has been defined and registered, can it be > incorporated into NL expressions via @defNLExpr? > Yes.

[julia-users] Re: GSOC2016

Hi Rohan, Head over to the julia-opt list and open a discussion there. Miles On Tuesday, March 8, 2016 at 8:56:39 AM UTC-5, Rohan Goel wrote: > > Hello everyone, > I want to contribute to the idea "Support for complex numbers within > Convex.jl". > Can someone please help me getting started

[julia-users] Re: ANN: JuMP 0.12 released

; > @setNLObjective( > nlp, > Min, > sum{f[i], i = 2:n-1} + f1 + fn > ) > > > status = solve(nlp); > > Thx, > > > JPD > > > Le samedi 27 février 2016 23:14:12 UTC+1, Miles Lubin

[julia-users] Re: How to have many arguments in a Jump.jl nonlinear model?

thanks again. > > > On Tuesday, March 1, 2016 at 12:24:05 PM UTC+11, Miles Lubin wrote: >> >> There's no syntax for this at the moment, it's a known issue. The problem >> is that JuMP's internal representation of nonlinear expressions doesn't >> allow vectors o

[julia-users] Re: How to have many arguments in a Jump.jl nonlinear model?

There's no syntax for this at the moment, it's a known issue. The problem is that JuMP's internal representation of nonlinear expressions doesn't allow vectors or matrices. For the moment we're targeting the use cases where the function is low dimensional. For box-constrained nonlinear

[julia-users] ANN: JuMP 0.12 released

The JuMP team is happy to announce the release of JuMP 0.12. This release features a complete rewrite of JuMP's automatic differentiation functionality, which is the largest change in JuMP's nonlinear optimization support since JuMP 0.5. Most of the changes are under the hood, but as

[julia-users] Re: ERROR: AssertionError: x.head == :escape

Could you provide some standalone code that reproduces this error? Hard to say without an example. On Sunday, October 25, 2015 at 12:24:34 PM UTC-4, Michela Di Lullo wrote: > > After the JuMP Package checkout > I get this error > > *ERROR: AssertionError: x.head == :escape* > > > while declaring

[julia-users] Re: Help on optimization problem

On Tuesday, October 20, 2015 at 11:12:44 PM UTC-4, Spencer Russell wrote: > > > 1. What solver is appropriate for this sort of problem? (or should I just > go with a spring-force algorithm like in GraphLayout.jl?) > That depends very directly on the precise formulation for the objective

[julia-users] Re: new function to calculate cycle basis in connected graphs

The Graphs.jl package is maintained by volunteers but has no active owner. You may want to look at contributing to the https://github.com/JuliaGraphs/LightGraphs.jl package instead. On Tuesday, October 13, 2015 at 6:28:53 AM UTC-4, Michela Di Lullo wrote: > > Hello, > > I would share into the

Re: [julia-users] Stateflow equivalent in Julia

It's not too clear to me what's trying to be accomplished here, but "gear" and "N" are optimization variables and can't be used inside conditional statements. You can use getValue() to query the value of the variables in an optimal solution after calling solve(). If these statements are

[julia-users] Re: ANN: JuMP 0.10 released

Hi Geoff, Very interesting work. I think it's fair to say that the current approach for modular models in JuMP is pretty ad hoc. The goal was develop a modeling system that's similar enough to what people are used to as opposed to putting forward a new philosophy for writing down and composing

Re: [julia-users] Creating a new version of a package

Pkg.tag(NaNMath) Pkg.publish() is all you should have to do.

[julia-users] Re: ANN: Major overhaul of ForwardDiff.jl

Thanks Jarrett! This work is the product of a Julia Summer of Code project. Please kick the tires, we're planning letting the new code sit on master for a bit before tagging a new release. Users of Julia 0.3 will be stuck on the old API, since the code makes extensive use the generated

Re: [julia-users] [ANN] SparseVectors

This was discussed pretty extensively in the github issue: https://github.com/JuliaLang/julia/issues/11324 On Wednesday, July 29, 2015 at 8:32:13 PM UTC-6, Christian Peel wrote: Nice! I saw somewhere recently that the plan was to move more out of base and into packages. I'm curious as to

Re: [julia-users] Re: Errors while trying to use JuMP

the julia directory and reinstalled it, but again with no luck and also again the old version of 0.1 was installed. On Sunday, June 14, 2015 at 2:34:30 PM UTC+2, Miles Lubin wrote: What is the output of versioninfo() (at the Julia command line)? Also, try deleting your ~/.julia directory

Re: [julia-users] Re: Organizing code around Jump models

, Miles Lubin a écrit : Hi Francois, Could you give an example of how you might organize this code in a more namespace heavy language like C++ or Java? I think this is quite a general question on how to structure large Julia applications, not really specific to JuMP. Best, Miles

[julia-users] Re: Organizing code around Jump models

Hi Francois, Could you give an example of how you might organize this code in a more namespace heavy language like C++ or Java? I think this is quite a general question on how to structure large Julia applications, not really specific to JuMP. Best, Miles On Wednesday, June 3, 2015 at

Re: [julia-users] JuliaCon registrations open

We now have a sheet to help JuliaCon attendees arrange for room sharing: https://docs.google.com/spreadsheets/d/1361VHhtFM5Cnu-S_cSUKFeUhrH33FepOKLCc0YLtosg/edit?usp=sharing On Friday, May 29, 2015 at 5:34:52 PM UTC+2, Matt Bauman wrote: Is anyone interested in splitting a room with two

[julia-users] Re: sparse Jacobian and/or coloring

Hi Mauro, Could you open an issue on ReverseDiffSparse to continue this discussion? Miles On Wednesday, May 27, 2015 at 3:42:36 PM UTC-4, Mauro wrote: For an implicit ode/dae solver, I want to provide automatic computation of a sparse Jacobian, either through automatic differentiation or

[julia-users] Re: Julia Summer of Code

We've put up a blog post (http://localhost:4000/blog/2015/05/jsoc-cfp/) with some more details on Julia Summer of Code. There are still a number of unfilled slots open; submission deadline is June 1st! On Friday, May 15, 2015 at 1:57:24 PM UTC-4, Viral Shah wrote: Folks, The Moore

[julia-users] Re: Julia Summer of Code

Sorry about that, here's the real link: http://julialang.org/blog/2015/05/jsoc-cfp/ On Sunday, May 24, 2015 at 12:42:27 AM UTC-4, Miles Lubin wrote: We've put up a blog post (http://localhost:4000/blog/2015/05/jsoc-cfp/) with some more details on Julia Summer of Code. There are still

[julia-users] Re: ANN node-julia 1.1.0

Very cool! On Friday, May 15, 2015 at 4:39:40 PM UTC-4, Jeff Waller wrote: This version supports 0.4.x using svec as well as previous 0.4.x and of course 0.3.x as well here's the docs http://node-julia.readme.io/ if interested. It's been a pretty long time + the svec change was breaking,

[julia-users] Re: Julia Summer of Code

for more info. Co-mentors: Miles Lubin and Theodore Papamarkou If this sounds cool and interesting to you, do get in touch!

[julia-users] Re: Error messages related to macros

This doesn't address all of your questions, but it's definitely a recognized issue that the error messages from macros leave much to be desired. See the discussions at: https://github.com/JuliaLang/julia/issues/1334 https://github.com/JuliaLang/julia/pull/6910

[julia-users] Re: Graphs.jl, Page Rank and Weighted Graphs

Hi Rita, Take a look at the LightGraphs package (https://github.com/JuliaGraphs/LightGraphs.jl), it should be a bit easier to use and less cranky than Graphs.jl. Miles On Wednesday, April 1, 2015 at 4:37:32 PM UTC-4, Rita Maria del Rio Chanona wrote: Hi, I was wondering if there is yet an

Re: [julia-users] how to model semicontinuous variables in Julia

Hi Michela, It looks like you're referring to JuMP, in which case: @defVar(m, a = x = b, SemiCont) should do the trick. See also the documentation: http://jump.readthedocs.org/en/release-0.8/refvariable.html?highlight=semicontinuous We prefer to direct optimization-related questions to

Re: [julia-users] how to model semicontinuous variables in Julia

, 2015 at 8:01:19 PM UTC-4, Michela Di Lullo wrote: Il giorno lunedì 30 marzo 2015 21:45:13 UTC+2, Miles Lubin ha scritto: Hi Michela, It looks like you're referring to JuMP, in which case: @defVar(m, a = x = b, SemiCont) should do the trick. See also the documentation: http

Re: [julia-users] How to test argument types available on a Function?

Take a look at applicable(). On Saturday, March 28, 2015 at 4:30:09 PM UTC-4, Sheehan Olver wrote: It currently works like this, which does work with f(x)=x, f(x,y) = x+y: try f(0) catch try f(0,0) catch f((0,0))

[julia-users] Re: Using JuMP module across multiple processes

Cross-post with https://groups.google.com/forum/#!topic/julia-opt/k19ZV2YgPgs On Sunday, March 22, 2015 at 12:15:37 PM UTC-4, Kirsten Westeinde wrote: Hello, I am trying to do some computations using JuMP and Cbc. Everything works as desired when I run it with one process, but the second I

[julia-users] Re: Question: `./` has no method matching ./(::Int64, ::AdditionAtom)

In the language of DCP, the division operator doesn't preserve convexity, so I'm assuming that's why it's not implemented. Looking at the list of available operations (http://convexjl.readthedocs.org/en/latest/operations.html), you may want to try inv_pos(x-t). On Saturday, March 21, 2015 at

[julia-users] Re: Point test convex hull

There are certainly more efficient specialized solutions for low dimensions like 3D, but you can test whether a point is contained in the convex hull by solving a linear programming feasibility problem: x = [1,1,1] # test point Z = [1 0 # points along columns 0 1 0 0] using JuMP m =

Re: [julia-users] Re: Community Support for Julia on Travis CI

Would that also run coverage by default when running Pkg.test() locally? That's probably not the expected behavior given that it will litter coverage files all over the place. On Saturday, February 14, 2015 at 7:22:24 AM UTC-5, Tony Kelman wrote: Making things too complicated will most likely

[julia-users] Re: benchmark with Julia 2x faster than C

Hi Arch, all, Thanks for looking into this, it's amazing to have experts here who understand the depths of compilers. I'm stubbornly having difficulty reproducing your timings, even though I see the same assembly generated for clang. I've tried on an i5-3320M and on an E5-2650 and on both,

Re: [julia-users] Announcing Convex.jl - Convex Optimization in Julia

I'm personally very pleased to see this. The JuMP team has worked closely with the Convex.jl team to make sure that we share a common infrastructure (through MathProgBase) to talk to solvers, and I don't think it's an exaggeration to say that this has resulted in an unprecedented level of

[julia-users] Re: Save state in a function

Just to clarify, my suggestion doesn't require any saved state between functions. In the context of the discussion, the state can be saved in the AbstractNLPEvaluator object. On Tuesday, February 3, 2015 at 1:11:36 PM UTC-5, Peter Simon wrote: I saw a suggestion from Miles Lubin in https

Re: [julia-users] benchmark with Julia 2x faster than C

. Though you probably did that for easier C compatibility. Not sure though. On Tue, Jan 27, 2015 at 3:39 PM, Miles Lubin miles...@gmail.com javascript: wrote: I'm working on a microbenchmark and found a surprising result (maybe not?) that Julia is about 2x faster than the same algorithm hand

Re: [julia-users] benchmark with Julia 2x faster than C

Here's the Julia assembly: function squareroot(x) it = x err = it*it-x # line 5 while abs(err) 1e-13 it = it - (err)/(2it) # line 7 err = it*it-x end return it end Source line: 5 push RBP mov RBP, RSP Source line: 5 vmulsd XMM1, XMM0, XMM0 vsubsd XMM2,

Re: [julia-users] Re: flip sparse matrices slow

It may be more efficient (given the current implementation) to filter the I, J, and V arrays directly and then call sparse() again. On Wednesday, January 7, 2015 12:48:31 PM UTC-7, Christoph Ortner wrote: Thanks for your replies - I will write a short piece of code that shows the

[julia-users] Re: pmap load balancing?

pmap already does dynamic scheduling, tasks get assigned dynamically as workers become available. This strategy can only do so much if the tasks are extremely imbalanced. On Thursday, January 1, 2015 9:10:51 PM UTC-5, Gabriel Mihalache wrote: Hello, all! I have a pmap over a large array,

[julia-users] Re: Why doesn't Dict() accept an output from zip?

This is a perfectly reasonable thing to want, and in fact has already been implemented in Julia 0.4-dev: julia Dict(zip(abc,123)) Dict{Char,Char} with 3 entries: 'b' = '2' 'c' = '3' 'a' = '1' On Friday, December 26, 2014 12:15:07 AM UTC-5, Ronald L. Rivest wrote: Since the Dict

[julia-users] Re: Gadfly in userimg.jl

I've been able to successfully precompile JuMP. What issues were you running into? On Friday, December 12, 2014 10:16:27 PM UTC-5, Matt Bauman wrote: I believe the trouble you're running into is that not all packages are precompile-friendly. While Gadfly by itself is, it dynamically tries

[julia-users] Re: how to test NaN in an array?

any(isnan,arr) is probably the most compact syntax. On Monday, December 15, 2014 3:33:55 PM UTC-5, Evan Pu wrote: 1 in [1,2,3] # returns true NaN in [NaN, 1.0, 2.0] # returns false how do I test if a float64 NaN is present in an array? I'm doing some numerical computation and it can have

[julia-users] Re: LIst Ipopt options when using JuMP

On Sunday, November 23, 2014 3:37:52 PM UTC-5, Pileas wrote: One last question that I have is whether we can use some kind of delimiter that can give us the result in a csv file, but not in one column, but rather in deifferent ones. It's better to just print out the file in the format

[julia-users] Re: C/MPI/SLURM = Julia/?/SLURM?

Two comments: 1) It's pretty easy to use MPI from Julia. For some use cases it may make more sense than Julia's built-in approach to parallelism, especially if you're already comfortable with MPI. Though if you're well served by pmap, that's much simpler. 2) It looks like the subproblem

[julia-users] Re: Bug in logdet?

In srand(1);a=rand(2,2);a=a*a';cholfact!(a);logdet(a) The issue is that cholfact!(a) destroys the data in a, you're not allowed to re-use it afterwords. You still need to save the result of cholfact! and use that object inside of logdet(), e.g., srand(1);a=rand(2,2);a=a*a';afact =

[julia-users] Re: ANN: FiniteDifferenceDerivatives.jl and questions about performance improvement

implementation of FDD.jl it would decrease the performance. Best, Paweł W dniu środa, 15 października 2014 22:06:40 UTC+2 użytkownik Miles Lubin napisał: Hi Paweł, How does your approach compare with the implementation of finite differences in Calculus.jl? It would be great if you could

[julia-users] Re: ANN: FiniteDifferenceDerivatives.jl and questions about performance improvement

Hi Paweł, How does your approach compare with the implementation of finite differences in Calculus.jl? It would be great if you could contribute any new techniques there. Miles On Wednesday, October 15, 2014 6:48:29 AM UTC-4, Paweł Biernat wrote: Hi, I have been experimenting with various

[julia-users] Re: Call for participation: High Performance Technical Computing in Dynamic Languages workshop (11/17/2014)

I'm not an organizer of this workshop, but just wanted to remind everyone of this and note that the deadline for submission is now August 25th, so get to writing those papers! On Monday, March 31, 2014 10:14:21 AM UTC-6, Jiahao Chen wrote: It is my great pleasure to announce a workshop at

Re: [julia-users] Optional import mechanism

This may do what you want (snippet from JuMP): if isdir(Pkg.dir(ArrayViews)) eval(Expr(:import,:ArrayViews)) const subarr = ArrayViews.view else const subarr = Base.sub end On Tuesday, August 19, 2014 7:38:27 PM UTC-6, Júlio Hoffimann wrote: Hi Joey, It’s a bit ugly, but this

Re: [julia-users] Suitable way to implement DSL for linear constraints

. On Monday, 18 August 2014 06:27:00 UTC+2, Miles Lubin wrote: Using eval() within a macro is almost always not what you should be doing. Macros should return code that may depend on the values of the input symbolically. In general, I would try writing down the expression that you want to generate

Re: [julia-users] Suitable way to implement DSL for linear constraints

JuMP is also a good place to look. On Saturday, August 16, 2014 8:11:04 AM UTC-6, Stefan Karpinski wrote: Have you looked at CVX? I feel like it may be relevant. On Aug 16, 2014, at 8:47 AM, Tomas Krehlik tomas@gmail.com javascript: wrote: I have a package that implements much of

Re: [julia-users] Package Popularity Ranking

It would be cool to be able to sort by github stars (and later, github clones) on pkg.julialang. On Wednesday, August 13, 2014 3:03:32 PM UTC-6, Iain Dunning wrote: Hi Michael, This is a problem I think about a lot, but definitely short on data. pkg.julialang.org has the GitHub stars,

Re: [julia-users] Behavior of size()

Perhaps the motivation for this behavior is that you can use 2d indexing on a 1d vector: julia x = [1,2,3]; julia x[1,1] 1 julia x[1,2] ERROR: BoundsError() Whether it makes sense to allow this is another question. On Wednesday, August 13, 2014 8:59:03 PM UTC-6, Isaiah wrote: size(x,n)

[julia-users] undergraduate research position available for JuMP (MIT only)

The JuMP team has undergraduate research positions (UROPs) for this fall, open to undergraduates at MIT. If you are interested, or know someone who may be interested in working on optimization software in Julia, please send us a note. Some background in numerical optimization would be

[julia-users] Re: what is the easiest way to convert array type?

How about convert(Vector{String},dt)? On Thursday, July 24, 2014 8:39:02 PM UTC-6, K leo wrote: I have an Array{Any,1}, but I want to convert to Array{String,1}. The only way I know is through comprehension. julia dt 3-element Array{Any,1}: 2010/1/4T15:00:00 2010/1/5T15:00:00

[julia-users] ANN: JuliaDiff -- differentiation tools in Julia

This is still a work in progress, but ahead of JuliaCon I'd like to announce JuliaDiff, a github organization and website (http://www.juliadiff.org/) for packages related to computing derivatives. This includes packages based on automatic differentiation. If you've never heard of AD, check out

Re: [julia-users] Optim/DualNumbers question (maybe a math question?)

There was a PR that was prematurely merged, it's still being discussed: https://github.com/JuliaDiff/DualNumbers.jl/pull/11 In the meantime, a generic Cholesky factorization has been proposed ( https://github.com/JuliaLang/julia/pull/7236) which would solve the original issue but not necessarily

Re: [julia-users] Re: ANN: Gumbo.jl (HTML parsing library)

On Thursday, June 19, 2014 8:34:46 PM UTC-6, Jameson wrote: poorly equipped to handle simultaneous multi-arch binary package installations agreed. I can't follow entirely why the GLPK steps are complicated, but I think it is because the person who packaged the binaries did not separate

[julia-users] Re: Least squares curve fitting package

Hi Paulo, This package seems to duplicate/complement the curve fitting functionality in Optim, which is capable of handling general models and providing statistical analysis of the fit. Please join us in the discussion here: https://github.com/JuliaOpt/Optim.jl/pull/56 about possibly merging

Re: [julia-users] Optim/DualNumbers question (maybe a math question?)

Any idea why type inference is failing on full(chol(A, :U)) ? It can't be determined from the types of the arguments what the result will be, because you could give different symbols to access different parts of the factorization. I think this was a design choice to make it easier to

[julia-users] Re: Optim/DualNumbers question (maybe a math question?)

Hi Thomas, This is definitely a reasonable thing to want to do, and there are a number of methods to differentiate through a matrix factorization. Julia doesn't yet have a generic chol() method (see https://github.com/JuliaLang/julia/issues/1629), but it does have a generic LU decomposition

Re: [julia-users] Optim.jl: unexpected results when using gradient-based methods

. Am Mittwoch, 21. Mai 2014 16:48:51 UTC+1 schrieb Miles Lubin: Just to extend on what John said, also think that if you can restructure the code to devectorize it and avoid using global variables, you'll see *much* better performance. The way to avoid globals is by using closures

Re: [julia-users] Optim.jl: unexpected results when using gradient-based methods

! into NumericExtensions Am Donnerstag, 22. Mai 2014 14:18:36 UTC+1 schrieb Miles Lubin: I was able to get a nearly 5x speedup by avoiding the matrix allocation and making the accumulators type stable: https://gist.github.com/mlubin/055690ddf2466e98bba6 How does this compare with Matlab now? On Thursday

Re: [julia-users] Optim.jl: unexpected results when using gradient-based methods

Just to extend on what John said, also think that if you can restructure the code to devectorize it and avoid using global variables, you'll see *much* better performance. The way to avoid globals is by using closures, for example: function foo(x, data) ... end ... data_raw =

[julia-users] Re: I cannot call using Requests in win XP

Relatedly, I think it's bad practice to simply put include(../deps/deps.jl) in the module, because this gives a meaningless error message to the user when the file doesn't exist. In Gurobi.jl, for example, we have: if isfile(joinpath(Pkg.dir(Gurobi),deps,deps.jl)) include(../deps/deps.jl)

Re: [julia-users] Re: How to use GLPK.exact ?

Looks good! On Friday, May 2, 2014 12:55:49 PM UTC-4, Stéphane Laurent wrote: Thank you everybody, almost every point discussed here is now written on my blog http://stla.github.io/stlapblog/posts/KantorovichWithJulia.html.

[julia-users] ANN: JuMP 0.5 released with support for nonlinear programming

We're happy to announce the release of JuMP 0.5, which includes support for modeling and solving convex and nonconvex nonlinear optimization problems which can be modeled by closed-form algebraic expressions. This extends JuMP's existing support for linear, mixed-integer, and second-order conic

[julia-users] Re: A Reader for the Harwell-Boeing Format

On Sunday, April 27, 2014 5:34:56 PM UTC-4, Dominique Orban wrote: Thanks a lot all for the feedback. There should now be a (yet unregistered) module named rb. It may not be the best name (it stands for Rutherford-Boeing). As this is my first module, I'd appreciate any feedback. Also I'm

[julia-users] Re: Trying to import a graph from NetworkX

Hi Ohad, There was indeed a bug in the graph() constructor. I've filed a PR: https://github.com/JuliaLang/Graphs.jl/pull/83 Miles On Saturday, April 26, 2014 12:19:41 PM UTC-4, Ohad wrote: To start playing with Julia I'm trying to import a graph from a NetworkX-generated JSON to the

Re: [julia-users] Re: How to use GLPK.exact ?

On Wednesday, April 23, 2014 3:40:02 PM UTC-4, Stéphane Laurent wrote: If I don't call GLPKMathProgInterface, does JuMP use an internal solver ? If a solver isn't specified, JuMP (actually MathProgBase) will search for an available solver and pick one by default. JuMP does not have an

Re: [julia-users] Re: How to use GLPK.exact ?

Cool! Glad to hear you got it working. Supporting exact coefficients in JuMP is technically possible, and I've opened an issue for it: https://github.com/JuliaOpt/JuMP.jl/issues/162. This will probably remain on the wishlist for a while. On Tuesday, April 22, 2014 2:28:01 PM UTC-4, Stéphane

Re: [julia-users] Re: How to use GLPK.exact ?

Where's the MathProgBase interface? :) On Wed, Apr 16, 2014 at 11:07 PM, Iain Dunning iaindunn...@gmail.com wrote: I implemented a version of simplex method for rational numbers - so you solve it exactly in pure Julia. https://github.com/IainNZ/RationalSimplex.jl Not for serious work - just

[julia-users] Re: Julia interface to the AMPL modeling language

Jacobian-vector products and Hessian-vector products (and Hessian of the Lagrangian - vector products). On Sunday, April 13, 2014 2:48:26 AM UTC-7, Miles Lubin wrote: Hi Dominique, This will definitely be very useful for accessing the large array of problem instances written in AMPL

[julia-users] Re: Julia interface to the AMPL modeling language

Hi Dominique, This will definitely be very useful for accessing the large array of problem instances written in AMPL. As for writing solvers in Julia around this format, I'm admittedly biased but I don't think it's an ideal approach. We already have a pure-Julia implementation of AD for

Re: [julia-users] Re: How to use GLPK.exact ?

By the way for another problem I need to get the vertices of the polyhedron defined by the linear constraints, as with the cddlib library, do you know how I could get that ? Enumerating vertices requires a very different algorithm from optimizing over polyhedra. The best way to do this

Re: [julia-users] Re: How to use GLPK.exact ?

When we have a simplex solver (either in Julia or external) that supports rational inputs, we could consider making this work with JuMP, but for now JuMP stores all data as floating-point as well. Stephane, nice work. LP definitely needs more exposure in the probability community. Please

Re: [julia-users] Re: Calculus2: A new proposed interface for Calculus

But once you have such a unified interface, why call the result Derivatives? What’s left for Calculus except the content of Derivatives? Will Calculus just be a symbolic calculus system? I don't have a strong opinion on the naming, but it seemed reasonable. Calculus also has simplification

[julia-users] Re: Calculus2: A new proposed interface for Calculus

I like the unified interface for differentiation. My only qualm is with flattening out the hierarchy of packages by including DualNumbers. There was some discussion on this here: https://github.com/johnmyleswhite/Calculus.jl/pull/44 To start out, what is the intended scope of Calculus.jl? Once

[julia-users] Re: example of thick Julia bindings

Hi Gour, It's pretty straightforward to wrap opaque C pointers with Julia types. For example, see https://github.com/JuliaOpt/Gurobi.jl. Adding a finalizer is a good way to avoid needing to manually free objects. It's also common to add multiple methods to perform type conversions and

Re: [julia-users] Re: Parallel sparse matrix vector multiplication

at 11:43 AM, Miles Lubin miles...@gmail.com wrote: Memory access is typically a significant bottleneck in sparse mat-vec, so unfortunately I'm skeptical that one could achieve good performance using Julia's current distributed memory approach on a multicore machine. This really calls

Re: [julia-users] Jacobian function?

We're still in the process of putting together a nice interface for this, but automatic differentiation is a good option that isn't available in most other languages. It will give you an *exact* numerical derivative, not subject to approximation error from finite differences. As an example of

[julia-users] Re: Parallel sparse matrix vector multiplication

Memory access is typically a significant bottleneck in sparse mat-vec, so unfortunately I'm skeptical that one could achieve good performance using Julia's current distributed memory approach on a multicore machine. This really calls for something like OpenMP. On Wednesday, February 5, 2014

Re: [julia-users] Re: Parallel sparse matrix vector multiplication

But speaking of writing parallel matrix vector products in native Julia, this might be a great use case for shared arrays (although right now I think only dense shared arrays exist). Amit, can you comment on this? Actually you could use a shared array for both the sparse matrix and output

Re: [julia-users] Generating Large LP Models Efficiently

Hi Andrew, Cool that you found JuMP! Please have a look at the docs: https://jump.readthedocs.org/en/latest/jump.html Some of the differences that you will encounter coming from AMPL are: - JuMP doesn't separate the model and the data, you just use the data directly in the model. - JuMP

Re: [julia-users] automatic differentiation for optimization

That would be great! On Wed, Dec 25, 2013 at 3:39 PM, Frederic Testard frederic.test...@gmail.com wrote: In case it helps : the autodiff code in the MCMC package is reverse mode AD (because it is well suited to gradient code generation), you may find some inspiration there. The current