Re: [julia-users] Best practice to type-check for array of floats?

FWIW, the read/write distinction really helps me to remember the terms: For reading, you want to be sure the array you’ve got contains values at least as specific as you were expecting. So, if you’re expecting Integer, it’s fine if you get Int. In that case, you might hope that Vector{Int} <: V

Re: [julia-users] Best practice to type-check for array of floats?

On Fri, Aug 15, 2014 at 9:53 PM, John Myles White wrote: > Hi Brendan, > > It looks like you’re hitting Julia’s invariance for the first time: > http://en.wikipedia.org/wiki/Covariance_and_contravariance_(computer_science) Oh nice. I could never keep the covariance/contravariance directions st

Re: [julia-users] Best practice to type-check for array of floats?

Hi Brendan, It looks like you’re hitting Julia’s invariance for the first time: http://en.wikipedia.org/wiki/Covariance_and_contravariance_(computer_science) If you’re working with inputs to a function, you can do the typecheck using a type parameter, T: function foo{T <: FloatingPoint}(x::Vec

[julia-users] Best practice to type-check for array of floats?

Hi, What's the best way to typecheck that an array is of floats, but any precision is OK? I've found it useful to do this because a number of operations fail when given integer vector inputs (for example, 10.^[-3:3]). I noticed that x::Vector{FloatingPoint} does not seem to cover them. I e

Re: [julia-users] Semi-OT: Recommendations for learning just enough C for Julia

João, I'd love to see your example. Even with light documentation, I think I could follow along. But from yours and everyone else's comments, it sounds like there really isn't too easy a way out here, that I should just break down and spend a few days/weeks playing around. Thanks for your detai

[julia-users] Package built errors

What to do with the following? I am on Xubuntu 14.04 64bits. == _ _ _ _(_)_ | A fresh approach to technical computing (_) | (_) (_)| Documentation: http://docs.julialang.org _ _ _| |_ __ _ | Type "help()" for help.

Re: [julia-users] Re: updating julia

Hah! Tony caught me! Yes, I made a small boo-boo yesterday and had to switch the order of two commits. ;) In addition to the in-depth answer above, I'd just like to add I typically `make clean` whenever I pull a new revision down. 80% of the time it's not necessary, but especially if you haven

Re: [julia-users] Re: updating julia

Thanks very much. It would be nice if this info were somewhere, e.g, README.md. On Fri, Aug 15, 2014 at 02:53:22PM -0700, Tony Kelman wrote: > You're fine, you're just seeing some consequences of the confusing way git > submodules work. For some dependencies that are managed within JuliaLang on

[julia-users] Re: updating julia

You're fine, you're just seeing some consequences of the confusing way git submodules work. For some dependencies that are managed within JuliaLang on github, they are set up as submodules in the deps folder so it's easier to work on them while still in version control inside a Julia source tree

[julia-users] updating julia

If I pull new source from julia can I just repeat the build procedure (without even, e.g. make clean)? Does it matter if it's within 0.3 or 0.3 to 0.4? (I plan to take the advice to stay off 0.4, but am wondering about the general principles.) I take it from previous discussion that I don't need

[julia-users] Command Line Julia Helper

I just ran across the very cool https://github.com/Russell91/pythonpy If someone has a free afternoon, then a Julia port would be cool (hint, hint, ;). As I've found myself in the past doing things like: cat numbers.csv | julia -E 'sum(map(int,readlines(STDIN)))' And a few more flags for conve

Re: [julia-users] Semi-OT: Recommendations for learning just enough C for Julia

I googled for "just enough C" (good book name, alas, taken) and found this, which looks very useful but does not cover shared libraries: http://www.slideshare.net/petdance/just-enough-c-for-open-source-programmers > Regarding the shared library part, you could probably live only knowing it > is

[julia-users] Re: ANN: Docile.jl Package Documentation (again)

Neither have I :) It's probably not the best solution for some projects, but I've seen mentions of docstrings here on a number of occasions so I thought I'd try them out.

Re: [julia-users] Literal numeric coefficient on RHS of variable

No worries. I had no idea it was a dup either. On Fri, Aug 15, 2014 at 4:17 PM, Peter Simon wrote: > Whoops, I see the issue was closed due to it being a duplicate of 6770 > . Sorry, I looked but > didn't find this before posting. > > --Pete > >

Re: [julia-users] Literal numeric coefficient on RHS of variable

Whoops, I see the issue was closed due to it being a duplicate of 6770 . Sorry, I looked but didn't find this before posting. --Pete On Friday, August 15, 2014 1:12:30 PM UTC-7, Stefan Karpinski wrote: > > Thanks! > > > On Fri, Aug 15, 2014 at 4:

[julia-users] Re: Semi-OT: Recommendations for learning just enough C for Julia

On Friday, August 15, 2014 10:15:55 AM UTC-7, Randy Zwitch wrote: > > I know the standard recommendation of K&R for people who want to learn C. > But what would people's recommendations be to learn *just enough* C to be > comfortable using C libraries from within Julia? > The problem with K&R is

Re: [julia-users] Literal numeric coefficient on RHS of variable

Thanks! On Fri, Aug 15, 2014 at 4:03 PM, Peter Simon wrote: > Issue 8017 has been > opened. > > --Peter > > > On Friday, August 15, 2014 12:56:50 PM UTC-7, Stefan Karpinski wrote: > >> Yeah, that seems pretty broken. Would be mind filing an issue

Re: [julia-users] Different behavior: "using" vs "include()"

Hi Tim, > I can't test it myself because I get > julia> using ImageQuilting > ERROR: syntax: invalid character "ₛ" > in include at ./boot.jl:245 > in include_from_node1 at ./loading.jl:128 > in include at ./boot.jl:245 > in include_from_node1 at ./loading.jl:128 > in reload_path at l

Re: [julia-users] .xlsx (excel)

I believe the Taro package can do this, though it depends on having Java available. On Fri, Aug 15, 2014 at 4:04 PM, Frederico Novaes wrote: > Hi, > > Any package/function to read .xlsx files ? > > Frederico.

[julia-users] .xlsx (excel)

Hi, Any package/function to read .xlsx files ? Frederico.

Re: [julia-users] Literal numeric coefficient on RHS of variable

Issue 8017 has been opened. --Peter On Friday, August 15, 2014 12:56:50 PM UTC-7, Stefan Karpinski wrote: > > Yeah, that seems pretty broken. Would be mind filing an issue? > > > On Fri, Aug 15, 2014 at 12:24 PM, Tomas Lycken > wrote: > >> I can't

Re: [julia-users] Literal numeric coefficient on RHS of variable

Yeah, that seems pretty broken. Would be mind filing an issue? On Fri, Aug 15, 2014 at 12:24 PM, Tomas Lycken wrote: > I can't say why, but it seems x.2 is parsed as .2x, which intentionally > evaluates as .2*x... > > // T >

[julia-users] Re: functions with state?

Forgive me if I'm oversimplifying, but will closures of the following form work for you? mod2 = x -> mod(x,2) mod2(3) Or you can even create closure generators: modcreator(m) = x->mod(x,2) mod2 = modcreator(2) mod3 = modcreator(3) ... If you want mutable state, you might prefer something like

[julia-users] Module system proposals

* https://github.com/JuliaLang/julia/issues/8014 * https://github.com/JuliaLang/julia/issues/8015

Re: [julia-users] moving from v0.3 to v0.4

Once we announce 0.3.0-final I will send out an email formalizing this a little bit. -E

Re: [julia-users] Re: functions with state?

I mean something like this: function f(x, state) # do something with x and state end type FunctionWithState state end call(fs::FunctionWithState, x) = f(x, fs.state) Now write call(fs, x) wherever you would have written f(x). On Fri, Aug 15, 2014 at 2:37 PM, Neal Becker wrote: > By clo

[julia-users] Re: functions with state?

By closure I believe you mean a function that returns a function which binds the 2nd argument of the function F. For example: function bound_phase_noise (spec::phase_noise_spec) function B (f) return compute_phase_noise_spec (spec, f) end end But I don't know what you mean by t

Re: [julia-users] functions with state?

The approaches that occur to me immediately are that you can either explicitly pass state or use a closure that closes over the state you need to update. Another option would be to represent your function as an object and use a generic function to apply that object to its argument. On Fri, Aug 15

Re: [julia-users] moving from v0.3 to v0.4

I agree. We've decided (I can't seem to find a link to it right now) that 0.4-dev is the "chaos period" when we'll be breaking things frequently. During this period, packages should continue to target 0.3. Once all the major breakages have occurred and we're ready to start stabilizing the master br

[julia-users] Re: my 2nd julia module (linear intepolator)

One warning I would give is about leaving a space between the macro name and the parenthesis with the arguments. This generally does not work in Julia because Julia has two different macro invocation forms and it mixes between them. For example a macro that returns a expression adding its two

RE: [julia-users] moving from v0.3 to v0.4

And I really hope that most package developers stay away from 0.4 for a long time, so that we have healthy package improvements working on 0.3 until 0.4 is actually released :) From: julia-users@googlegroups.com [mailto:julia-users@googlegroups.com] On Behalf Of Iain Dunning Sent: Friday, Au

[julia-users] functions with state?

I'm trying to do numerical integration. I want a function that has state information. Let's say I'm trying to integrate some function F over x. In addition, F has some state function F (x, state) = In python (and in c++), one way is to make F a class (which can have state), and overload th

[julia-users] Re: my 2nd julia module (linear intepolator)

Seems like you have a habit of adding a space between the function name and the opening parenthesis. Julia is sometimes sensitive to extra space, so if you do this in the wrong context, you might get subtle bugs. Especially macros have both a statement (parentheses less form) and a function for

Re: [julia-users] moving from v0.3 to v0.4

Absolutely. Already had ~10 packages break today due to 0.4 stuff, I expect plenty more. On Friday, August 15, 2014 11:24:18 AM UTC-4, John Myles White wrote: > > I agree: I think most people should not use 0.4 beause it’s going to > change so frequently. > > — John > > On Aug 15, 2014, at 12:

Re: [julia-users] Semi-OT: Recommendations for learning just enough C for Julia

I once wrote a simple example of how to write and compile C code as a shared library and call from Julia using ccall. I could share it if you are interested, although it has no documentation at all at the moment besides a couple of comments. Main things you have to know from C are types, structs,

Re: [julia-users] Different behavior: "using" vs "include()"

I can't test it myself because I get julia> using ImageQuilting ERROR: syntax: invalid character "ₛ" in include at ./boot.jl:245 in include_from_node1 at ./loading.jl:128 in include at ./boot.jl:245 in include_from_node1 at ./loading.jl:128 in reload_path at loading.jl:152 in _require at load

Re: [julia-users] Equivalent of c++ functor objects, std::bind, and lambdas

Thus is not an issue with pass-by-sharing so much as the existence of mutable types. You can avoid needing to define (or use) the copy method by structuring your logic to generally avoid mutating your data structures. I know this may sound strange at first, but if you are going to be making copies

[julia-users] my 2nd julia module (linear intepolator)

As another learning exercise, I converted my python 1-d linear interpolator module to julia. Any suggestions/comments on improvements? https://gist.github.com/anonymous/19526d1455ba7dc9f48f

[julia-users] Semi-OT: Recommendations for learning just enough C for Julia

I know the standard recommendation of K&R for people who want to learn C. But what would people's recommendations be to learn *just enough* C to be comfortable using C libraries from within Julia? For example, the manual states: "The code to be called must be available as a shared library. Most

Re: [julia-users] Vectorised boolean access to array elements

Because that might be subtle to see if you're not looking closely, and you mentioned a MATLAB background, note that there is an additional dot before the comparison operator--Julia uses the elementwise operator notation for comparison operators as well as for elementwise multiplication and divi

Re: [julia-users] Equivalent of c++ functor objects, std::bind, and lambdas

Ok thanks. I guess the heart of the question is overcoming Julia's builtin pass-by-reference behavior. I would be fine using an explicit copy function, but is there any way I can avoid defining a copy function for all my types, which would be annoying? OK, it seems that deepcopy corresponds bette

Re: [julia-users] We have typed functions, don't we?

Love it! I get your point much better now. I'll be using it in the future :). Re Iterators, yes, your guess is spot on---it's those pesky tuples (which are way better than they used to be, see https://github.com/JuliaLang/julia/pull/4042). In high performance code, iterators from Iterators are

Re: [julia-users] Equivalent of c++ functor objects, std::bind, and lambdas

Ok thanks. I guess the heart of the question is overcoming Julia's builtin pass-by-reference behavior. I would be fine using an explicit copy function, but is there any way I can avoid defining a copy function for all my types, which would be annoying?

Re: [julia-users] Equivalent of c++ functor objects, std::bind, and lambdas

If the function "copy" is implemented for z: z = ... newfun = let zz = copy(z); (x, y) -> f(zz, x, y) end I think I understood that lambdas are less efficient than functions so this may be faster: let zz = copy(z) global newfun newfun(x, y) = f(zz, x, y) end

[julia-users] Re: Equivalent of c++ functor objects, std::bind, and lambdas

function make_counter(x) return (delta) -> begin x[1] += delta x end end x = [0] c0 = make_counter(x) c1 = make_counter(x) c0(1) # [1] c1(1) # [2] c0(1) # [3] c1(1) # [4] That seems to be the case with scalars, but not with a vector (or other objects I assume). See the above code

[julia-users] Different behavior: "using" vs "include()"

Dear all, Consider this package: https://github.com/juliohm/ImageQuilting.jl If I do: using ImageQuilting ImageQuilting.example() then the images are not shown by ImageView, even though this package is installed on my system. If however I do: include("src/example.jl") example() then

[julia-users] Literal numeric coefficient on RHS of variable

I can't say why, but it seems x.2 is parsed as .2x, which intentionally evaluates as .2*x... // T

[julia-users] Re: Equivalent of c++ functor objects, std::bind, and lambdas

Here's a quick test: function make_counter(x) return (delta) -> begin x += delta x end end c0 = make_counter(0) c1 = make_counter(0) c0(1) # 1 c1(1) # 1 c0(1) # 2 c1(1) # 2 So it looks like they don't share the bound 'x' variable. On Friday, August 15, 2014 11:50:03 AM UTC-4, Noah

Re: [julia-users] We have typed functions, don't we?

> I've not used NumericFuctors so I can't comment on your main question. If > it's of any use, there's an entirely different approach (more of a dirty > trick, really) to inlining functions passed in as arguments. Here's a gist > that shows the trick: > https://gist.github.com/timholy/bdcee95f9b772

[julia-users] Literal numeric coefficient on RHS of variable

Is the following behavior expected? julia> x = 2 2 julia> x.2 0.4 I'm running Version 0.3.0-rc4 (2014-08-15 04:01 UTC) Thanks, --Peter

[julia-users] Equivalent of c++ functor objects, std::bind, and lambdas

suppose I have a function with prototype "f(x, y, z)". How can bind one of the arguments to a new function argument by copying vs reference? In c++11 I could do z = ... ; auto newfun = [=](x,y) { f(x,y, z)}; if I wanted newfun to have its own copy of z. Then I could make some changes to ori

Re: [julia-users] moving from v0.3 to v0.4

I agree: I think most people should not use 0.4 beause it’s going to change so frequently. — John On Aug 15, 2014, at 12:28 AM, Tobias Knopp wrote: > I think we should discourage using 0.4-dev version until feature freeze. > There might be a lot of package breakage (wasn't there once a term

[julia-users] Grammatical Evolution

For anyone who is interested, I've been working on a Grammatical Evolution (GE) [Conor Ryan, JJ Collins and Michael O'Neill, 1998] library: https://github.com/abeschneider/GrammaticalEvolution It is not quite ready to reach versio

[julia-users] Re: Running example code in packages

> > I do this via Pkg.dir()*"/MyPackage". > If there is any better solution, I'd be quite curious to know! > Thank you Simon, this solved the wrong path problem. > This looks like a cool package! =) > I just implemented an OpenGL image viewer, which theoretically can tile > images very easily

Re: [julia-users] We have typed functions, don't we?

I guess I replied to the subject of your post rather than the detail :). Sorry for the noise. On Fri, Aug 15, 2014 at 4:37 PM, Rafael Fourquet wrote: > > Hi Rafael, I recently posted an example of using function types, see: > > Thanks Adrian, but I couldn't see how it helps to make a > function

Re: [julia-users] We have typed functions, don't we?

> Hi Rafael, I recently posted an example of using function types, see: Thanks Adrian, but I couldn't see how it helps to make a function specialize on its (higher-order) function parameter (and possibly inline it).

Re: [julia-users] Re: Proposal for inargs

So, what's stopping you from writing that package? Since much of the code in Base (and in many packages) takes an AbstractArray input, you can protect any array you care about by wrapping it in ReadOnlyArray, and (mostly) it should Just Work. You don't need to convince anyone of anything. Besid

Re: [julia-users] How to save sparse in HDF5? ERROR: no method write(HDF5File, ASCIIString, SparseMatrixCSC{Int64,Int64})

Use JLD, not HDF5. HDF5 is only for very basic types, or reading HDF5 files written from other languages. --Tim On Thursday, August 14, 2014 11:21:20 PM paul analyst wrote: > How to save sparse in HDF5? > julia> D=sparse(I,J,V); > julia> h5write("D.h5","D", D); > ERROR: no method write(HDF5File,

Re: [julia-users] Vectorised boolean access to array elements

a .< 5 On Aug 15, 2014, at 3:53 AM, David Higgins wrote: > Hi, > > Is there a mechanism for vectorised logical access to array elements in > Julia? I basically mean, is there an equivalent to the Matlab notation > a[a<5] > which should give all of the elements of a[] which have value less tha

[julia-users] Vectorised boolean access to array elements

Hi, Is there a mechanism for vectorised logical access to array elements in Julia? I basically mean, is there an equivalent to the Matlab notation a[a<5] which should give all of the elements of a[] which have value less than 5. I guess the feature I've not found is the ability to create the boo

[julia-users] Re: Proposal for inargs

Tamas Papp asks what problem would be solved by being able to specify which function arguments are inargs. Here are three possible problems solved: (1) To catch programmer blunders, e.g., which is the correct order of the three matrices passed to 'matmul'. (2) To make debugging easier. For e

[julia-users] Re: Possibility for an MPI-based cluster manager for use on Cray systems?

Connections to Julia workers are regular socket streams. ClusterManagers.jl has modules to launch Julia workers on supported grid engines. We still need regular socket connectivity to the launched workers, following which pmap / @parallel can be used. ssh to launch workers is supported as part o

Re: [julia-users] 0.3.0 Release Candidate 4 released

Windows binaries are up. On Fri, Aug 15, 2014 at 12:44 AM, Elliot Saba wrote: > Mac and Ubuntu binaries are up, Windows not yet. Barring any significant > problems cropping up over the next few days, this RC is expected to be > promoted to 0.3.0-final. > > Happy coding, > -E >

Re: [julia-users] We have typed functions, don't we?

Rafael, thanks for the very kind note. I've not used NumericFuctors so I can't comment on your main question. If it's of any use, there's an entirely different approach (more of a dirty trick, really) to inlining functions passed in as arguments. Here's a gist that shows the trick: https://gist

Re: [julia-users] Proposal for inargs

Even C++ doesn't do nested constness. Hence you have double const * double * const double const * const double * const * const ... all meaning different things. --Tim On Friday, August 15, 2014 10:26:48 AM Tamas Papp wrote: > Suppose you have an array of arrays (or any other kind o

[julia-users] Re: Gadfly, draw, and font embedding.

Hello colleague, can you please post an example that creates rasterized fonts? I had a theory something could happen in Cairo.jl, but all my Gadfly examples (rendered to .svg, .pdf and .eps) actually contain an embedded font (.svg als Glyphs). So my next guess would be, it's an issue with the

Re: [julia-users] Re: 0.3.0 Release Candidate 4 released

I kept a separate directory for a checkout of the release-0.2 branch in order to be able to test things with old Julia. If you switch branches regularly, you will spend much time recompiling dependencies when they change and you are likely to hit trouble if we fail to future proof the make rule

[julia-users] Re: Incorrect type conversion for ImmutableArrays

Filled the issue on the Julia side: https://github.com/JuliaLang/julia/issues/8010 Am Freitag, 15. August 2014 12:13:42 UTC+2 schrieb Tobias Knopp: > > This is a bug but I am not sure on which side. The convert function is > wrong because it does not create the type that is passed. But maybe Jul

[julia-users] Re: Incorrect type conversion for ImmutableArrays

This is a bug but I am not sure on which side. The convert function is wrong because it does not create the type that is passed. But maybe Julia should not create the immutable type when seeing this. Could you please file a bug on the ImmutableArrays.jl project page? I will file one on the Juli

Re: [julia-users] We have typed functions, don't we?

Hi Rafael, I recently posted an example of using function types, see: https://groups.google.com/forum/#!searchin/julia-users/abstract$20functions/julia-users/lGRfDiJDX8E/jmahvLDldAUJ On Fri, Aug 15, 2014 at 10:23 AM, Rafael Fourquet wrote: > Oh, and by the way functions specialized on values

[julia-users] Re: How quickly subtract the two large arrays ?

This might be a bit faster: function sub!(A,B,C) for j=1:size(A,2) for i=1:size(A,1) @inbounds C[i,j] = A[i,j] - B[i,j] end end end C = zeros(size(A)); sub!(A,B,C) Do you have enough RAM to store these matrices though ? 10^5 * 10^5 Float64 seems rather large.

Re: [julia-users] Re: 0.3.0 Release Candidate 4 released

Yes, but once the release has been made one could also use the tagged version or use the binaries. Am Freitag, 15. August 2014 11:22:27 UTC+2 schrieb Carlos Becker: > > To keep up-to-date with the latest changes for v0.3, would it suffice to > checkout branch release-0.3 periodically? > Is that

[julia-users] How quickly subtract the two large arrays ?

How quickly subtract the two large arrays ? A = random (10 ^ 5.10 ^ 5); B = rand (10 ^ 5.10 ^ 5); @time C = A-B ; very long: / Paul

Re: [julia-users] Re: 0.3.0 Release Candidate 4 released

To keep up-to-date with the latest changes for v0.3, would it suffice to checkout branch release-0.3 periodically? Is that the one supposed to have the latest development code for v0.3? Thanks. El viernes, 15 de agosto de 2014 07:25:28 UTC+2, Elliot Saba escribió: > > Your packages should remain

[julia-users] Incorrect type conversion for ImmutableArrays

Dear all, I am seeing an unexpected behaviour when combining ImmutableArrays and struct like types (I forgot what the proper name is). In particular, the following code: using ImmutableArrays immutable Mytype pos::Vector2{Float32} end println(Mytype([4.; 5.])) incorrectly prints "Mytype

Re: [julia-users] Proposal for inargs

Suppose you have an array of arrays (or any other kind of containers). If I understand correctly, even if I wrap the outer array as read-only, your proposal would still allow changing the elements in the inner arrays. Given this, I wonder if your proposal would be a solution to your problem (I have

Re: [julia-users] We have typed functions, don't we?

Oh, and by the way functions specialized on values can be emulated, e.g. type plusN{N} plusN(x) = x+N end plus{10}(1) And writing a constrained function can be slightly simpler than in my previous post: # constrained function: f{F<:BinaryFunctor}(::Type{F}, x) = F(x, x) f(plus, 1) Or, as a

[julia-users] Re: moving from v0.3 to v0.4

I think we should discourage using 0.4-dev version until feature freeze. There might be a lot of package breakage (wasn't there once a term "chaos month"?). Am Freitag, 15. August 2014 09:16:20 UTC+2 schrieb Andreas Lobinger: > > Hello colleague, > > i was aware that there are parallel installa

[julia-users] Re: moving from v0.3 to v0.4

Hello colleague, i was aware that there are parallel installations With the 'smallest problem' i meant i immediately recognized a bunch of other problems, alike WARNING: split(x,y,k::Bool) is deprecated, use split($(Expr(:parameters, :(keep=k))),x,y) instead. in _atox at /home/lobi/.julia/v0.