Re: [julia-users] Re: Article on `@simd`

This are great news! :) One question: Why is "tuple vectorization" needed for fast 3D vector calculations? On Wednesday, September 24, 2014 12:05:47 AM UTC+2, Arch Robison wrote: > > Update on 64-bit support for vectorizing loops: The support just went into > the Github sources. See https://gi

[julia-users] Re: Current state of 3rd party APIs in Julia?

The EC2 API is pretty complete in Amit Murthy's AWS.jl, but it needs to be updated, and takes a long time to compile. I don't know of Google APIs. We do want to have stable and well-tested packages for some widely used AWS APIs to start with. -viral On Monday, September 22, 2014 5:24:21 AM UT

Re: [julia-users] Re: Article on `@simd`

ispc has some amazing capabilities. With the stuff that Arch is doing, we are already incrementally getting there. Perhaps someone can also convince the ispc team to spend some of their time on Julia. :-) -viral On Wednesday, September 24, 2014 12:22:49 AM UTC+5:30, Jeff Waller wrote: > > Could

Re: [julia-users] Re: Pretty Printing of Results

Le mardi 23 septembre 2014 à 17:45 -0700, Jason Knight a écrit : > Donald, > > > > I would enjoy seeing your code once you release it. I'm doing research > with Bayesian bioinformatics models using MCMC. > > > You can place all your MCMC samples in a DataFrame and then call > describe on it t

[julia-users] congratulations to Indian dost

for the wonderful achievement with Mangalyaan! With a budget less than a Hollywood movie, I bet they must have largely used (and supported?) open sources - Julia included?

[julia-users] @grisu_ccall not defined when using Gadfly

I've uploaded julia to the newest development version yesterday. Julia Version 0.4.0-dev+728 Commit f7172d3* (2014-09-22 12:08 UTC) Platform Info: System: Linux (x86_64-redhat-linux) CPU: Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz WORD_SIZE: 64 BLAS: libopenblas (USE64BITINT NO_AFFINITY NEH

[julia-users] Re: DataArray Concatenation

b = [@data([4]),a[1:end-1]] b = [@data([NA,4]),a[1:end-1]] work and return a DataArray. But b = [@data([NA]),a[1:end-1]] says "ERROR: Type of literal data cannot be inferred" and I don't get why. Maybe you know how to fix the latter ? On Wednesday, 24 September 2014 03:11:20 UTC+2, Li Zhang wr

Re: [julia-users] Re: Article on `@simd`

On Wednesday, September 24, 2014 12:08:38 AM Uwe Fechner wrote: > One question: Why is "tuple vectorization" needed for fast 3D vector > calculations? 3 doesn't fit naturally into the hardware width. So the way to vectorize 3d computations is with 3 tuples, one per coordinate. @Arch, I noticed a

[julia-users] ArrayViews and "end" keyword

Hello, I am using ArrayViews but it seems that statements with "end" keyword are not supported e.g. view(A, :, 2:end) throws an error and I have to use view(A, :, 2:size(A,2)). Is this the way it is meant to be ? No "end" keyword with ArrayViews ? Thanks, Jan

Re: [julia-users] @grisu_ccall not defined when using Gadfly

Please take a look at this link: https://github.com/dcjones/Gadfly.jl/issues/409#issuecomment-56643684 I was facing the same problem and everything was solved after doing this: Pkg.free("Gadfly") Pkg.free("Showoff") Pkg.checkout("Gadfly") Pkg.chekcout("Showoff") I hope it helps! Charles

[julia-users] cumsum and Multinomial distribution with a huge Sparse matrix in Julia

Dear all, I am working with huge matrices (65600 x 65600) that represent the euclidean distances between sites. So, mat[i,j] = mat[j,i]. I would like to use a multinomial distribution to define to which site an element from site i can move: the closer is the target site j, the higher is the proba

Re: [julia-users] ArrayViews and "end" keyword

Yes, I think this is a limitation of Julia. `end` is really just syntactic sugar: expressions like `A[1,end]` will be replaced by `A[1,size(A,2)]` before being further parsed. (at least that is my understanding.) If there is no `[]` indexing syntax then `end` cannot work`. Thus to construct the

Re: [julia-users] cumsum and Multinomial distribution with a huge Sparse matrix in Julia

Presumably the problem is that cumsum generates a dense matrix? I think what you'll need to do is create a new "sparse cumsum" type: type StairVector{T} <: AbstractArray{T,1} index::Vector{Int} value::Vector{T} end v.index will be a sorted vector of indexes at which the cumsum increments

Re: [julia-users] cumsum and Multinomial distribution with a huge Sparse matrix in Julia

Amazing!! Thank you, Tim! Actually my main problem is "time consuming", but of course it may be associated to a dense matrix, so I think your suggestions will help me a lot!!! I have a lot of things to study from your email, I will let you know how it evolves. Thanks a lot! Best, Charles On Wed

Re: [julia-users] ArrayViews and "end" keyword

OK, thanks for this info. I wonder whether e.g. view(A, :, 2:size(A,2))[end-10:end] will create a copy of the data or only a reference. In Julia v0.3 bracket expressions like A[:,2:end] will actually create a copy what imposes quite an overhead if A is large. Jan Dňa streda, 24. septembra 2014

Re: [julia-users] ArrayViews and "end" keyword

> OK, thanks for this info. I wonder whether e.g. view(A, :, > 2:size(A,2))[end-10:end] will create a copy of the data or only a > reference. That depends on the implementation of getindex for ArrayView which could do either. Easy to check though: typeof(view(A, :, 2:size(A,2))[end-10:end]) B

[julia-users] Re: Pretty Printing of Results

Jason, Thank you for this information. I will try to get this working soon. If you would like the code for the Bayesian spatial autoregressive model, please send me an email: donald.lacombe-at-mail.wvu.edu and I can get these and the support functions out to you. The code currently uses simula

Re: [julia-users] Re: Pretty Printing of Results

Dear Milan, Thank you for this information. I think that it will prove to be very useful for my needs. Regards, Don On Wednesday, September 24, 2014 3:29:00 AM UTC-4, Milan Bouchet-Valat wrote: > > Le mardi 23 septembre 2014 à 17:45 -0700, Jason Knight a écrit : > > Donald, > > > > I wou

Re: [julia-users] @grisu_ccall not defined when using Gadfly

However, julia> Pkg.free("Gadfly") ERROR: Gadfly cannot be freed – repo is dirty in free at pkg/entry.jl:204 in anonymous at pkg/dir.jl:28 in cd at ./file.jl:20 in __cd#230__ at ./pkg/dir.jl:28 in free at pkg.jl:37 julia> Pkg.free("Showoff") ERROR: Showoff is not a git repo in free at pkg/e

[julia-users] Re: DataArray Concatenation

It looks like a bug that @data([NA]) throws an error, should an issue be filed ?

Re: [julia-users] Re: DataArray Concatenation

What would @data([NA]) even mean? — John On Sep 24, 2014, at 7:12 AM, muraveill wrote: > It looks like a bug that @data([NA]) throws an error, should an issue be > filed ?

Re: [julia-users] Re: Article on `@simd`

I've been thinking about this a bit, and as usual, Julia's multiple dispatch might make such a thing possible in a novel way. The heart of ISPC is allowing a function that looks like int addScalar (int a, int b) { return a + b; } effectively be vector addVector (vector a, vector b) { return /*AV

Re: [julia-users] @grisu_ccall not defined when using Gadfly

Have you tried to remove your .julia directory and add the package by using: Pkg.checkout("Gadfly") ? Charles On Wed, Sep 24, 2014 at 3:59 PM, wrote: > However, > > julia> Pkg.free("Gadfly") > ERROR: Gadfly cannot be freed – repo is dirty > in free at pkg/entry.jl:204 > in anonymous at pkg/di

Re: [julia-users] DataArray Concatenation

Here’s a supported way to do this operation. — John julia> using DataArrays julia> a=@data([NA,3,5,7,NA,3,7]) 7-element DataArray{Int64,1}: NA 3 5 7 NA 3 7 julia> b = copy(a) 7-element DataArray{Int64,1}: NA 3 5 7 NA 3 7 julia> pop!(b) 7 julia> unshift!(

Re: [julia-users] Re: DataArray Concatenation

To my naive view, a data array with cells containing ony value NA. Well, it works with numbers, why not NA. The error thrown is the same in two dimensions with arrays of length > 1.

Re: [julia-users] Re: DataArray Concatenation

julia> @data([NA NA NA; NA NA NA]) ERROR: Type of literal data cannot be inferred julia> @data([NA NA NA; NA NA 1]) 2x3 DataArray{Int64,2}: NA NA NA NA NA 1 On Wednesday, 24 September 2014 16:22:48 UTC+2, muraveill wrote: > > To my naive view, a data array with cells containing ony value N

Re: [julia-users] Re: DataArray Concatenation

Naivete isn’t a big deal. Just try to be very precise. Any literal in Julia should produce a value V of type T. What’s the type T that @data([NA]) would produce? — John On Sep 24, 2014, at 7:22 AM, muraveill wrote: > To my naive view, a data array with cells containing ony value NA. Well, it

Re: [julia-users] Re: DataArray Concatenation

I was about to say DataArray{NAtype,1}. But then the type cannot be changed according to what is added to it, right ? Then DataArray{Any,1}. Just as @data(["asdf" NA; NA 1.4]). On Wednesday, 24 September 2014 16:25:17 UTC+2, John Myles White wrote: > > Naivete isn’t a big deal. Just try to be ver

Re: [julia-users] Re: DataArray Concatenation

I think that DataArray{Any, 1} is probably the best thing you could possibly do. But it’s still going to cause people lots of problems, because there’s almost never a time when you’d want to work with DataArray{Any, 1}. At some point, we have to improve the @data macro. But for this use case, I

[julia-users] Re: PathPatch object in julia with PyPlot

On Tuesday, September 23, 2014 4:17:22 PM UTC-4, Diego Tapias wrote: > > using PyCall > > @pyimport matplotlib.patches as patch > circle=patch.pymember(“Circle”) > > But when I try circle[:set_radius](1.0) or set!(circle, radius, 1.0) I > got an error. Any ideas? > > Circle is the class, not an

Re: [julia-users] Re: a good IDE for Julia ? (Julia Studio does not work with Julia v 0.3.0)

Yes, having a debugger is important and it's being actively worked on. On Tue, Sep 23, 2014 at 4:03 AM, Ján Dolinský wrote: > Hello, > > Thanks for another tip. I read a recent post of Stefan Karpinski > concerning debugger and IDE and I assume all of these options (LightTable, > SublimeText) do

Re: [julia-users] Re: DataArray Concatenation

@data([NaN]) already correctly returns DataArray(Float64, 1). If people mix NA's inside, it should return an Any type. The problem is when it comes from data that is all NA in some parts, like results of statistical tests with zero variance etc., and you want @data(subset_of_the_data). But stat

Re: [julia-users] Re: DataArray Concatenation

Please don’t conflate NaN and NA. You should _never_ use NaN unless you’re working with an Array, not a DataArray. — John On Sep 24, 2014, at 7:48 AM, muraveill wrote: > @data([NaN]) already correctly returns DataArray(Float64, 1). If people mix > NA's inside, it should return an Any type. >

Re: [julia-users] Re: Current state of 3rd party APIs in Julia?

Julia Ferraioli wrote a blog post on running Julia on the Google Compute Engine earlier in 2014: http://www.blog.juliaferraioli.com/2014/02/julia-on-google-compute-engine-parallel.html I'm not sure how up-to-date it is, but it may be helpful. On Wed, Sep 24, 2014 at 3:10 AM, Viral Shah wrote:

Re: [julia-users] congratulations to Indian dost

I think this is the flight to Mars that India just finsihed. — John On Sep 24, 2014, at 8:04 AM, Stefan Karpinski wrote: > I have no idea what this is about. Can you clarify? > > On Wed, Sep 24, 2014 at 3:47 AM, K Leo wrote: > for the wonderful achievement with Mangalyaan! > > With a budget

Re: [julia-users] congratulations to Indian dost

I have no idea what this is about. Can you clarify? On Wed, Sep 24, 2014 at 3:47 AM, K Leo wrote: > for the wonderful achievement with Mangalyaan! > > With a budget less than a Hollywood movie, I bet they must have largely > used (and supported?) open sources - Julia included? > > >

Re: [julia-users] congratulations to Indian dost

Is there any evidence that Julia was used to accomplish that? On Wed, Sep 24, 2014 at 11:05 AM, John Myles White wrote: > I think this is the flight to Mars that India just finsihed. > > — John > > On Sep 24, 2014, at 8:04 AM, Stefan Karpinski > wrote: > > I have no idea what this is about. Ca

Re: [julia-users] congratulations to Indian dost

Not that I’m aware of. I’d say the thread is probably off-topic. — John On Sep 24, 2014, at 8:10 AM, Stefan Karpinski wrote: > Is there any evidence that Julia was used to accomplish that? > > On Wed, Sep 24, 2014 at 11:05 AM, John Myles White > wrote: > I think this is the flight to Mars t

Re: [julia-users] congratulations to Indian dost

Seems wildly unlikely. The lead time and certification hoops for software that has to run successfully a quarter-billion kilometers from earth are, well, out of this world. "Sorry, sir, but our Mars probe crashed. It's a complete loss." "Oh no! Why?" "Well, just before it was about to dock, Ti

Re: [julia-users] congratulations to Indian dost

I should have known that Tim Holy is the real Ghost of Mars. — John On Sep 24, 2014, at 8:27 AM, Tim Holy wrote: > Seems wildly unlikely. The lead time and certification hoops for software > that > has to run successfully a quarter-billion kilometers from earth are, well, > out > of this w

Re: [julia-users] Re: Article on `@simd`

... though I suspect to really profit from masked vectorization like this, it needs to be tackled at a much lower level in the compiler, likely even as an LLVM optimization pass, guided only by some hints from Julia itself. *Sebastian Good* On Wed, Sep 24, 2014 at 10:16 AM, Sebastian Good < seba

Re: [julia-users] module definition within macro

Thanks for these pieces of wizardry. I had a need for also adding a body to the module being created. Got it to work based on the code below; sharing if someone else needs more examples than in the ones Patrick linked to: julia> function create_module(modname::Symbol) Expr(:toplevel, :(

Re: [julia-users] congratulations to Indian dost

Since we're off the ranch anyway, the brochure is a neat read: http://www.isro.gov.in/pslv-c25/pdf/pslv-c25-brochure.pdf (the MAR1750 processor is a 16-bit ISA from 1980, implemented in a radiation-hardened package and widely used for spacecraft control) On Wed, Sep 24, 2014 at 11:28 AM, John My

Re: [julia-users] Re: Article on `@simd`

You couldn't really preserve the semantics as Julia is a much more dynamic language. ISPC can do what it does because the kernel language is fairly restrictive. On Wednesday, September 24, 2014 11:30:56 AM UTC-4, Sebastian Good wrote: > > ... though I suspect to really profit from masked vector

Re: [julia-users] Re: Article on `@simd`

This is an important part. One of the most important pieces of functionality in vectorizing compilers is explaining how and why they did or didn't vectorize your code. It can be terrifically complicated to figure out. With ISPC, the language is constrained such that everything can be vectorized and

[julia-users] Re: PathPatch object in julia with PyPlot

(Please don't double-post questions.)

Re: [julia-users] Re: Pretty Printing of Results

Dear Milan, Do you have a link to the "coeftable" code? I looked here but could not find it: https://github.com/JuliaStats/StatsBase.jl Thanks, Don On Wednesday, September 24, 2014 3:29:00 AM UTC-4, Milan Bouchet-Valat wrote: > > Le mardi 23 septembre 2014 à 17:45 -0700, Jason Knight a écrit

Re: [julia-users] Re: Pretty Printing of Results

> > Do you have a link to the "coeftable" code? I looked here but could not > find it: > Try: https://github.com/JuliaStats/StatsBase.jl/blob/a2ba466f0b1ba82ed45f35918a9c5804ccc11d6b/src/statmodels.jl#L50

Re: [julia-users] Re: PathPatch object in julia with PyPlot

Thanks Steven... I will take your advice into account! 2014-09-24 13:33 GMT-05:00 Steven G. Johnson : > (Please don't double-post questions.) >

[julia-users] Re: congratulations to Indian dost

Government research organizations in India often use open source software just like everywhere else, where applicable. I am pretty sure that there was no Julia usage given that the mission would have been planned and its software was developed when Julia had barely been around. -viral On Wedne

[julia-users] Strange vector operation

I have been updated to version 0.4 though I seem to remember it was said the PPA archive will not move to 0.4 before a stable prerelease has been reached. Now I get the following strange behavior in vector operations: julia> x1 = [1.0, 1, 1]; x2 = [1:3]; julia> x1 / x2 3x3 Array{F

Re: [julia-users] Strange vector operation

No comment on the rest, but the element-wise division operator is `./`, which does work: ~~~ julia> x1 = [1.0, 1, 1]; julia> 1.0 ./ x1 3-element Array{Float64,1}: 1.0 1.0 1.0 ~~~ -- Leah On Wed, Sep 24, 2014 at 5:17 PM, Hans W Borchers wrote: > I have been updated to version 0.4 though I s

Re: [julia-users] Strange vector operation

Yes, I know. But the manual says: "Some operators without dots operate elementwise anyway when one argument is a scalar. These operators are *, /, \, and the bitwise operators." On Thursday, September 25, 2014 12:55:16 AM UTC+2, Leah Hanson wrote: > > No comment on the rest, but the element-wis

Re: [julia-users] Strange vector operation

The manual could possibly be a bit more clear on this, but I don't think much has changed since 0.3. The forward slash in your first example gets computed as x1/x2=(x2'\x1')' which Julia interprets as an underdetermined system with three right hand sides. The description in the manual about defau

Re: [julia-users] Re: Pretty Printing of Results

Jason, I copied lines 26-82 from the Git Hub page you kindly referenced and did the following: In [4]: mat = [1 2 3;4 5 6;7 8 9] Out [4]: 3x3 Array{Int64,2}: 1 2 3 4 5 6 7 8 9 In [5]: colnms =["a";"b";"c"] Out [5]: 3-element Array{ASCIIString,1}: "a" "b" "c" In [6]: rownms = [

Re: [julia-users] Re: Pretty Printing of Results

> > In [11]: show(CoefTable) > > Out [11]: ERROR: `show` has no method matching show(::Type{CoefTable}) > you may have intended to import Base.show > while loading In[11], in expression starting on line 1 > > > In [12]: import Base.show > > Warning: import of Base.show into Main conflicts with an

Re: [julia-users] Strange vector operation

Rather than complain, please help clarify the manual. Like Andreas said, I doubt much about those operations has changed between 0.3 and 0.4. But if you don't want to be surprised by changes in behavior, you might want to stick with 0.3; more disruptive changes are in the works. --Tim On Wedne

Re: [julia-users] Re: Pretty Printing of Results

Jason, I think you may have saved my sanity! Here is the output: In [22]: temp = CoefTable(mat,colnms,rownms) Out [22]: a b c a1 2 3 b4 5 6 c7 8 9 In [23]: show(temp) a b c a1 2 3 b4 5 6 c7 8 9 These actually line up perfectly in my Sublime Editor so I thin

Re: [julia-users] Re: Pretty Printing of Results

Hey Don, Have you considered putting your CoefTable function in a Julia package somewhere? I'm not sure whether there's an existing one that would be appropriate, but this looks like a useful bit of code that could go in a library. -- Leah On Wed, Sep 24, 2014 at 7:49 PM, Donald Lacombe wrote:

Re: [julia-users] Re: Pretty Printing of Results

Dear Leah, I actually used the following code that Jason told me about: https://github.com/JuliaStats/StatsBase.jl/blob/a2ba466f0b1ba82ed45f35918a9c5804ccc11d6b/src/statmodels.jl#L50

[julia-users] Re: DataArray Concatenation

i tried methods in base for Dequeues, but it seems some of them are not implemented for DataArray. For example: a=@data([NA,3,5,7,NA,3,7]) b=copy(a[2:end]) append!(b,[NA]) append! works, but prepend! gives no methods error. also deleteat! works but insert! no methods. anyway, other methods wo

Re: [julia-users] Re: DataArray Concatenation

Would be great to submit a pull request implementing the missing funtionality. -- John On Sep 24, 2014, at 6:06 PM, Li Zhang wrote: > i tried methods in base for Dequeues, but it seems some of them are not > implemented for DataArray. > For example: > > a=@data([NA,3,5,7,NA,3,7]) > b=copy(a[

Re: [julia-users] Re: Pretty Printing of Results

I'd highly recommend learning how to make a package and putting it up on Github. And no better way to learn than to do. Then you're one step closer to contributing back to the Julia ecosystem! I don't see the CoefTable code having *too* much additional utility over the describe function in DataFra

Re: [julia-users] Re: Pretty Printing of Results

Jason, I appreciate the kind words and advice but the code I'm using is from StatsBase and there is another version in Mixed Models that looks nice as well. It's all already out there and I just wanted to modify it for my own use. The only thing I cannot figure out with a DataFrame is how to a

Re: [julia-users] Re: Pretty Printing of Results

There is no Row column. That's just a printing convention. -- John On Sep 24, 2014, at 7:20 PM, Donald Lacombe wrote: > Jason, > > I appreciate the kind words and advice but the code I'm using is from > StatsBase and there is another version in Mixed Models that looks nice as > well. It's a

[julia-users] Package installation directory: dealing with multiple Julia version

The issue is concerned with Julia's packages install directory. We successfully installed Julia on the department Linux server ("central Julia") and I have a personal installation on a local area ("my Julia"). At the moment my install directory is ~/.julia/version/ for both mine and central Juli

[julia-users] Nested parallelism possible?

Hi all: I have a question regarding nested @parallel macros. Let's say I have a cluster and I use ssh or Condor to log into the nodes. Each node has multiple cores, of course, so on each node I would like to run a few simulations in parallel or run a parallel solver. I can add workers by usi

Re: [julia-users] Strange vector operation

I wanted to stick with 0.3. I was under the impression (from thread "Re: Announcing Julia 0.3.0 final.") that the PPA would change to 0.4 only when a kind of prerelease is nearing. In a phase of disruptive changes, it's difficult to tell whether the manual is not quite accurate or the behavio