If someone else needs something like this I ended up copying and modifying
pmap to suit my needs. I'm sure it can be simplified and shortened but it
works.
Example code below and I get healthy speedups all the way up to 20 procs:
feldt:~/tmp$ julia03 test_pfind_colwise.jl
0.5952118622068966
BTW, can anyone explain why it segfaults if one continues increasing the
number of procs available:
feldt:~/tmp$ julia03 -p 32 test_pfind_colwise.jl
0.05507445327586207
feldt:~/tmp$ julia03 -p 64 test_pfind_colwise.jl
0.0358273456
/Users/feldt/feldt/bin/julia03: line 2: 68828
@Roger Herikstad
Do you get a warning, that you should install GLFW 3.0.4 instead of 2.x ?
That would explain that error.
@J Luis
Are there any build errors?
And what windows version are you on?
I'm a little clueless here, as I don't really know, what there is that can
go wrong with the build
On 7 Jul 2014, at 16:00, Simon Danisch wrote:
@Roger Herikstad
Do you get a warning, that you should install GLFW 3.0.4 instead of
2.x ?
That would explain that error.
Got it to work after installing GLFW 3.0.4 via homebrew and checking out
the latest master of GLFW.jl
I now see a white
Great! =)
To add to Milan’s answer, it’s worth noting that you need type parameters
on your methods if you want to define them for MyType{AbstractT}, where
AbstractT is any abstract type:
julia f{T:FloatingPoint}(x::MyType{T}) = c
f (generic funciton with 3 methods)
julia f(b)
b
julia c =
After running `make cleanall`, I'm stuck on this error (apparently) while
compiling fftw3f. Does anybody else see this, or have suggestions for how
to solve it. I have tried to checkout old (previously working) commits, but
it did not help. `release-0.2` apparently compiles fine.
Making
I'm having build problems as well, but on Ubuntu. Don't know if it's
related - I get a completely different message - but I also tried checking
out old, previously working, commits with no success. I can `make clean` or
`make cleanall` to get all the way through building `Base`, but then
```
We've discussed various ways of making this simpler, but we weren't able to
settle on a good interface.
Currently, I think most readable way to access the coordinates of the
vertices is this:
```
cs = contours(x,y,z,N)
for c in cs
lvl, lines = c.level, c.lines
for line in lines
For reference: https://github.com/tlycken/Contour.jl/issues/9
On Monday, July 7, 2014 11:14:38 AM UTC+2, Tomas Lycken wrote:
We've discussed various ways of making this simpler, but we weren't able
to settle on a good interface.
Currently, I think most readable way to access the coordinates
On Sunday, July 06, 2014 02:52:45 PM Andrei Zh wrote:
Is there any way to update definitions that where imported into Main with
using statement?
Not that I know of. https://github.com/JuliaLang/julia/pull/7487 may be of
interest to you.
--Tim
For this example, yes. Poor type choices make for slow code.
I did the beginner tutorials on the Julia Studio website before I wrote this.
I may do a far more complex GA which involves variable length chromosomes.
If I am lucky enough, I will still get 10x the performance of Python (which
Just wondering, can metaprogramming be used to shorten code such as this
one:
*for i = 1:2axes[1][:scatter](kx, real([OnePump.λ1(0., momx, np) for
momx in kx]), s=15, alpha=0.4, color=orange)axes[1][:scatter](kx,
real([OnePump.λ2(0., momx, np) for momx in kx]), s=15, alpha=0.2)
Here's another version. Yes, there's a lot of cruft in there, and I'm sure
one could trim it down to a smaller example of the same issue, but (1) it
seems that removing things tend to remove the problem (… and I don't really
understand what the problem is), and (2) it sort of illustrates how
Hi Magnus,
Definitely looks like a bug. Would you be able to submit a bug report at
https://github.com/JuliaLang/julia/issues/new ? Your first example is
probably the easiest, since it's easier to follow.
Cheers,
Kevin
On Mon, Jul 7, 2014 at 4:09 AM, Magnus Lie Hetland m...@idi.ntnu.no
Sure thing. Just wanted to make sure I wasn't missing some PEBKAC here ;-)
I'll post a bug report, with a reference to this discussion.
I have a function which is suppose to take a variety of inputs, one of them
is an array of file names. The problem I'm running into is that sending in
an ASCII string array, Array{ASCIIString,1}, to a function with a more
general definition, Array{String,1}, it give me a method error.
julia y
This is because of invariance:
This last point is very important:
Even though Float64 : Real we DO NOT have Point{Float64} : Point{Real}.
In other words, in the parlance of type theory, Julia’s type parameters
are invariant, rather than being covariant (or even contravariant).
This is a FAQ and covered in the manual, in the section on parametric types.
If you don't find that description clear, please edit it to make it better.
Short answer: use a function signature like this
function myfunc{S:String}(a::Array{S,1})
Best,
--Tim
On Monday, July 07, 2014 04:50:14
To get dispatch working for arrays of any subtype of String, you have to
write your function using a type parameter:
f{T:String}(a::Array{T,1}) = a[1]
should work for both Array{ASCIIString,1} and Array{String,1}
On Monday, July 7, 2014 12:54:19 PM UTC+1, Mauro wrote:
This is because of
Also worked for me on Mac 10.9.3 with
Version 0.3.0-prerelease+3551 (2014-06-07 20:57 UTC)
Commit 547facf* (29 days old master)
x86_64-apple-darwin12.5.0
I also got the messages
INFO: loaded GLFW 3.0.4 Cocoa NSGL chdir menubar dynamic from
/Users/abahm/.julia/GLFW/deps/usr/lib/libglfw
I'm not sure I quite understand your example (your `i` variable is never
used?), but macros just take zero or more expression/symbol arguments and
return expressions, so you could easily do.
macro orangeaxes()
:(*axes[1][:scatter](kx, real([OnePump.λ1(0., momx, np) for momx in
kx]), s=15,
Cool! As someone who uses Processing to prototypes pretty much everything,
I'll have a look.
One important feature is that it makes super easy to set up an interactive
loop - your description makes it sound like it doesn't do that (yet).
Of course, the biggest draw to Processing is that it has
Also worked on my windows laptop, but I should note that the fading color
of the background happened at much higher frequency, like x50, as a
flicker, rather than a nice fade in/out.
Version 0.3.0-prerelease+3718 (2014-06-17 13:52 UTC)
Commit c5107b3 (19 days old master)
x86_64-w64-mingw32
The OS was Win7, SP1.
I'm excited for the upcoming release. Kudos on the work you've done so
far!
Cheers.
Alan
On Monday, July 7, 2014 6:58:53 AM UTC-7, Alan Bahm wrote:
Also worked on my windows laptop, but I should note that the fading color
of the background happened at much higher
I'm stumped by this upon loading a module I'm working on:
Warning: New definition
+(Union(Number,NAtype),AbstractDataArray{T,N}) at deprecated.jl:26
is ambiguous with:
+(Number,AbstractArray{T,N}) at array.jl:771.
To fix, define
+(Number,AbstractDataArray{T,N})
before the new
Maybe this issue is related:
https://github.com/JuliaLang/julia/issues/6190
On Mon, 2014-07-07 at 15:06, tony.hf.f...@gmail.com wrote:
I'm stumped by this upon loading a module I'm working on:
Warning: New definition
+(Union(Number,NAtype),AbstractDataArray{T,N}) at deprecated.jl:26
is
This seems to be related to DataFrames, (or rather DataArrays). Often such
issues will arrive at an update, and be fixed when you update again. If you
make sure you are on the last version and the problem persists, you should
report it in the DataFrames.jl issue tracker.
Jacob, tnx for the quick reply and sorry about the redundant for loop. This
is how I decided to do it:
macro scatter(axno, reim, λno, col)
:(*axes[$axno][:scatter](sbox.kx, $reim([OnePump.λ$λno(0., momx, np)
for momx in sbox.kx]), s=15, alpha=0.2, color=$col) )*
end
fig, axes =
Yes, that is probably it. However, it's still unclear to me how that comes
about. Does using actually scan all the packages installed?
My module only has the following using statements
using Lint # it in turns uses nothing else
using TermWin # also it uses nothing else
using LightXML # a simple
Why do you have the asterisks (`*`) in your macro definition?
I'm pretty sure the OnePump.λ$λno interpolation isn't going to work. I
don't think you can't interpolate into a symbol like that. You may have to
try renaming those variables more generically, or do some manual symbol
construction;
After (locally) registering a package Stan (
https://github,com/goedman/Stan.jl.git ) I'm trying to publish it (untagged) to
METADATA.jl, but continue to get below error message.
Github does accept my password (otherwise it gives another error message).
Tagging the package does not seem to
Instructions on fixing it in
https://github.com/JuliaLang/julia/issues/5998
--Tim
On Monday, July 07, 2014 08:14:30 AM Rob J. Goedman wrote:
After (locally) registering a package Stan (
https://github,com/goedman/Stan.jl.git ) I'm trying to publish it
(untagged) to METADATA.jl, but continue
The asterisks were an artefact of the copy-pasting it seems :) Anyway,
you're right about the symbol construction, it failed. So I tried
macro scatter(axno, reim, λno, col)
:(axes[$axno][:scatter](sbox.kx, $reim([symbol(OnePump.λ *
string($λno))(0., momx, np) for momx in sbox.kx]), s=15,
Couldn't you just use a function? (If possible use functions and not macros.)
function scatter(axes, axno, reim::Function, λno::Symbol, col)
axes[axno][:scatter](sbox.kx, reim([OnePump.(λno)(0., momx, np) for momx in
sbox.kx]), s=15, alpha=0.2, color=col)
end
On Mon, 2014-07-07 at 16:05,
Thanks Tim,
That fixed it and clearly the steps have changed a bit over the last few weeks!
Rob J. Goedman
goed...@icloud.com
On Jul 7, 2014, at 8:17 AM, Tim Holy tim.h...@gmail.com wrote:
Instructions on fixing it in
https://github.com/JuliaLang/julia/issues/5998
--Tim
On Monday,
Mea culpa: I missed a using SQLite in a sub-directory. So that's it.
Sorry about the confusion.
On Monday, July 7, 2014 10:13:48 PM UTC+7, Tony Fong wrote:
Yes, that is probably it. However, it's still unclear to me how that comes
about. Does using actually scan all the packages installed?
Yup, that will do just fine it seems, ty!
On Monday, July 7, 2014 5:27:24 PM UTC+2, Mauro wrote:
Couldn't you just use a function? (If possible use functions and not
macros.)
function scatter(axes, axno, reim::Function, λno::Symbol, col)
axes[axno][:scatter](sbox.kx,
Thanks you Jameson but i am here to learn and I woud like to see what's
happening in details.
What code line is giving my color to my graphic cart and display the pixels
I wanna understand very closesly.
Cairo likely doesn't provide the right abstraction for coloring a pixel on
the screen, since its focus is on drawing vector shapes.
Similarly, Tk forces interaction with the library to be through strings,
which is highly inefficient for passing in large chunks of data.
Although if you only want
Ivar: I think Keno ran into the same -malign-double thing before
here https://github.com/JuliaLang/julia/pull/6809#issuecomment-44178953
Are you switching between clang and gcc? You probably need to rerun FFTW's
configure.
Tomas: There's still some lingering fragility in a few places of the
Is there any possibility to define a theme that specifies the color of
labels of legends or any other text? For instance, in the example
plot(dataset(car, SLID), x=Wages, color=Language, Geom.histogram )
I would like to change the color of the text Language to white. So far I
have only been
Any suggestions about how to do this? Thanks in advance.
-Thom
On Friday, July 4, 2014 9:15:27 AM UTC-5, Thomas Covert wrote:
Hi Julia-Users,
I'm trying to parallelize a likelihood calculation in which each step
requires read-only access to several large matrices (10,000 by 3000 in one
Hmm, it must have felt ashamed from my last message because it just let me
delete on a second attempt. However, I now get clear build errors
=[ ERROR:
GLFW ]=
could
Not ideal, but I do this:
using Gadfly, RDatasets
theme = Theme(
minor_label_color=color(#aa),
major_label_color=color(#dcdccc),
point_label_color=color(#dcdccc),
key_title_color=color(#cc),
key_label_color=color(#dcdccc))
plot(dataset(car, SLID), x=Wages, color=Language,
For (1), perhaps something like the parallel map mentioned here
(http://julia.readthedocs.org/en/latest/manual/parallel-computing/) would
work?
On Monday, July 7, 2014 2:00:43 PM UTC-4, Thomas Covert wrote:
Any suggestions about how to do this? Thanks in advance.
-Thom
On Friday, July 4,
I was indeed planning on using parallel map, but I'm still not clear on
whether it will allocate copies of arrays for each worker.
On Monday, July 7, 2014 1:21:13 PM UTC-5, Sid wrote:
For (1), perhaps something like the parallel map mentioned here (
Also check out Distributed Arrays.
On Monday, July 7, 2014 2:21:13 PM UTC-4, Sid wrote:
For (1), perhaps something like the parallel map mentioned here (
http://julia.readthedocs.org/en/latest/manual/parallel-computing/) would
work?
On Monday, July 7, 2014 2:00:43 PM UTC-4, Thomas Covert
I can try to move things manually, but that's not a good solution
but if do it, I get a nice triangle and a gentle background color changing
but note that I used the MSVC dll not he the MinGW one
julia Pkg.test(ModernGL)
INFO: Testing ModernGL
INFO: loaded GLFW 3.0.4 Win32 WGL VisualC DLL
Assuming all your cores are on the same host, then SharedArrays should work
well. I'm not sure @parallel will be as helpful as you might hope---I haven't
used it much yet myself, but I am not sure it will distribute this kind of
workload sensibly. I seem to remember seeing a chunked version of
The short answer: I re-created .julia. I've never accessed METADATA from
another computer.
The longer answer:
My problem was that the first call to Pkg.publish() failed on a a previous
package of mine, HyperDualNumbers. Somehow when that package became part of
JuliaDiff, I think version
1) Is there a way to disable assertions in julia (like an optimize option)?
There is the @inbounds macro which disables bounds checking for built-in
arrays.
For your own code you could have an assertion function which you would
set to the identity function when running optimised and to checking
The flickering reflects the framerate, so congratulations to all of the
people which got intense, unbearable flickering! =)
I run Ubuntu and Windows on my laptop, and on windows the flickering is at
least two times as fast.
Sad though, that Apple is on the Linux side in this regard. I hope that
There is an open issue https://github.com/JuliaLang/julia/issues/7054 for
Pkg.rm()
not removing broken packages.
On Monday, July 7, 2014 10:32:00 PM UTC+3, Simon Danisch wrote:
The flickering reflects the framerate, so congratulations to all of the
people which got intense, unbearable
Re I seem to remember seeing a chunked version of @parallel
somewhere -- Maybe https://github.com/tanmaykm/Blocks.jl?
But on the other hand, this is working perfectly fine for me right now. Do
you have spaces in your absolute path?
Oh the HORROR, never ever spaces in names or paths. I rarely let anything
but MS progs even go to Program Files
For #2, see https://github.com/JuliaLang/julia/pull/6176
--Tim
On Monday, July 07, 2014 09:53:22 AM Sid wrote:
1) Is there a way to disable assertions in julia (like an optimize option)?
2) Something I like to do in C/C++ is have a macro to print a message which
includes the file and the
So what would be the easiest way of plotting the contours obtained with
contour(x, y, Z, h) using PyPlot?
What won’t be “sensible” about the way @parallel distributes the work?
On Jul 7, 2014, at 1:57 PM, Tim Holy tim.h...@gmail.com wrote:
Assuming all your cores are on the same host, then SharedArrays should work
well. I'm not sure @parallel will be as helpful as you might hope---I haven't
Because I haven't used @parallel myself, I don't know what will actually
happen. But an example of nonsensical work distribution would be consecutive
indexes assigned to alternating cores---you'll have far too many cache misses.
--Tim
On Monday, July 07, 2014 02:12:49 PM Thomas Covert wrote:
I had a similar issue (former Matlab user, big likelihood computation that
needs a giant read-only matrix for each observation.) I haven't dealt with
preallocating outputs, but for the read-only matrices can you do something
like:
mybigmatrix = ... #
@everywhere const mybigmatrix =
Hi,
if you feel adventurous, you can try to use my OpenGL packages for this.
In a few days, this might even be a pleasant experience!
Here is how you would do this, right now:
https://gist.github.com/SimonDanisch/98aee37ddb76279cf774
still very messy, but this will ultimately be reduced to
Damn, I accidentally hit post!
Well there is not much to add, besides that you can also use signals from
React for the custom parameters:
customparam1 = Input(1f0)
customparam2 = Input(Vector3(0f0))
const fullscreenquad = RenderObject(
[
:position = GLBuffer(GLfloat[-1,-1, -1,1,
Thank you so much Tony
The problem was (probably) that clang was updated and changed behaviour.
Apparently FFTW felt like recompiling without running configure again, and
then I was in trouble.
For the record, the easiest fix is
make -C deps distclean-fftw
If you want to avoid downloading
Thank you so much Tony
The problem was (probably) that clang was updated and changed behaviour.
Apparently FFTW felt like recompiling without running configure again, and
then I was in trouble.
For the record, the easiest fix is
make -C deps distclean-fftw
If you want to avoid downloading
Julia macros does not work on text, but on the parsed AST. If the modified
syntax parses to something legally, you could do the transformation on the
AST, but you should have a really good reason for wanting to do that. The
syntax (for the user of your macro) would look something like:
Yes, Julia doesn't do reader macros or textual preprocessing, so there's no
way to do this (shorting of preprocessing yourself).
On Mon, Jul 7, 2014 at 2:10 PM, Ivar Nesje iva...@gmail.com wrote:
Julia macros does not work on text, but on the parsed AST. If the modified
syntax parses to
Thanks Tim, this is another interesting capability, and I'm definitely
looking forward to try it out in Julia REPL. It's a bit unusual though, so
I'm still looking for smoother way to work interactively.
I checked one of popular packages (namely, Distributions.jl) and I like
their approach.
Hi Andrei, Tim,
another related issue I sometimes have with interactive development in
Matlab is to load/save workspaces entirely.
For example, in a typical experiment I will have a short driver script
run_exp1.m which contains as its last line a statement like save
result_exp1.mat, saving
For now, maybe something like:
```
cs = contours(X,Y,Z,h)
for c in cs
lvl,lines = c.level, c.lines
for line in lines
line_color = string((lvl - minimum(h))/maximum(h)); # Shades of gray
v = reinterpret(Float64, line.vertices, (2, length(line.vertices)))
A while ago above question came up on the Julia MCMC issue list (
https://github.com/JuliaStats/MCMC.jl/issues/45 ).
I have no idea how much interest there is in such an interface, but I wanted it
at least for my own use, in addition to a similar interface to Jags (which I'm
working on), and
https://github.com/timholy/HDF5.jl
--Tim
On Monday, July 07, 2014 11:14:50 PM Sebastian Nowozin wrote:
Hi Andrei, Tim,
another related issue I sometimes have with interactive development in
Matlab is to load/save workspaces entirely.
For example, in a typical experiment I will have a
Well for GLPlot, Vector{Vector2} is just perfect ;)
Color information could be given via a color map for the iso values, or
just another Vector{Union(Vector3/4, ColorValue/AlphaColorValue)}
Am Sonntag, 29. Juni 2014 00:34:28 UTC+2 schrieb Tomas Lycken:
Huzzah!
We’ve just released
I see.
So the optimal solution would be to do everything inside a shader, or with
OpenCL, as uploading a complete image to the framebuffer is extraordinary
expensive.
Here's a demo of raytracing the Julia-Set with OpenCL and then displaying
the resulting image in OpenGL, staying completely in
Thanks you very mutch I will look this closely in order to understand the
basics.
I mean for exemple:
why a shader is fast and what it is...
how windows is using and talking with the graphic card...
what's happening in my computer
and more important how do I write only the code line I
It is great to have a Stan interface. I personally do not do MCMC stuff,
but there are mailing list threads discussing it. Julia and Stan also
shared a grant proposal at one point - so now this makes everything come a
full circle!
-viral
On Monday, July 7, 2014 3:28:17 PM UTC-7, Rob J Goedman
This is really nice to have. I'll definitely be using it for some of my work.
Now that Keno's been working on a C++ FFI, we might even be able to avoid using
CmdStan.
-- John
On Jul 7, 2014, at 4:40 PM, Viral Shah vi...@mayin.org wrote:
It is great to have a Stan interface. I personally do
Dear Julia Users,
I am currently developing/converting several Bayesian spatial econometric
models from MATLAB into Julia. I have successfully coded the spatial
autoregressive model (SAR) with diffuse priors in Julia but have a question
regarding the use of sparse matrix algorithms. The models
I absolutely love Jewel and where this project is headed. Thanks for all
the effort!
On Sunday, June 29, 2014 2:46:21 AM UTC-7, Mike Innes wrote:
Hey all,
I've released the latest version of the Julia environment
https://github.com/one-more-minute/Jupiter-LT I'm building. There are a
Andrei,
Not sure if my way is better, but it seems to work ok for me. It is also biased
towards OSX, although I know a similar approach has been used on other
platforms (for R).
I have Julia's REPL window side by side to TextMate (or similar editor). I
extended Julia's plugin for Textmate
This is a pretty impressive range of different talks for a single event. Very
cool.
— John
On Jul 6, 2014, at 5:35 PM, Sorami Hisamoto therem...@gmail.com wrote:
Last week's JuliaTokyo meetup in Japan ended in great success!
Around 40 people attended, 6 main talks and 4 lightning talks.
81 matches
Mail list logo
