If I understand correctly, julia reads ~/.juliarc.jl, but I would like it,
on start-up, to find the innermost containing directory that contains a
juliarc file, and read that. That way, I can have project-specific settings
for my repl.
Is there a julia-standard way to do that? It doesn't seem t
in[2:end] - in[1:end-1]
On Saturday, February 7, 2015 at 7:19:39 AM UTC-5, Evan Pu wrote:
>
> say I want to compute a pair-wise diff for all the elements in the array.
> input:[1, 2, 4, 7, 8]
> output:[1, 2, 3, 1]
>
> is there some kind of "beautiful" way of doing it? i.e. w/o u
On Saturday, 7 February 2015 15:56:02 UTC-5, Chris wrote:
>
> I have recently started trying to use the ODE solvers in Sundials.jl, as
> there are several features I need that are not yet implemented in ODE.jl.
> One of these is the ability to set the maximum step size. I see there is a
> method
You can try build with ieee-fp set for icc.
glenn
> On Feb 7, 2015, at 2:44 PM, Mark Wells wrote:
>
> Hi everyone,
>
> I am rebuilding Julia on my arch linux machine. I am compiling with the
> Intel icc/i++ compilers and the Intel MKL using the instructions on
> https://github.com/JuliaLang
Docs added:
https://github.com/JuliaLang/julia/commit/15c739ccbae9d7b42a876655df249607d500537f
On Sat, Feb 7, 2015 at 8:45 PM, Jacob Quinn wrote:
> Great to hear about the performance gains. I tried to push as much of the
> "parsing" work into the actual creation of the DateFormat object for thi
What version (sha and/or branch) of Julia are you building? linalg4.jl does
not have a line 263 on master. Can you edit test/linalg4.jl and change
debug from false to true and try again? So we can at least see exactly what
operation causes the mismatch. It's likely a problem in complex calling
On Saturday, February 7, 2015 at 3:12:35 PM UTC-5, Kevin Squire wrote:
>
> I have mixed feelings, as I'm somehow one of those people who frequently
> breaks out of loops and wants to know the value of the variable at the last
> iteration.
>
Even more commonly, I want to know (and use) the value
Great to hear about the performance gains. I tried to push as much of the
"parsing" work into the actual creation of the DateFormat object for this
purpose. The vectorized versions use this performance trick, but like I
said, I probably need to add a little documentation for cases like this.
-Jaco
Or: "Take Julia for a spin and see how it corners!"
On Saturday, February 7, 2015 at 2:31:42 PM UTC-8, J Luis wrote:
>
> or
> "That corner is mine"
> or
> "The Julia corner"
>
> sábado, 7 de Fevereiro de 2015 às 20:31:41 UTC, Tobias Knopp escreveu:
>>
>> :-) this is so funny! But still not a marke
On Saturday, February 7, 2015 at 3:03:14 PM UTC-8, Eric S wrote:
>
> That is what I get running Julia from the Terminal.
>
> Eric
>
>
Can you run git from the command line? I'm wondering whether you need to
accept the license agreement or something.
So far, the only things I'm missing in ODE.jl are things that are already
planned or seem to be in progress already: step size is one, and output at
specified times is the other major one.
On Saturday, February 7, 2015 at 5:47:41 PM UTC-5, Alex wrote:
>
> I think (almost) all those functions tak
Call the general version pairwise() then?
On Sunday, February 8, 2015 at 6:01:43 AM UTC+11, Stefan Karpinski wrote:
>
> I actually was just wondering if we shouldn't add an "op" argument to diff
> to allow it to do other things besides the differences between adjacent
> pairs. The name becomes a
That is what I get running Julia from the Terminal.
Eric
On Saturday, February 7, 2015 at 4:34:04 PM UTC-6, Kevin Squire wrote:
>
> Hi Eric,
>
> What happens if you run the command at the Terminal command line?
>
> Kevin
>
> On Sat, Feb 7, 2015 at 1:55 PM, Eric S >
> wrote:
>
>> Just installed
Hi everyone,
I am rebuilding Julia on my arch linux machine. I am compiling with the
Intel icc/i++ compilers and the Intel MKL using the instructions on
https://github.com/JuliaLang/julia. I am not using the intel fortran
compiler (I set that option equal to 0).
Building Julia seems to work
I think (almost) all those functions take a pointer to a "memory structure",
which was previously created by *Create. But I see that Mauro already answered
that one.
May I ask what you are missing in ODE.jl besides setting the maximum step size?
I think it might be helpful if you could file an
Hi Eric,
What happens if you run the command at the Terminal command line?
Kevin
On Sat, Feb 7, 2015 at 1:55 PM, Eric S wrote:
> Just installed Julia and Juno on OS X 10.10.2. I don't seem to be able to
> install packages. For example:
>
> *julia> **Pkg.add("Jewel")*
>
> *ERROR: failed process
or
"That corner is mine"
or
"The Julia corner"
sábado, 7 de Fevereiro de 2015 às 20:31:41 UTC, Tobias Knopp escreveu:
>
> :-) this is so funny! But still not a marketing slogan. Whats about:
>
> "The speed lays in the corner" or
> "The truth lays in the corner"...
>
> Am Samstag, 7. Februar 2015 2
That does appear to work, thank you!
On Saturday, February 7, 2015 at 5:11:27 PM UTC-5, Mauro wrote:
>
> I used this to set another property, following their example (not the
> one with the simple interface):
>
> mem = Sundials.CVodeCreate(Sundials.CV_BDF, Sundials.CV_NEWTON)
>
I used this to set another property, following their example (not the
one with the simple interface):
mem = Sundials.CVodeCreate(Sundials.CV_BDF, Sundials.CV_NEWTON)
...
flag = Sundials.CVodeSetMaxNumSteps(mem, -1)
...
I would think that adapting this to your case
I've been doing a lot of streaming data analysis in Julia lately, so I
finally put together a package with some core functionality for working
with data streams:
https://github.com/johnmyleswhite/StreamStats.jl
My hope is that the community can help refine the design over time. After
it's sett
Just installed Julia and Juno on OS X 10.10.2. I don't seem to be able to
install packages. For example:
*julia> **Pkg.add("Jewel")*
*ERROR: failed process: Process(`git
--git-dir=/Users/ericshain/.julia/.cache/Stats merge-base
78f5810a78fa8bee684137d703d21eca3b1d8c78
8208e29af9f80ef633e50884
You can try using Gt or Lt or Ge or Le to compare syboliv values. That should
work.
What happens when a "for" loop is parallelized? (Not at a computer, so
can't check.)
Kevin
On Saturday, February 7, 2015, Stefan Karpinski
wrote:
> When I talked with Jeff about this yesterday, he felt that since the
> current behavior is exactly how a while loop with a let block around the
>
I have recently started trying to use the ODE solvers in Sundials.jl, as
there are several features I need that are not yet implemented in ODE.jl.
One of these is the ability to set the maximum step size. I see there is a
method Sundials.CVodeSetMaxStep(::Any,::Any), and looking at the source it
When I talked with Jeff about this yesterday, he felt that since the
current behavior is exactly how a while loop with a let block around the
loop body would behave (this is essentially what our for loops are), the
behavior is in line with all the other scoping rules. Parity with
comprehensions is
:-) this is so funny! But still not a marketing slogan. Whats about:
"The speed lays in the corner" or
"The truth lays in the corner"...
Am Samstag, 7. Februar 2015 20:54:11 UTC+1 schrieb J Luis:
>
>
>
> I'm going to refer to that plot from now on as "nobody puts Julia in a
>> corner".
>>
>
> mi
I have mixed feelings, as I'm somehow one of those people who frequently
breaks out of loops and wants to know the value of the variable at the last
iteration. That said, I can see how nonlocal semantics can be confusing
(and the workarounds suggested by Stephen and Mauro).
David, to move this di
I'm going to refer to that plot from now on as "nobody puts Julia in a
> corner".
>
might be more faithful if you call it "Julia puts itself in a corner"
>
> On Sat, Feb 7, 2015 at 2:41 PM, Stefan Karpinski > wrote:
>
>> That's a good one too, Jacob, but it was indeed Simon Danisch's plot
I'm going to refer to that plot from now on as "nobody puts Julia in a
corner".
On Sat, Feb 7, 2015 at 2:41 PM, Stefan Karpinski
wrote:
> That's a good one too, Jacob, but it was indeed Simon Danisch's plot I was
> thinking of. Thanks, Andreas!
>
> On Sat, Feb 7, 2015 at 2:33 PM, Jacob Quinn
>
That's a good one too, Jacob, but it was indeed Simon Danisch's plot I was
thinking of. Thanks, Andreas!
On Sat, Feb 7, 2015 at 2:33 PM, Jacob Quinn wrote:
> Was is the graph I posted? It compares several of the "Great Computer
> Shootout" benchmarks amongst top languages.
>
>
>
> On Sat, Feb 7,
Was is the graph I posted? It compares several of the "Great Computer
Shootout" benchmarks amongst top languages.
On Sat, Feb 7, 2015 at 12:31 PM, Andreas Noack wrote:
> Simon Danisch in
> https://groups.google.com/forum/#!topic/julia-users/BYRAeQJuvTw
>
> 2015-02-07 14:26 GMT-05:00 Stefan Kar
Simon Danisch in
https://groups.google.com/forum/#!topic/julia-users/BYRAeQJuvTw
2015-02-07 14:26 GMT-05:00 Stefan Karpinski :
> There was a thread at some point where someone posted a plot comparing
> lines of code versus performance for a bunch of benchmarks (probably our
> microbenchmarks), wh
Le samedi 07 février 2015 à 11:23 -0800, Keith Kee a écrit :
> Hi Milan,
>
>
> I used the too csvfix 1.6 validate function by Neil Butterworth for
> the csv file (only for csv, not for wsv) and discovered two corrupted
> lines and then it worked.
>
>
> The link to the manual:
> http://neilb.bit
There was a thread at some point where someone posted a plot comparing
lines of code versus performance for a bunch of benchmarks (probably our
microbenchmarks), where Julia was alone in the very fast & very small line
count corner. I can't for the life of me find that thread now. Does anyone
recal
Hi Milan,
I used the too csvfix 1.6 validate function by Neil Butterworth for the csv
file (only for csv, not for wsv) and discovered two corrupted lines and
then it worked.
The link to the manual:
http://neilb.bitbucket.org/csvfix/manual/csvfix16/csvfix.html
As a feature request, it would be
A subtle difference for Julia v0.3:
const nchans = size(userData.samples, 2)
const samples = sub(userData.samples, tuple(rng, ntuple(nchans - 1, _
->:)...)...)
because (I think) of https://github.com/JuliaLang/julia/issues/4869
On Saturday, February 7, 2015 at 1:44:28 PM UTC-5, Daniel Casimiro w
I actually was just wondering if we shouldn't add an "op" argument to diff
to allow it to do other things besides the differences between adjacent
pairs. The name becomes a bit odd then, but might be ok.
On Sat, Feb 7, 2015 at 12:44 PM, Steven G. Johnson
wrote:
> In general, I think people need
Hi Joseph,
I'm no expert, but, I managed to install Juno on Mac OS X 10.9 with no
problem. Have you tried installing Julia first then Juno. If you and your
students are working on Apple Macs I would recommend installing Julia via
Homebrew because it then becomes easier to maintain any availabl
FWIW, I came up with the following snippet to get the behavior that I want:
const nchans = size(userData.samples, 2)
const samples = sub(userData.samples, tuple(rng, ntuple(nchans - 1, _
->:)...))
It appears to work.
On Sunday, February 1, 2015 at 8:38:09 PM UTC-5, Tim Holy wrote:
>
> SubArrays
In general, I think people need to unlearn the prejudices against loops
that have been instilled by languages like Matlab; loops are compact, loops
are beautiful, loops are fast.
On Saturday, February 7, 2015 at 6:45:40 AM UTC-8, Evan Pu wrote:
>
> i see, maybe my example is too special.
> What i meant is I have an array of elements and I have a binary operator
> which I want to apply to every adjacent elements.
>
No, I think you have to use a loop or comprehension, e.g
Hi Milan,
No worries, greatly appreciate that you have taken time to help, will also
do as you advised.
Thanks.
Keith
On Saturday, 7 February 2015 08:17:51 UTC-8, Milan Bouchet-Valat wrote:
>
> Le vendredi 06 février 2015 à 15:01 -0800, Keith Kee a écrit :
> > Hi Milan,
> >
> >
> > Thanks
Take this with a grain of salt, but I have had someone else tell me that
Julia failed to install packages on OS X because of git permission errors.
Le vendredi 06 février 2015 à 15:01 -0800, Keith Kee a écrit :
> Hi Milan,
>
>
> Thanks for your advice.
>
>
> I spotted one corruption in a smaller sample of 3000 lines and then it
> worked.
>
>
> Then a tried a larger number of 1 lines and it gave the following:
> Saw 1 rows, 4 colu
I've used Winston and Tk for this kind of stuff. I have code for panning
through an iterable using simple controls. The code is messy, but if it is
of interest, you can find it here
https://github.com/grero/Visualizer.jl
On Saturday, February 7, 2015 at 2:07:23 PM UTC+8, Christoph Ortner wrote:
i see, maybe my example is too special.
What i meant is I have an array of elements and I have a binary operator
which I want to apply to every adjacent elements.
On Saturday, February 7, 2015 at 7:26:55 AM UTC-5, Tamas Papp wrote:
>
> diff([1, 2, 4, 7, 8])
>
> I don't think it gets more beautif
Perhaps an addition to Color.jl's documentation would be in order, then? It's
super-easy to edit docs even if you don't have commit privileges (just click
on the file on the GitHub site, then click on the pencil icon).
--Tim
On Friday, February 06, 2015 02:38:22 PM Christopher Lee wrote:
> That
diff([1, 2, 4, 7, 8])
I don't think it gets more beautiful than this :D
Best,
Tamas
On Sat, Feb 07 2015, Evan Pu wrote:
> say I want to compute a pair-wise diff for all the elements in the array.
> input:[1, 2, 4, 7, 8]
> output:[1, 2, 3, 1]
>
> is there some kind of "beautifu
say I want to compute a pair-wise diff for all the elements in the array.
input:[1, 2, 4, 7, 8]
output:[1, 2, 3, 1]
is there some kind of "beautiful" way of doing it? i.e. w/o using a for
loop nor using explicit indecies
Can you remind me what change I need to make to my Make.user file to build
with the standard LLVM that Julia currently uses, instead of llvm-svn. I
will try changing that, just so we can eliminate the possibility that this
is an LLVM bug.
Note that the example above is complete. I don't load any m
No, definitely didn't do that.
On 7 February 2015 at 01:32, Stefan Karpinski wrote:
> Did you call gc_disable() at some point?
>
>
> On Friday, February 6, 2015, Bill Hart
> wrote:
>
>> Oh wait a minute. Running gc() did clean up the specific example that I
>> gave, just not the original code t
> I would be ok with loop variables always being local to the loop so that
> this would happen:
>
> julia> i = 0
> 0
>
> julia> for i = 1:5
>println(i)
>end
> 1
> 2
> 3
> 4
> 5
>
> julia> i
> 0
>
>
> That would simplify the semantics of for loops at the cost of making it
> harde
52 matches
Mail list logo
