I'm considering changing the type of string literals. That requires this
change to make tests continue to pass. Whether that happens or not, this
change makes the tests independent of the type of string literals, which is
a good thing.
On Fri, Aug 21, 2015 at 7:42 PM, Scott Jones
https://github.com/JuliaLang/julia/issues/269
On Sat, Aug 22, 2015 at 1:33 AM, Paul Thompson pm...@case.edu wrote:
Hi:
I want to define two types, and each will have a field that is the other
type. For instance:
type Foo
bar::Bar
otherfield1
otherfield2
end
type Bar
foo::Foo
No, I haven't yet – but might still be interested :)
Currently Julia only drops trailing singleton dimensions in indexing. So if
F is n-by-m-by-k, then F[:, i, :] is n-by-1-by-k. You need to use reshape
or squeeze to convert that to an n-by-k 2-dimensional matrix.
Note that writing code in this direct-port slicing style is going to
involve a lot
Not sure how that's really a response to what I said, which acknowledges
that it's an ambiguity...
On Sat, Aug 22, 2015 at 12:58 AM, Michele Zaffalon
michele.zaffa...@gmail.com wrote:
Verbosity aside, (which may also be disputable since errors are supposed
to be rare...), I still do not see
You are right, my reply was unnecessary.
On Sat, Aug 22, 2015 at 10:14 AM, Stefan Karpinski ste...@karpinski.org
wrote:
Not sure how that's really a response to what I said, which acknowledges
that it's an ambiguity...
On Sat, Aug 22, 2015 at 12:58 AM, Michele Zaffalon
Hi I tried using Gadfly and Interact and tried the example from Interact.jl
github notebooks.
My code doesn't evaluate. It just keeps processing and doesn't plot
anything, the kernel however is shown to be busy
@manipulate for ϕ = 0:π/16:4π, f = [:sin = sin, :cos = cos], both = false
if
Hi,
I caught myself wondering about the correct way to use function and
function! -- or rather, how other people deal with this.
Let's say I have a simple function that operates on an array, and I
want a version to modify the original object, and one that doesn't.
Is this the correct way of
Will `ArrayView` overcome the unnecessary copying?
On Saturday, August 22, 2015 at 11:54:39 AM UTC+2, Tony Kelman wrote:
Currently Julia only drops trailing singleton dimensions in indexing. So
if F is n-by-m-by-k, then F[:, i, :] is n-by-1-by-k. You need to use
reshape or squeeze to
On Sat, Aug 22, 2015 at 2:49 PM, Tero Frondelius
tero.frondel...@gmail.com wrote:
Why testvariable is not incremented?
function incrementvariable(numb)
numb += 1
end
function testing()
global testvariable = 0
for i = 1:3
incrementvariable(Ptr{testvariable})
end
On Sat, Aug 22, 2015 at 6:34 PM, Tero Frondelius
tero.frondel...@gmail.com wrote:
Can you show how to use the Ref in practice, this is merely for academic
purposes, because I had to use the list style anyways for further use of my
code?
For your questions: this is no performance critical.
Oups, a typo ``testvariable+=1`
On Saturday, August 22, 2015 at 11:35:33 PM UTC+2, Sisyphuss wrote:
You can write `testvariable=incrementvariable(testvariable)`.
But it's wield to write a dedicated function to implement the increment.
Can't you just write `testvariable+=testvariable`?
In particular, this comment has a work-around:
https://github.com/JuliaLang/julia/issues/269#issuecomment-68421745
On Saturday, August 22, 2015, Stefan Karpinski ste...@karpinski.org wrote:
https://github.com/JuliaLang/julia/issues/269
On Sat, Aug 22, 2015 at 1:33 AM, Paul Thompson
Hi Felipe,
This looks great. Thanks for sharing. We'll see if we can help develop this
further.
Thanks also for pointing to Imanuel's FinancialMarkets.jl
https://github.com/imanuelcostigan/FinancialMarkets.jl. His package on
OTC derivative data (dataonderivatives.jl
You can write `testvariable=incrementvariable(testvariable)`.
But it's wield to write a dedicated function to implement the increment.
Can't you just write `testvariable+=testvariable`?
On Saturday, August 22, 2015 at 8:49:33 PM UTC+2, Tero Frondelius wrote:
Why testvariable is not
On Saturday, August 22, 2015 at 9:47:56 AM UTC+2, Stefan Karpinski wrote:
I'm considering changing the type of string literals. That requires this
change to make tests continue to pass. Whether that happens or not, this
change makes the tests independent of the type of string literals,
Bright ideas? I am afraid not.
Had you said dumb ideas, I would have suggested either to remove it
completely, leave it as a shortname for throw(ErrorException) or to replace
with something that cannot be caught, but for this one I cannot suggest a
use case (maybe writing way past array
Also, Julia compiles code on the fly, so you should put it in a function,
then call it once before timing it to let it warm up. You can also
profile the code and use ProfileView.jl to view where the bottleneck is.
There are many tricks to making things faster.
On Saturday, August 22, 2015 at
Thanks for these infos. I will definitely pre-allocate memory (in fact
I do in the real code).
t
Le samedi 22 août 2015 à 14:24 -0700, Sisyphuss a écrit :
I don't think it's much less efficient to copy` in your second
example. In
the second function., you should allocate memory for `y`
Hi all,
I've started working on a Julia PLplot wrapper
https://github.com/wildart/PLplot.jl. PLplot is a powerful cross-platform
library
that can be used to create standard x-y plots, semi-log plots, log-log
plots, contour plots, 3D surface plots, mesh plots, bar and pie charts,
dynamic
What do you mean copying the object is inefficient? Do you mean that it is
cheaper to build it from scratch than to start with the copy?
On Saturday, August 22, 2015 at 9:51:48 PM UTC+2, Timothée Poisot wrote:
Hi,
I caught myself wondering about the correct way to use function and
BTW, I'm not sure I use @parallel correctly, so I tried to start tasks
manually with @async too:
@time @sync for url in urls
@async begin
resp = get(url)
println(Status: $(resp.status))
end
end
But I didn't notice any difference in performance.
On Sunday, August 23,
Most likely your github ssh keys were not set up? Or needed to be reset?
On Saturday, August 22, 2015 at 11:02:53 AM UTC-7, Uwe Fechner wrote:
Ok, publishing METADATA manually worked:
Publishing METADATA manually
If Pkg.publish()
Hello,
just added *InterestRates.jl* package to METADATA. It provides tools for
term structure models (Nelson Siegel, Svensson) and interpolation of
yield curves in finance (Linear, FlatForward, natural Cubic Spline).
Hopefully helpful to someone.
Installation: *Pkg.add(InterestRates)*. Docs
Why testvariable is not incremented?
function incrementvariable(numb)
numb += 1
end
function testing()
global testvariable = 0
for i = 1:3
incrementvariable(Ptr{testvariable})
end
println(testvariable)
end
Or actually what should I change to get the testvariable
Another (maybe better) option is
```
baz!(z,x) = begin ... return z; end
baz(x) = baz!(similar(x),x)
```
You can call them by
```
baz!(x,x)
z = baz(x)
```
On Saturday, August 22, 2015 at 11:24:51 PM UTC+2, Sisyphuss wrote:
I don't think it's much less efficient to copy` in your second
Can you show how to use the Ref in practice, this is merely for academic
purposes, because I had to use the list style anyways for further use of my
code?
For your questions: this is no performance critical.
On Sunday, August 23, 2015 at 1:18:00 AM UTC+3, Yichao Yu wrote:
On Sat, Aug 22,
On Sat, Aug 22, 2015 at 6:55 PM, Michael Wang
cwang.michael...@gmail.com wrote:
I am new to Julia. I heard that Julia has the performance with Cpp even
though it is a high level language. I tested an example on my machine,
however, the result was that Julia was in the same ballpark with R not
I'm writing a kind of a web scanner that should retrieve and analyze about
100k URLs as fast as possible. Of course, it will take time anyway, but I'm
looking for how to utilize my CPUs and network as much as possible.
My initial approach was to add all available processors, pack urls into
Pkg.tag(NaNMath)
Pkg.publish()
is all you should have to do.
Well, it is not that easy.
I did:
Pkg.tag(NaNMath,:minor)
which worked well.
Pkg.pubish() did not work.
First I had to do:
git config --global github.user ufechner7
which worked fine.
Than I had to do:
Pkg.add(JSON)
It would be nice if the error message of the missing JSON package would
Ok, publishing METADATA manually worked:
Publishing METADATA manually
If Pkg.publish()
http://julia.readthedocs.org/en/latest/stdlib/pkg/#Base.Pkg.publish fails
you can follow these instructions to manually publish your package.
By “forking” the main METADATA repository, you can create a
Here they are! https://github.com/quinnj/TimeZones.jl
IMO it's better than pytz, which is exciting.
All credit goes to Curtis Vogt @omus.
No worries, if you've got any bright ideas how we should resolve this
throw/error thing...
On Sat, Aug 22, 2015 at 7:06 AM, Michele Zaffalon
michele.zaffa...@gmail.com wrote:
You are right, my reply was unnecessary.
On Sat, Aug 22, 2015 at 10:14 AM, Stefan Karpinski ste...@karpinski.org
I'd really appreciate any ideas I know this is very little to go on but
maybe some of you can just spitball some ideas that might work. I really
would like to use this functionality
On Saturday, August 22, 2015 at 5:40:13 PM UTC+5:30, Shubham Bhushan wrote:
Hi I tried using Gadfly and
Hello,
I was invited to tag a new version of the package NaNMath.jl .
https://github.com/mlubin/NaNMath.jl
Does anyone know, how to do this?
Best regards:
Uwe Fechner
Yes! There is a good documentation here:
http://julia.readthedocs.org/en/latest/manual/packages/
It works for 0.3 as well as 0.4.
t
Le 2015-08-22 12:20, Uwe Fechner uwe.fechner@gmail.com a écrit :
Hello,
I was invited to tag a new version of the package NaNMath.jl .
Ok. As far as I understand, I first have to create a tag in the NaNMath.jl
repository.
Then I have to fork METADATA.jl. (I did this).
But what next?
Am Samstag, 22. August 2015 18:31:03 UTC+2 schrieb Timothée Poisot:
Yes! There is a good documentation here:
Which version of Jupyter / IPython are you using?
This is all going through an upheaval at the moment. My guess is that you
need
Pkg.checkout(Interact)
to get the latest version of Interact.jl that works with version 3.0 of the
Notebook. (But not, apparently, with the very latest version 4.0
Uwe,
The way I proceed is to:
1) Clone the forked version of METADATA.jl to your desktop.
2) Remove the entry for NaNMath.jl in this local, cloned version of METADATA.jl.
3) Replace it with the entry from your NanMath.jl in .julia/v0.x METADATA.jl.
4) Merge back to your forked verion on
40 matches
Mail list logo
