Thanks. Indeed I missed this in the docs, so no ticket needs to be opened.
The fact that overloading deecopy_internal is fully supported is perfectly
fine.
However, could you please explain to me what is involved in updating dict?
I understand an ObjectIdDict is a hash table whose keys are object
Ok, thanks!
matrix = [ Int64[] for i=1:2, j=1:2 ]
Friday, February 26, 2016 at 11:51:09 AM UTC+11, Ilya Orson wrote:
>
> Hello everyone,
>
> I am working with a matrix of initially empty vectors and wanted to fill
> each vector with specific elements using push!, but I get a weird result:
>
>
> matrix = fi
On Thu, Feb 25, 2016 at 7:51 PM, Ilya Orson wrote:
> Hello everyone,
>
> I am working with a matrix of initially empty vectors and wanted to fill
> each vector with specific elements using push!, but I get a weird result:
>
>
> matrix = fill(Int64[],2,2)
>
> 2x2 Array{Array{Int64,1},2}:
> Int64[]
Hello everyone,
I am working with a matrix of initially empty vectors and wanted to fill
each vector with specific elements using push!, but I get a weird result:
matrix = fill(Int64[],2,2)
2x2 Array{Array{Int64,1},2}:
Int64[] Int64[]
Int64[] Int64[]
push!(matrix[1,1],3,2)
2-element A
I believe 2014 was around 80 people. Not sure about 2015 but upwards of 200
sounds right.
On Thu, Feb 25, 2016 at 4:49 PM, Pontus Stenetorp
wrote:
> On 25 February 2016 at 20:12, Cedric St-Jean
> wrote:
> >
> > Out of curiosity, how many people attended last year?
>
> There are not official num
as Erik says:
immutable Point # immutable is the kind of type to use when you want
memory to be immediate
x::Cdouble
y::Cdouble
end
a = [Point(0.0, 0.0),Point(0.2,0.3),Point(1.1, 1.3)]
sizeof(a)
48
On Thursday, February 25, 2016 at 4:34:20 PM UTC-5, Martin Kuzma wrote:
>
> I have the the foll
Your type `Point` has in C a layout comparable to `struct*`, not
`struct. You could use `immutable Point` instead, and what you're
trying to achieve should work.
-erik
On Thu, Feb 25, 2016 at 4:34 PM, Martin Kuzma wrote:
> I have the the following type defined:
>
> type Point
> x::Cdouble
> y::C
On 25 February 2016 at 20:12, Cedric St-Jean wrote:
>
> Out of curiosity, how many people attended last year?
There are not official numbers by any means, but around 250 in 2015.
This was quite a significant increase compared to 2014 when I think we
were about 50 people or so.
Pontus
I have the the following type defined:
type Point
x::Cdouble
y::Cdouble
end
Calling sizeof returns 16 as expected. But when I have for example:
a = [Point(0.0, 0.0),Point(0.2,0.3),Point(1.1, 1.3)]
sizeof returns *24*. But I expected it to return 3 * 16 = *48*. (It looks
like in the array are st
The `MPI` package contains this macro:
```Julia
macro mpi_do(mgr, expr)
expr = Base.localize_vars(:(()->$expr), false)
:(mpi_do($(esc(mgr)), $(esc(expr
end
```
This fails on the current master with
```
julia> x=1
julia> @mpi_do MPIManager() info(x)
ERROR: syntax: unhandled expr (locali
Yes, that would help.
On Thursday, February 25, 2016 at 11:48:44 AM UTC-5, Josh Langsfeld wrote:
>
> Maybe a solution to this kind of design problem would be to allow multiple
> top-level modules per package, with syntax to let you load the particular
> one you want/need.
>
> On Thursday, Februa
On Thu, Feb 25, 2016 at 2:57 PM, Toivo Henningsson wrote:
> It seems very reasonable that you should be able to overload deepcopy for a
> given type, and that if that has to be done on a specific way, it should be
> mentioned in the documentation for deepcopy. Open an issue?
Please read the doc
It seems very reasonable that you should be able to overload deepcopy for a
given type, and that if that has to be done on a specific way, it should be
mentioned in the documentation for deepcopy. Open an issue?
Maybe a solution to this kind of design problem would be to allow multiple
top-level modules per package, with syntax to let you load the particular
one you want/need.
On Thursday, February 25, 2016 at 1:10:28 AM UTC-5, John Myles White wrote:
>
> I don't think Julia is really amenable to this k
On Thu, 2016-02-25 at 15:57, J Luis wrote:
> Incidentally, this errors in 0.5 (but works fine in 0.4)
>
> julia> s=0.0;
>
> julia> m=rand(1,1000);
>
> julia> @time for el in m
>s = s + el
>end
> ERROR: UndefVarError: s not defined
> [inlined code] from .\none:2
> in anonymous a
Nope
julia> s=0.0;
julia> m=rand(1,1000);
julia> @time for i in 1:1, j in 1:1000
s = s + m[i,j]
end
ERROR: UndefVarError: s not defined
[inlined code] from .\none:2
in anonymous at .\no file:4294967295
in eval at :0
quinta-feira, 25 de Fevereiro de 2016 às 14:58:44 UT
On Thu, Feb 25, 2016 at 9:57 AM, Yichao Yu wrote:
> On Thu, Feb 25, 2016 at 9:33 AM, abc wrote:
>> Ah, yes, this seems to have fixed the problem, thank you.
>>
>> Now, when not in global scope, using eachindex is definitely the fastest
>> approach.
>> I made some measurements over 1000 runs for e
On Thu, Feb 25, 2016 at 9:33 AM, abc wrote:
> Ah, yes, this seems to have fixed the problem, thank you.
>
> Now, when not in global scope, using eachindex is definitely the fastest
> approach.
> I made some measurements over 1000 runs for each of the approaches in the
> original post, here are the
Hello all,
I am collecting as many user codes as I can lay my hands on for the
Indexing benchmarks. Primarily the goal is to evaluate array views and
performance / correctness impacts.
If this effort is successful, it will also create a body of user codes that
we can use for longer term perfo
Incidentally, this errors in 0.5 (but works fine in 0.4)
julia> s=0.0;
julia> m=rand(1,1000);
julia> @time for el in m
s = s + el
end
ERROR: UndefVarError: s not defined
[inlined code] from .\none:2
in anonymous at .\no file:4294967295
in eval at :0
quinta-f
Ah, yes, this seems to have fixed the problem, thank you.
Now, when not in global scope, using eachindex is definitely the fastest
approach.
I made some measurements over 1000 runs for each of the approaches in the
original post, here are the averages:
Using eachindex: 0.0026683
Using ranges: 0.
I'm glad to announce you that I recently tagged a new package called
StringEncodings.jl. As its name implies, it allows converting text
between character encodings: this includes decoding vectors of bytes to
Julia strings, encoding strings to such vectors, as well as reading
from and writing to enc
Can you try it not in global scope?
On Thu, Feb 25, 2016 at 5:17 AM, abc wrote:
> For the following matrix
> my_matrix = randn(1, 1000)
> using eachindex to access all elements is much slower than using ranges
> or even for el in my_matrix, even though it says in the documentation (
> http:/
At JuliaCon 2015 this last summer, after seeing Keno's presentation on Cxx,
some of us were talking about how it would be nice to also have a simpler
C REPL for Julia, just like his C++ REPL in Cxx.jl.
People even said that having the C REPL like that would even be something
they'd be happy to pa
For the following matrix
my_matrix = randn(1, 1000)
using eachindex to access all elements is much slower than using ranges or
even for el in my_matrix, even though it says in the documentation
(http://docs.julialang.org/en/release-0.4/stdlib/arrays/#Base.eachindex)
that eachindex uses rang
I am relieved to hear you what not use it in practice, I was trying really
hard to grasp the utility of it :-) It is a very elegant piece of code,
though.
Den torsdag den 25. februar 2016 kl. 01.18.16 UTC+1 skrev Tom Breloff:
>
> I think he was just referring to the fact that it's "one more thin
27 matches
Mail list logo
