Thanks Tim,
I need the single column matrix literal in one place only, so it's not
worth the use of a function.
But I came to understand that there is no need to have a way to write
literals for everything, and that the "right/standard/official" way to do
what I want is in fact the use of reshape.
reshape let us do
reshape([1, 2, 3], (3,1))
and can be used to any other shapes one can need.
Additionally, we can do
[1 2 3]'
as a trick.
Thanks!
On Monday, January 27, 2014 11:59:29 AM UTC, Tim Holy wrote:
>
> You have several options, e.g., define a
>
> colv(v::AbstractVector) = reshape(v, length(v), 1)
>
> function. Or you can define a comparison function,
>
> cmpcolv(v1, v2)
>
> which will do the comparison with what you receive from SymPy, accepting
> either AbstractVector or AbstractMatrix arguments.
>
> The latter would be preferable because you can implement it without
> triggering
> any memory allocation.
>
> --Tim
>
> On Sunday, January 26, 2014 10:47:43 PM Cristóvão Duarte Sousa wrote:
> > BTW, I need that to compare against a column vector with 2
> > dimensions received from Python through SymPy.jl.
> >
> > It is not a big issue, I can perfectly use [1 2 3]'.
> > I'm just wondering if it is the best way to do it.
> >
> > On Monday, January 27, 2014 6:39:54 AM UTC, Cristóvão Duarte Sousa
> wrote:
> > > Hi,
> > >
> > > What is the best way to write down a "single column matrix" literal?
> > >
> > > For exampe, to obtain this:
> > > 3x1 Array{Int64,2}: # 2 dimensions
> > >
> > > 1
> > > 2
> > > 3
> > >
> > > This
> > >
> > > [[1], [2], [3]]
> > >
> > > doesn't work,
> > > this
> > >
> > > [1, 2, 3]''
> > >
> > > works but seems it can stop working in the future
> > > (#2686<https://github.com/JuliaLang/julia/issues/2686> ),
> > > and, finally, this
> > >
> > > [1 2 3]'
> > >
> > > indeed does what I want.
> > > However, is it the use of the transpose the only option?
> > >
> > > Thanks.
>
