You can do B[E[:],:] or better write E as a vector (rather than a 1x10
array) directly:
E = [5, 5, 5, 4, 5, 5, 5, 5, 1, 5]
Then B[E,:] works as expected.
Den tisdagen den 10:e juni 2014 kl. 15:40:50 UTC+2 skrev Bruno Rodrigues:
>
> Hi, I posted a first thread but can't find it anywhere, so I probably
> messed up somewhere. So here is my problem: let's say we have following
> arrays:
>
> B = [17 24 1 8 15; 23 5 7 14 16;4 6 13 20 22; 10 12 19 21 3;11 18 25 2 9]
>
> E =[5 5 5 4 5 5 5 5 1 5]
>
> If we do this:
>
> B[E]
>
> we get:
>
> B[E]
> 1x10 Array{Int64,2}:
> 11 11 11 10 11 11 11 11 17 11
>
> No problems. But how would I do that for all columns of B? In matlab, I'd
> do this: B(E,:) and get:
>
> ans =
>
> 11 18 25 2 9
> 11 18 25 2 9
> 11 18 25 2 9
> 10 12 19 21 3
> 11 18 25 2 9
> 11 18 25 2 9
> 11 18 25 2 9
> 11 18 25 2 9
> 17 24 1 8 15
> 11 18 25 2 9
>
> In Julia I must do a loop:
>
> S = zeros(10,5)
> x = [1 2 3 4 5]
>
> for i in x
> S[:,i] = B[:,i][E[end,:]]
> end
>
>
> Is there a way to avoid this loop and obtain something like in Matlab
> using simple indexing notation?
>
