Hi,
I have a dataframe looking like:
using DataFrames
df = DataFrame(a=1:4, b=5:8)
4×2 DataFrame
Row │ a b
│ Int64 Int64
─────┼──────────────
1 │ 1 5
2 │ 2 6
3 │ 3 7
4 │ 4 8
From that, I would like to create a vector/array like:
[1, 5, 2, 6, 3, 7, 4, 8]
Tried many hcat, vcat, values etc. combinations without success…
julia> v = Int[]
Int64[]
julia> for i in eachindex(df.a, df.b)
push!(v, df.a[i])
push!(v, df.b[i])
end
julia> v
8-element Vector{Int64}:
1
5
2
6
3
7
4
8
I’m not sure about performance, but here’s a one-liner:
df |> Matrix |> vec
and if you need the specific ordering of the elements:
vec(Matrix(df)')
# or df |> Matrix |> transpose |> vec
Thanks everyone !
julia> vcat(df.a, df.b)
8-element Vector{Int64}:
1
2
3
4
5
6
7
8
Ahem! Didn’t read the Q carefully enough!
other ways
vcat(eachcol(permutedims(df))...)
stack(permutedims(df),:).value
r,c=size(df)
vc=Vector{typeof(df[1,1])}(undef,r*c)
idx=0
for i in 1:r, j in 1:c
vc[idx+=1]=df[i,j]
end
vc
Matrix(df)'[:]
julia> df |> eachrow |> Iterators.flatten |> collect
8-element Vector{Int64}:
1
5
2
6
3
7
4
8