Deleting rows of matrix based on duplicated values of a column in Julia

Question

Here is another challenge in Julia. Imagine when have the following matrix:

 5.0  3.54924
 6.0  3.54702
 6.0  3.54655
 7.0  3.54168
 7.0  3.0

I want to delete the rows of the matrix for which the elements of the first column are duplicated. That will produce the following matrix, for example:

 5.0  3.54924
 6.0  3.54702
 7.0  3.0

Deletion of rows are arbitrary since we don't keep repeated values for the first column. How can I achieve that?

Answer 1

Also you can simply do it using DataFrames.jl (the performance will be worse as you perform the conversion twice, but the code is simpler):

julia> A = [5.0  3.54924
                   6.0  3.54702
                   6.0  3.54655
                   7.0  3.54168
                   7.0  3.0]
5×2 Array{Float64,2}:
 5.0  3.54924
 6.0  3.54702
 6.0  3.54655
 7.0  3.54168
 7.0  3.0

julia> Matrix(unique(DataFrame(A), 1))
3×2 Array{Float64,2}:
 5.0  3.54924
 6.0  3.54702
 7.0  3.54168

Alternatively you could write

julia> A[.!nonunique(DataFrame(A[:,1:1])),:]
3×2 Array{Float64,2}:
 5.0  3.54924
 6.0  3.54702
 7.0  3.54168

which is a bit faster and uses less memory but more messy.