Combine grouped DF in Julia with Floats and Strings

Question

I have a bunch of Grouped DataFrames gdf that I want to combine. I want to combine the GDF with the mean var1 which is a Float and the first element of var2 which is a String.

I tried

combine(gdf, :var1 .=> mean, :var2 .=> first(:var2))

But getting the error ERROR: MethodError: no method matching iterate(::Symbol) I also tried first(:var2, 1) .

Thanks for any help.

Answer 1

This is the way to do it with DataFrames.jl:

julia> using DataFrames

julia> using Statistics

julia> df = DataFrame(id=[1,2,1,2,1,2], var1=1.5:1:6.5, var2=string.(1:6))
6×3 DataFrame
 Row │ id     var1     var2
     │ Int64  Float64  String
─────┼────────────────────────
   1 │     1      1.5  1
   2 │     2      2.5  2
   3 │     1      3.5  3
   4 │     2      4.5  4
   5 │     1      5.5  5
   6 │     2      6.5  6

julia> gdf = groupby(df, :id)
GroupedDataFrame with 2 groups based on key: id
First Group (3 rows): id = 1
 Row │ id     var1     var2
     │ Int64  Float64  String
─────┼────────────────────────
   1 │     1      1.5  1
   2 │     1      3.5  3
   3 │     1      5.5  5
⋮
Last Group (3 rows): id = 2
 Row │ id     var1     var2
     │ Int64  Float64  String
─────┼────────────────────────
   1 │     2      2.5  2
   2 │     2      4.5  4
   3 │     2      6.5  6

julia> combine(gdf, :var1 => mean, :var2 => first)
2×3 DataFrame
 Row │ id     var1_mean  var2_first
     │ Int64  Float64    String
─────┼──────────────────────────────
   1 │     1        3.5  1
   2 │     2        4.5  2

(there is no need of . before => and no need to pass argument to first explicitly)

If you would prefer to use assignment style (instead of functional style with => pairs) use DataFramesMeta.jl:

julia> using DataFramesMeta

julia> @combine(gdf, :var1_mean=mean(:var1), :var2_first=first(:var2))
2×3 DataFrame
 Row │ id     var1_mean  var2_first
     │ Int64  Float64    String
─────┼──────────────────────────────
   1 │     1        3.5  1
   2 │     2        4.5  2