What would be the function to create this new table ?

Hello there, my question is the following :

I have a table (let's say 20x2). I have duplicates on the first column but not on the second (for example the first 10 lines are x (each of these x have a their own value on the second column) and the 10 last lines are y (same logic).

With this function : Table2_grouped = varfun(@sum,Table2_bis,'InputVariables','second column','GroupingVariables','first column') I could regroup and cancel the duplicates and sum their own values. (I now have a 2x2 tables with x and y and their sums)

But here's my issue : i want now to create a new table where i once again have duplicates (so the 20 lines) but i want to keep the new value (meaning i want to keep to sum of the ten x and ten y). Hopefully this is clear and you can help me. thanks a lot

Answer 1

[KL]

Do you want to repeat the new value ten times on the second column of the new table?

sumrepArray = [repmat(sumTable{1,2},10,1); repmat(sumTable{2,2},10,1)] 
newTable = table(oldTable{:,1},sumrepArray);

but I'd recommend getting the size of each repeating elements from the oldtable and use this count to create the sumrep array.

[Loïc Boga]

Maybe my example was not the best one indeed. My original table has more than 2 duplicated variables which have their own values. Once i grouped these variables and get the sum I just want to go back to my original table and have as a second column the sum corresponding to the variable (second column will contain duplicates obviously). So yes to answer your question, i would like to repeat ten times the new value (so i can work on it). Could you please give me more details on " * but I'd recommend getting the size of each repeating elements from the oldtable and use this count to create the sumrep array."* thanks a lot

[KL]

Aha, it's not so hard but I cannot test my code at the moment. Try something like the following (k is the 1 to number of repeating variables),

 oldTable{oldTable{:,1}==sumTable{k,1}} = sumTable{k,2};

[Loïc Boga]

thanks for the answers ! :)

[KL]

You're welcome. If it solved your problem, you could also accept the answer:)