How to group a matrix based on one column?

19 次查看(过去 30 天)
Hi,
I have a 64600x59 matrix data. I need unique values and also non-unique values grouped. Usually i use unique() function and then check if my rows matches with unique value rows, then i group by this using for loop. But this time my data is very big and it takes too much time.
I am curious about if there is any shortcut way or a function in matlab for me to group every same value in a different cell or table. This way i will be seeing non-unique values as grouped.
Do you have any ideas or suggestions?
Thank you
  3 个评论
Sevgi
Sevgi 2023-5-9
编辑:Sevgi 2023-5-9
Thank you for this suggestion. I study on it and the code work for now.
T=readtable("a.xslx");
[percombine, combinenames]= findgroups(T.Combine);
meanfirstaxle= splitapply(@mean,T.("1. Axle"),percombine);
I can take mean of each groups too. But I still can't see what are group members. Is there a way for it?
Peter Perkins
Peter Perkins 2023-6-5
There are any number of ways to do grouped calculations, includng groupsummary/transform/filter/counts, varfun/rowfun, and (as Stephen123 says) findgroups/splitapply.
But first you need to show exactly what you have and exactly what you want.

请先登录,再进行评论。

回答(1 个)

Matt J
Matt J 2023-5-9
编辑:Matt J 2023-5-9
Usually i use unique() function and then check if my rows matches with unique value rows, then i group by this using for loop.
If using unique(), the 3rd output argument gives the group label. There is no need for a for-loop.
[values, ~, groupLabel]=unique(data);

类别

Help CenterFile Exchange 中查找有关 Creating and Concatenating Matrices 的更多信息

产品


版本

R2021b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by