I have numeric data.I'd like to group the data.It is easy to use 'Kmeans' command, but it cannot be necessarily k=3.I want to set k automatically.
Or Is there any other good idea to group?
kmeans has an option for setting a range of ks to attempt.
| kmeans k=3-12
Just feed it different data a few times and see what it does for you. Here's the reference.