This will run the K means algorithm with the given number of clusters (K). It will output the centers of the clusters and will also use the Freq column to assign a cluster number to each row of the stats table. The initial values of the Freq column are NOT used so you have to actually type in multiple times any point with multiplicity instead of using the Freq column.
(It's a bit big so if you encounter any errors, please point them out, maybe I've mistyped something.)
FreqOff n→N FreqOn 0→L "K="?K 0→S 0→M For 1→I To T RanInt#(1,K)→List Freq[I] Next Do S→L M+1→M 0→DimZ 3K+3→DimZ For 1→I To T List Freq[I]→J Z[3J]+List X[I]→Z[3J] Z[3J+1]+List Y[I]→Z[3J+1] Z[3J+2]+1→Z[3J+2] Next For 1→I To K If Z[3I+2]=0:Then List X[RanInt#(1,T)]→Z[3I] List X[RanInt#(1,T)]→Z[3I+1] Else Z[3I]÷Z[3I+2]→Z[3I] Z[3I+1]÷Z[3I+2]→Z[3I+1] IfEnd Next 0→S For 1→I To T 1×10^99→B For 1→J To K √((List X[I]-Z[3J])²+(List Y[I]-Z[3J+1])²)→D If D<B:Then D→B J→W IfEnd Next W→List Freq[I] S+B→S Next Cls Locate 1,1,"S=" Locate 3,1,S÷T Locate 1,2,"ITERATION " Locate 11,2,M LpWhile S≠L Locate 1,3,"CONVERGED."◣ For 1→I To K Cls Locate 1,1,"K=" Locate 3,1,I Locate 1,2,"X=" Locate 3,2,Z[3I] Locate 1,3,"Y=" Locate 3,3,Z[3I+1]◣ Next Cls "DONE."