GLA
Start with an initial codebook c1, c2, ... cn and training set X.
Iterate:
- Partition X into X1, X2, ..., Xn where Xi includes the members of X that are closest to ci.
- Let xi be the centroid of Xi .
-
-
- if is sufficiently small then quit.otherwise continue the interation with the new codebook c1, c2, ... cn = x1, x2, ... xn .