ShiftScaleTransformer Scaling Options "maxnorm" and "maxnormsym"

[nicolearetz]

option "maxnorm" for transformers.
Snapshots are transformed such that the maximum Euclidean norm of the transformed snapshots (with or without centering) is at most 1. This is useful because the norm of the Kronecker product of two vectors is equal to the product of the two vectors' norms: $|x \otimes y| = |x| |y|$. Scaling the snapshots to have maximum norm 1 guarantees that the entries of the OpInf data matrix remain bounded even for high-dimensional polynomial operators.