Functional Principal Least Squares

Partial Least Squares using SVD

moduleauthor:: J. Derek Tucker <jdtuck@sandia.gov>

fPLS.pls_svd(time, qf, qg, no, alpha=0.0)[source]

This function computes the partial least squares using SVD

Parameters:
  • time – vector describing time samples
  • qf – numpy ndarray of shape (M,N) of N functions with M samples
  • qg – numpy ndarray of shape (M,N) of N functions with M samples
  • no – number of components
  • alpha – amount of smoothing (Default = 0.0 i.e., none)
Return type:

numpy ndarray

Return wqf:

f weight function

Return wqg:

g weight function

Return alpha:

smoothing value

Return values:

singular values