Tue May 20 22:34:08 2025

svd_truncated_test():
  python version: 3.10.12
  numpy version:  1.26.4
  Demonstrate the use of the truncated
  Singular Value Decomposition (SVD) for cases where
  the sizes of M and N are very different.

svd_truncated_u_test():
  M = 4
  N = 3

  Original matrix A:
[[0.15333306 0.53370181 0.2489525 ]
 [0.6978926  0.33925746 0.68306671]
 [0.78401704 0.86739345 0.71723311]
 [0.74840143 0.48749487 0.96011544]]

  Maximum error |A - U*S*V| =  4.163336342344337e-16

  Recomputed A = U * S * V:
[[0.15333306 0.53370181 0.2489525 ]
 [0.6978926  0.33925746 0.68306671]
 [0.78401704 0.86739345 0.71723311]
 [0.74840143 0.48749487 0.96011544]]

svd_truncated_v_test():
  M = 3
  N = 4

  Original matrix A:
[[0.8200497  0.73700845 0.61438166 0.73738404]
 [0.63573015 0.57336277 0.14576005 0.56786714]
 [0.2106874  0.85797394 0.00794206 0.40436866]]

  Maximum error |A - U*S*V| =  5.551115123125783e-16

  Recomputed A = U * S * V:
[[0.8200497  0.73700845 0.61438166 0.73738404]
 [0.63573015 0.57336277 0.14576005 0.56786714]
 [0.2106874  0.85797394 0.00794206 0.40436866]]

svd_truncated_test():
  Normal end of execution.
Tue May 20 22:34:08 2025