Skip to content
GitLab
Projects Groups Snippets
  • /
  • Help
    • Help
    • Support
    • Community forum
    • Submit feedback
    • Contribute to GitLab
  • Sign in
  • A armawrap
  • Project information
    • Project information
    • Activity
    • Labels
    • Members
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributors
    • Graph
    • Compare
  • Issues 0
    • Issues 0
    • List
    • Boards
    • Service Desk
    • Milestones
  • Merge requests 0
    • Merge requests 0
  • Deployments
    • Deployments
    • Releases
  • Monitor
    • Monitor
    • Incidents
  • Analytics
    • Analytics
    • Value stream
    • Repository
  • Activity
  • Graph
  • Create a new issue
  • Commits
  • Issue Boards
Collapse sidebar
  • FSLFSL
  • armawrap
  • Issues
  • #1
Closed
Open
Issue created Dec 16, 2020 by Matthew Webster@mwebsterOwner

Bug in main SVD wrapper

The line

https://git.fmrib.ox.ac.uk/fsl/armawrap/-/blob/master/armawrap/function_svd.hpp#L25

resizes U to the dimensions of At ( the input Data ). However this will crash for over-complete matrices ( e.g. 12x10 ). This line may have originally been introduced when the function used arma:svd for compatibility with the newmat interface. However svd_econ appears to always output a U of the expected size, including for the case of an over-complete input ( 12x10 -> 10x10 ). One fix would be to make the first limit min(At.n_rows-1, At.n_cols-1) but it may be simpler to just drop the submatrix call - perhaps avoid the copy altogether and pass U straight in, setting U=0 if !withU similar to V.

To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information
Assignee
Assign to
Time tracking