Skip to content

Latest commit

 

History

History
12 lines (9 loc) · 862 Bytes

File metadata and controls

12 lines (9 loc) · 862 Bytes

McCormick_LinAlg: Linear Algebra For McCormick Relaxations

McCormick_LinAlg is an auxillary library of linear algebra functions designed for the global optimization solver EAGO (Easy Advanced Global Optimization). For more information about EAGO, click here.

Features

  • BLAS styled linear algebra routines meant to handle the McCormick relaxation data type found in EAGO.
  • Test files to check correctness and efficiency of each function's implementations
  • Estimation methods for the determinant of an interval matrix. Still in progress

Status of Package

  • Majority of methods in the BLAS library up to TRMV are running.
  • Preconditioned Gaussian Elimination seems to be the most functional and accurate Interval Determinant Estimation method