TODO

   1 1. Add unit testing for crazier things like random invertible matrices.
   2
   3 2. Add real docstrings everywhere.
   4
   5 3. Try to eliminate the code in matrices.py.
   6
   7 4. Make it work on a cartesian product of cones in the correct order.
   8
   9 5. Make it work on a cartesian product of cones in the wrong order
  10    (apply a perm utation before/after).
  11
  12 6. Rename all of my variables so that they don't conflict with CVXOPT.
  13    Maybe x -> xi and y -> gamma in my paper, if that works out.
  14
  15 7. Make sure we have the dimensions of the PSD cone correct.
  16
  17 8. Use a positive tolerance when comparing floating point numbers.
  18
  19 9. Come up with a fast heuristic (like making nu huge and taking e1 as
  20    our point) that finds a primal feasible point.
  21
  22 10. Ensure that we only compute eigenvalues of *symmetric* matrices
  23     (so that the eigenvalues are real).