[Octave-bug-tracker] [bug #31731] Test failed for sqp

[Marco Caliari]

Follow-up Comment #4, bug #31731 (project octave):

Dear Rik,

I got my result with 


tol = eps


If I run the test with the standard tolerance sqrt (eps), I get


norm(x-x_opt,inf)
ans =  9.87475148317429e-08


which is greater than


5*sqrt(eps)


So, nothing strange here. Can you please run


[x, obj, info, iter, nf, lambda] = sqp (x0, @phi, @g,
[],-realmax,realmax,100,eps);


and compare the result with x_opt?

    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?31731>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/