Seite 1 von 1

Assignment 3 Problem 1, warning using quadprog

Verfasst: 26. Jun 2013 13:43
von fy95olok
Hi,
using quadprog in our Program produces:

Code: Alles auswählen

Warning: Large-scale algorithm does not currently solve this problem formulation,
using medium-scale algorithm instead. 
  In quadprog at 291
  In train_svm_s at 20
  In problem1a at 6
Exiting: the constraints are overly stringent;
 no feasible starting point found.
Does anyone know what might cause this problem?

Re: Assignment 3 Problem 1, warning using quadprog

Verfasst: 26. Jun 2013 14:43
von lustiz
fy95olok hat geschrieben:Hi,
using quadprog in our Program produces:

Code: Alles auswählen

Warning: Large-scale algorithm does not currently solve this problem formulation,
using medium-scale algorithm instead. 
  In quadprog at 291
  In train_svm_s at 20
  In problem1a at 6
Exiting: the constraints are overly stringent;
 no feasible starting point found.
Does anyone know what might cause this problem?

Hi, it means the quadprog solver is not able to find a feasible solution because your constraints cannot be satisfied. It's probably a mistake in either A or b or both!

Re: Assignment 3 Problem 1, warning using quadprog

Verfasst: 28. Jun 2013 14:17
von hymGo
I get a warning, too. Anyway, the displayed results seems to be correct.

Code: Alles auswählen

Warning: Trust-region-reflective algorithm does not solve this type of problem, using active-set
algorithm. For more help, see Choosing the Algorithm in the documentation. 
> In quadprog at 371
  In train_svm_s at 37
  In problem1a at 6 
Warning: Your current settings will run a different algorithm (interior-point-convex) in a
future release. 
> In quadprog at 375
  In train_svm_s at 37
  In problem1a at 6 

Re: Assignment 3 Problem 1, warning using quadprog

Verfasst: 28. Jun 2013 18:40
von lustiz
You can get rid of the warning by calling quadprog with:

Code: Alles auswählen

options = optimset('quadprog');
options.Algorithm = 'interior-point-convex';
quadprog(......, options);