Git Product home page Git Product logo

Comments (8)

pelesh avatar pelesh commented on July 18, 2024

I can confirm this issue existed before #65. It seems that it went undetected because floating point zero was not exactly zero.

from hiop.

cnpetra avatar cnpetra commented on July 18, 2024

Can you please run the same example with these two options (hiop.options file):

verbosity_level 5
time_kkt on
compute_mode hybrid

and then

verbosity_level 5
time_kkt on
compute_mode cpu

from hiop.

pelesh avatar pelesh commented on July 18, 2024

Hybrid mode:

...

NLP resid [update]: inf norm rszu=5.00623181190249064e-06
NLP resid [update]: inf norm rsvl=2.76028290659870840e-06
NLP resid [update]: inf norm rsvu=2.15816015869067584e-04
nrmOneDualEqu 1092.91   nrmOneDualBo 777.895
nlpoverall 0.0466209  nloptim 0.00202824  sd 1  nlpfeas 0.0466209  nlpcomplem 0.00036624  sc 1
  Nlp    errs: pr-infeas:4.66209154235927986e-02   dual-infeas:2.02824197585146871e-03  comp:3.66240139592523403e-04  overall:4.66209154235927986e-02
  LogBar errs: pr-infeas:4.66209154235927986e-02   dual-infeas:2.02824348009270593e-03  comp:2.15816015869067584e-04  overall:4.66209154235927986e-02
  85 -1.3560463e+03 4.662e-02  2.028e-03  -3.82  1.000e+00  5.551e-17  54(?)
Couldn't solve the problem.
Linesearch returned unsuccessfully (small step). Probable cause: inaccurate gradients/Jacobians or infeasible problem.
Total time 5.514 sec 
Hiop internal time:     total 5.490 sec     avg iter 0.065 sec 
    internal total std dev across ranks 0.000 percent
Fcn/deriv time:     total=0.003 sec  ( obj=0.000 grad=0.000 cons=0.001 Jac=0.001 Hess=0.000) 
    Fcn/deriv total std dev across ranks 0.000 percent
Fcn/deriv #: obj 172 grad 86 eq cons 173 ineq cons 173 eq Jac 86 ineq Jac 86
Total KKT time 5.471 sec 
	update init 0.001sec     update linsys 0.063 sec     fact 5.387 sec 
	solve rhs-manip 0.007 sec     triangular solve 0.005 sec 

solve4 trouble: returned -4 (with objective is -1.356046289760e+03)

CPU mode:

...

NLP resid [update]: inf norm rszu=8.82172655451939644e-12
NLP resid [update]: inf norm rsvl=1.22717251926128643e-12
NLP resid [update]: inf norm rsvu=7.24712484747036049e-11
nrmOneDualEqu 1094.5   nrmOneDualBo 778.5
nlpoverall 2.57837e-09  nloptim 4.98946e-14  sd 1  nlpfeas 2.22461e-14  nlpcomplem 2.57837e-09  sc 1
  Nlp    errs: pr-infeas:2.22460938559265742e-14   dual-infeas:4.98946459851671800e-14  comp:2.57837480815476538e-09  overall:2.57837480815476538e-09
  LogBar errs: pr-infeas:2.22460938559265742e-14   dual-infeas:4.98946459851671800e-14  comp:7.24712484747036049e-11  overall:7.24712484747036049e-11
  85 -1.3565000e+03 2.225e-14  4.989e-14  -8.60  1.000e+00  1.000e+00  1(f)
Successfull termination.
Total time 0.194 sec 
Hiop internal time:     total 0.182 sec     avg iter 0.002 sec 
    internal total std dev across ranks 0.000 percent
Fcn/deriv time:     total=0.002 sec  ( obj=0.000 grad=0.000 cons=0.001 Jac=0.001 Hess=0.000) 
    Fcn/deriv total std dev across ranks 0.000 percent
Fcn/deriv #: obj 87 grad 86 eq cons 88 ineq cons 88 eq Jac 86 ineq Jac 86
Total KKT time 0.167 sec 
	update init 0.004sec     update linsys 0.040 sec     fact 0.108 sec 
	solve rhs-manip 0.006 sec     triangular solve 0.004 sec 

Optimal objective 1:  -1.24881064528112e+01. Solver status: 0
Optimal objective 2:  -1.24881064633628e+01. Solver status: 0
Optimal objective 3:  -1.24099999964871e+03. Solver status: 0
Optimal objective 4:  -1.35649999989221e+03. Solver status: 0

from hiop.

ashermancinelli avatar ashermancinelli commented on July 18, 2024

Also reporting failures for this test on ORNL Ascent system.

from hiop.

pelesh avatar pelesh commented on July 18, 2024

Also reporting failures for this test on ORNL Ascent system.

In which mode?

from hiop.

cnpetra avatar cnpetra commented on July 18, 2024

this should be fixed now, please check against the master

from hiop.

pelesh avatar pelesh commented on July 18, 2024

I am not sure if this is fixed. All but one test in Example 5 are disabled. The remaining test passes.

from hiop.

cnpetra avatar cnpetra commented on July 18, 2024

the others tests are not ran part of make test anymore because they give too many false negatives. they're there to serve other purposes...

from hiop.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.