Skip to content

Fix tests that do not converge #881

@cmbenne3

Description

@cmbenne3

Description

We need to add in checks to all the benchmark tests to ensure that the problem converges successfully before comparing expected values.

self.assertTrue(prob.result.success)

Example

The following tests do not currently give prob.result.success = True:

test_bench_FwFm.py:TestBenchFwFmSerial.test_bench_FwFm_IPOPT
test_bench_FwFm.py:TestBenchFwFmSerial.test_bench_FwFm_SNOPT exit code 3 requested accuracy not achieved
test_off_design_FwFm.py:TestOffDesign.test_off_design_IPOPT
test_off_design_FwFm.py:TestOffDesign.test_off_design_SNOPT
test_off_design_GwGm.py:TestOffDesign.test_off_design_IPOPT
test_off_design_GwGm.py:TestOffDesign.test_off_design_SNOPT
test_bench_FwGm.py:* (close, but doesn't get a 0,1) [added by KTM]
test_bench_GwFm.py:ProblemPhaseTestCase.bench_test_swap_1_GwFm_IPOPT (doesn't run under testflo runs in PR).
test_bench_GwFm.py:ProblemPhaseTestCase.bench_test_swap_1_GwFm_SNOPT (doesn't run in testflo or PR, runs in merge queue).
test_FLOPS_detailed_landing.py:TestFLOPSDetailedLanding.bench_test_IPOPT
test_FLOPS_based_sizing_N3CC.py:ProblemPhaseTestCase.bench_test_sizing_N3CC exit code 41 current point cannot be improved

Aviary Version

0.10.0-dev

Relevant environment information

No response

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't working

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions