-
Notifications
You must be signed in to change notification settings - Fork 135
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Fix defect ensemble model (with Code) and genetic algorithm #2317
Conversation
.../GeneticAlgorithms/continuous/unconstrained/test_ensembleModel_withGA_Code_and_Functions.xml
Outdated
Show resolved
Hide resolved
…trained/test_ensembleModel_withGA_Code_and_Functions.xml
…ven into alfoa/genetic_ensemble_model
...lgorithms/continuous/unconstrained/metaModelWithCodeAndFunctionsAndGenetic/decayConstantB.py
Outdated
Show resolved
Hide resolved
…trained/metaModelWithCodeAndFunctionsAndGenetic/decayConstantB.py
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Just one question, changes are good to me.
optimize | ||
</Sequence> | ||
<WorkingDir>metaModelWithCodeAndFunctionsAndGenetic</WorkingDir> | ||
<batchSize>1</batchSize> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Have you tested it with different batch size?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
ops. Not yet. let me test it
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@wangcj05 I fixed the parallel execution and converted the test into a parallel (batchSize > 1) to test it.
Ready for re-review
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The changes are good. @alfoa Could you document how you define job '' prefix'' and 'uniqueHandler' inside Ensemble? This will help for future rework on the job submission function.
type = 'RavenFramework' | ||
input = 'continuous/unconstrained/test_ensembleModel_withGA_Code_and_Functions.xml' | ||
[./csv] | ||
type = UnorderedCsv |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@alfoa FYI, you need to change 'UnorderedCsv' to 'UnorderedCSV'. There is an inconsistent naming in the rook test system.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done and added an error msg in ROOK (to avoid a blind crash)
I can add the info in the Issue I created? |
Job Test Ubuntu 20-2 Optional on a522493 : invalidated by @wangcj05 failed at fetch |
Job Test qsubs sawtooth on a522493 : invalidated by @alfoa Set python environment taking 6+ hrs? |
@wangcj05 There is a problem with HPC testing (set up enviroment) that is not related to this PR. (It does not seem to be a spurious problem since I invalidated the job multiple times and it gets stuck at the same step) |
@wangcj05 @joshua-cogliati-inl It seems that now in devel/main there is a cascade of failures (https://civet.inl.gov/branch/2903/) after merge of PR #2309 (I dnk if it is related...probably it is not) |
@wangcj05 now it fails because |
scripts/library_handler.py
Outdated
@@ -54,7 +54,7 @@ | |||
# some bad actors can't use the metadata correctly | |||
# and so need special treatment | |||
# -> see findLibAndVersion | |||
metaExceptions = ['pyside2', 'AMSC', 'PIL'] | |||
metaExceptions = ['pyside2', 'AMSC', 'PIL', 'cvxpy'] |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
'cvxpy' is added via FARM plugin. It cannot be checked with the metadata. I followed the approach in the library_handler but this should be revised since it is not easily maintainable (overall for plugins). @mandd @wangcj05 @joshua-cogliati-inl . I wonder why the regression system (for PR) did not catch this.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It seems this is due to an update of the library in conda-forge 2 days ago (see https://anaconda.org/conda-forge/cvxpy-base/files).
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
'cvxpy' is added via FARM plugin. It cannot be checked with the metadata. I followed the approach in the library_handler but this should be revised since it is not easily maintainable (overall for plugins). @mandd @wangcj05 @joshua-cogliati-inl . I wonder why the regression system (for PR) did not catch this.
I changed the approach..if a library through a metadata is not found, we try the "slow approach".
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
skip_check
is another way to fix this. Example: <hdf5 skip_check='True'/>
I have no objection to fixing it in metaExceptions however.
Job Mingw Test on 7ff7659 : invalidated by @alfoa |
@wangcj05 the |
Job Mingw Test on 7ff7659 : invalidated by @alfoa |
How many tests are failing? (And if only a few, which ones?) Thanks. |
FYI, there are around 33 tests failed in previous MingW test. Many for optimizations, such as simulated annealing. I have checked the new tests right now, it seems the runs are ok for now. Let's see if the test on Mingw can pass or not. @joshua-cogliati-inl |
@wangcj05 @joshua-cogliati-inl unfortunately they are still failing...now in the TSA, PostProcessors, Datamining etc. |
@joshua-cogliati-inl Let's try to test it on devel first. Can you pull the changes for cvxpy and test it on your trivial white space branch? |
Hm, which set of changes for cvxpy? d3c7069 ? |
|
|
Thanks: #2113 |
Job Mingw Test on 7ff7659 : invalidated by @alfoa |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
changes are good.
Changes are good, PR checklist is good. PR can be merged. |
Pull Request Description
What issue does this change request address? (Use "#" before the issue to link it, i.e., #42.)
Closes #2304
What are the significant changes in functionality due to this change request?
The EnsembleModel (with code) uses a different parallelization strategy. The batching mode has been enabled in that strategy (parallelMode ==2)
I also created an issue #2318 to better re-design the batching system
For Change Control Board: Change Request Review
The following review must be completed by an authorized member of the Change Control Board.
<internalParallel>
to True.raven/tests/framework/user_guide
andraven/docs/workshop
) have been changed, the associated documentation must be reviewed and assured the text matches the example.