Parse noSolution status from CPLEXShell and CPLEXDirect solvers correctly
#1313
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Codecov Report
@@ Coverage Diff @@
## master #1313 +/- ##
==========================================
+ Coverage 71.57% 72.75% +1.17%
==========================================
Files 547 491 -56
Lines 83583 79087 -4496
==========================================
- Hits 59821 57536 -2285
+ Misses 23762 21551 -2211
Continue to review full report at Codecov.
|
ruaridhw
force-pushed
the
fix/cplex_shell_logfile_1210
branch
from
8b61118 to
7e8d86c
Compare
ruaridhw
force-pushed
the
fix/cplex_shell_logfile_1210
branch
from
7e8d86c to
12cdacb
Compare
|
@michaelbynum, sorry about that. Branch is fixed now. |
- It looks as though IBM have updated the schema that the shell logfile uses and therefore the parsing is not quite done correctly
- Notably, 'CPLEX Error; no integer solution' is being flagged and returned as ('ok', 'unknown') for the status and termination condition
- The Direct solver should return the same solver and solution status as the Shell solver when no solution is found - These error codes are currently unhandled and just return `TerminationCondition.error` - It seems reasonable to assume that if a solve reaches max time limit and is infeasible that we cannot yet determine whether it is actually infeasible and therefore makes sense to return `noSolution`.
- I've elected not to add this to the return Bunch from `_apply_solver()` because that would change the behaviour of the downstream code (which raises an ApplicationError)
- Do not adjust `SolverInformation` since this is called `return_code` here
- Reaching these limits could also be reasons for having not found a solution
…ation - If we have already determined a reasonable termination condition that would have otherwise returned an `ok` SolverStatus, return `warning` instead of `error`
- In the case of `noSolution`, CPLEXSHELL now considers this a `warning` solver status - Similarly, `GAMSDirect` returns a `SolutionStatus.unknown` so we should be consistent with that
- Constraint `C` is being overwritten
- This is a much harder model that is more likely to not be solved within the allowed time
ruaridhw
force-pushed
the
fix/cplex_shell_logfile_1210
branch
from
12cdacb to
33b0b2d
Compare
|
@michaelbynum, this should be ready to review whenever you get a chance |
|
@ruaridhw Thanks. I will review these PRs early next week. |
|
This PR is going to be superseded by work to complete #1030 - which builds in base support for "no solution" to all solver interfaces. As such, I am going to close this PR. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes
Summary/Motivation:
Currently when CPLEX returns "No Solution", this isn't captured in any of the Pyomo solution or solver data objects. We can identify "No Solution" because when both of (1) max time limit is reached and (2) the model is infeasible occur.
Changes proposed in this PR:
CPLEXDirect_postsolve()to use the actual CPLEX library constantsMIP_time_limit_infeasibleandMIP_dettime_limit_infeasibleas indicating a solvertermination_conditionofnoSolutionCplexSolverErrors (but don't add this to the returnedBunchdue to downstream code raisingApplicationErrors)results.solver.return_codeandresults.solver.termination_messageusing the available informationCPLEXSHELLprocess_logfile()for CPLEX 12.10 as it looks like IBM have updated the log file schema and is no longer being parsed correctly'no integer solution.'in the log file as indicating a solvertermination_conditionofnoSolutionCPLEX Errorlines in the log fileLegal Acknowledgement
By contributing to this software project, I agree to the following terms and conditions for my contribution: