[
https://issues.apache.org/jira/browse/MATH-439?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12933268#action_12933268
]
Gilles commented on MATH-439:
-----------------------------
And another regression (maybe).
{noformat}
Tests run: 12, Failures: 1, Errors: 0, Skipped: 0, Time elapsed: 0.271 sec <<<
FAILURE!
testCircleFitting(org.apache.commons.math.optimization.general.NonLinearConjugateGradientOptimizerTest)
Time elapsed: 0.012 sec <<< FAILURE!
java.lang.AssertionError: expected:<69.960161753> but was:<5.9097119121754055E9>
at org.junit.Assert.fail(Assert.java:74)
at org.junit.Assert.failNotEquals(Assert.java:448)
at org.junit.Assert.assertEquals(Assert.java:315)
at org.junit.Assert.assertEquals(Assert.java:346)
at
org.apache.commons.math.optimization.general.NonLinearConjugateGradientOptimizerTest.testCircleFitting(NonLinearConjugateGradientOptimizerTest.java:355)
[...]
{noformat}
The problem is at those lines:
{code}
NonLinearConjugateGradientOptimizer optimizer =
new
NonLinearConjugateGradientOptimizer(ConjugateGradientFormula.POLAK_RIBIERE);
optimizer.setMaxEvaluations(100);
optimizer.setConvergenceChecker(new SimpleScalarValueChecker(1.0e-30,
1.0e-30)); // <--- here
UnivariateRealSolver solver = new BrentSolver(1e-15, 1e-13);
optimizer.setLineSearchSolver(solver);
RealPointValuePair optimum =
optimizer.optimize(circle, GoalType.MINIMIZE, new double[] { 98.680, 47.345
});
{code}
Now, when, in the above, the line marked "<--- here" is commented out, the test
passes.
However, if the line is left in but another solver is used (I tried
{{BisectionSolver}} and {{SecantSolver}}), the test also passes!
Does someone have an idea of what could be going on?
> Refactoring of solvers (package "analysis.solvers")
> ---------------------------------------------------
>
> Key: MATH-439
> URL: https://issues.apache.org/jira/browse/MATH-439
> Project: Commons Math
> Issue Type: Improvement
> Reporter: Gilles
> Priority: Minor
> Fix For: 3.0
>
> Attachments: AbstractUnivariateRealSolver.java
>
>
> The classes in package "analysis.solvers" could be refactored similarly to
> what was done for package {{optimization}}.
> * Replace {{MaxIterationsExceededException}} with
> {{TooManyEvaluationsException}}:
> Apart from the class {{MaxIterationsExceededException}} being deprecated,
> this approach makes it difficult to compare different algorithms: While the
> concept of iteration is algorithm-dependent, the user is probably mostly
> interested in the number of function evaluations.
> * Implement the method {{solve}} in the base class
> ({{UnivariateRealSolverImpl}}) and define an abstract method {{doSolve}} to
> be implemented in derived classes. This method would then use a new
> {{computeObjectiveFunction}} method that will take care of the counting of
> the function evaluations.
> * Remove "protected" fields (the root is unnecessary since it is returned by
> {{solve}}). Arguingly the function value is also not very useful (as we know
> what it should be), except for debugging purposes (in which case, it might
> not be a problem to call the function's {{value}} method once more).
> * Remove the tolerance setter (accuracy) and make the corresponding fields
> "final".
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
