/**
* Create the ECDF attribute
*
* @param xAxisTransformation the transformation to be applied to the {@code x}-axis
* @param yAxisInputTransformation the transformation to be applied to the data of the {@code y}
* -axis before being fed to the actual computation
* @param yAxisOutputTransformation the transformation of the result of the function applied to
* the data on the {@code y}-axis.
* @param goalValue the goal value
* @param criterion the goal comparison criterion
* @param aggregate the method to aggregate the ECDFs
*/
public ECDF(
final DimensionTransformation xAxisTransformation,
final DimensionTransformation yAxisInputTransformation,
final Transformation yAxisOutputTransformation,
final Number goalValue,
final EComparison criterion,
final StatisticalParameter aggregate) {
super(
EAttributeType.NEVER_STORED,
xAxisTransformation,
yAxisInputTransformation,
yAxisOutputTransformation);
final IDimension goalDim;
if (goalValue == null) {
throw new IllegalArgumentException( //
"Goal value of ECDF cannot be null."); //$NON-NLS-1$
}
if (criterion == null) {
throw new IllegalArgumentException( //
"Comparison criterion of ECDF cannot be null."); //$NON-NLS-1$
}
if (aggregate == null) {
throw new IllegalArgumentException( //
"Aggregate to join ECDFs of different instance sets cannot be null."); //$NON-NLS-1$
}
goalDim = yAxisInputTransformation.getDimension();
switch (goalDim.getDataType()) {
case BYTE:
case SHORT:
case INT:
case LONG:
{
if (yAxisInputTransformation.isLongArithmeticAccurate()) {
this.m_useLongGoal = true;
if ((NumericalTypes.getTypes(goalValue) & NumericalTypes.IS_LONG) != 0) {
this.m_goalValueLong = goalValue.longValue();
} else {
this.m_goalValueLong = ECDF.__doubleToLong(goalValue.doubleValue(), criterion);
}
this.m_goalValueDouble = this.m_goalValueLong;
break;
}
// fall through
}
// $FALL-THROUGH$
default:
{
this.m_goalValueDouble = goalValue.doubleValue();
this.m_useLongGoal = false;
this.m_goalValueLong = ECDF.__doubleToLong(this.m_goalValueDouble, criterion);
}
}
this.m_criterion = criterion;
this.m_aggregate = aggregate;
}
/**
* The Estimated Cumulative Distribution Function returns, for an experiment or instance runs set,
* the fraction of runs which have reached a specified goal.
*/
public final class ECDF extends FunctionAttribute<IElementSet> {
/** the criterion parameter */
public static final String CRITERION_PARAM = "criterion"; // $NON-NLS-1$
/** the goal parameter */
public static final String GOAL_PARAM = "goal"; // $NON-NLS-1$
/** the aggregate parameter */
public static final String AGGREGATE_PARAM = "aggregate"; // $NON-NLS-1$
/** the references to be used for the ecdf */
public static final Bibliography REFERENCES = ECDF.__buildReferences();
/** the goal value as {@code double} */
private final double m_goalValueDouble;
/** the goal value as {@code long} */
private final long m_goalValueLong;
/**
* should we use {@link #m_goalValueDouble} as goal ({@code true}) or {@link #m_goalValueLong}
* ({@code false}?
*/
private final boolean m_useLongGoal;
/** the comparison used to determine whether the goal criterion was met */
private final EComparison m_criterion;
/** the way to aggregate the different ECDFs */
private final StatisticalParameter m_aggregate;
/**
* Create the ECDF attribute
*
* @param xAxisTransformation the transformation to be applied to the {@code x}-axis
* @param yAxisInputTransformation the transformation to be applied to the data of the {@code y}
* -axis before being fed to the actual computation
* @param yAxisOutputTransformation the transformation of the result of the function applied to
* the data on the {@code y}-axis.
* @param goalValue the goal value
* @param criterion the goal comparison criterion
* @param aggregate the method to aggregate the ECDFs
*/
public ECDF(
final DimensionTransformation xAxisTransformation,
final DimensionTransformation yAxisInputTransformation,
final Transformation yAxisOutputTransformation,
final Number goalValue,
final EComparison criterion,
final StatisticalParameter aggregate) {
super(
EAttributeType.NEVER_STORED,
xAxisTransformation,
yAxisInputTransformation,
yAxisOutputTransformation);
final IDimension goalDim;
if (goalValue == null) {
throw new IllegalArgumentException( //
"Goal value of ECDF cannot be null."); //$NON-NLS-1$
}
if (criterion == null) {
throw new IllegalArgumentException( //
"Comparison criterion of ECDF cannot be null."); //$NON-NLS-1$
}
if (aggregate == null) {
throw new IllegalArgumentException( //
"Aggregate to join ECDFs of different instance sets cannot be null."); //$NON-NLS-1$
}
goalDim = yAxisInputTransformation.getDimension();
switch (goalDim.getDataType()) {
case BYTE:
case SHORT:
case INT:
case LONG:
{
if (yAxisInputTransformation.isLongArithmeticAccurate()) {
this.m_useLongGoal = true;
if ((NumericalTypes.getTypes(goalValue) & NumericalTypes.IS_LONG) != 0) {
this.m_goalValueLong = goalValue.longValue();
} else {
this.m_goalValueLong = ECDF.__doubleToLong(goalValue.doubleValue(), criterion);
}
this.m_goalValueDouble = this.m_goalValueLong;
break;
}
// fall through
}
// $FALL-THROUGH$
default:
{
this.m_goalValueDouble = goalValue.doubleValue();
this.m_useLongGoal = false;
this.m_goalValueLong = ECDF.__doubleToLong(this.m_goalValueDouble, criterion);
}
}
this.m_criterion = criterion;
this.m_aggregate = aggregate;
}
/** {@inheritDoc} */
@Override
protected void yAxisRenderYAxisSourceAsParameter(final IMath out) {
try (final IMath math = out.compare(EMathComparison.fromEComparison(this.m_criterion))) {
super.yAxisRenderYAxisSourceAsParameter(math);
try (final IText number = math.number()) {
if (this.isGoalValueLong() || (this.m_goalValueLong == this.m_goalValueDouble)) {
number.append(this.m_goalValueLong);
} else {
number.append(this.m_goalValueDouble);
}
}
}
}
/** {@inheritDoc} */
@Override
protected void yAxisRenderYAxisSourceAsParameter(final ITextOutput out) {
super.yAxisRenderYAxisSourceAsParameter(out);
out.append(EMathComparison.fromEComparison(this.m_criterion).getOperatorChar());
if (this.isGoalValueLong() || (this.m_goalValueLong == this.m_goalValueDouble)) {
out.append(this.m_goalValueLong);
} else {
out.append(this.m_goalValueDouble);
}
}
/** {@inheritDoc} */
@Override
protected ETextCase printNameInDescription(
final ITextOutput textOut,
final ETextCase textCase,
final boolean fromYAxisSemanticComponent) {
ETextCase next;
next = super.printNameInDescription(textOut, textCase, fromYAxisSemanticComponent);
if (!fromYAxisSemanticComponent) {
if (textOut instanceof IComplexText) {
try (final BibliographyBuilder builder =
((IComplexText) textOut).cite(ECitationMode.ID, next, ESequenceMode.COMMA)) {
builder.addAll(ECDF.REFERENCES);
}
}
}
return next.nextCase();
}
/** {@inheritDoc} */
@Override
public ETextCase printDescription(final ITextOutput textOut, final ETextCase textCase) {
final DimensionTransformation xIn, yIn;
final Transformation yOut;
ETextCase use;
use = super.printDescription(textOut, textCase).nextCase();
yIn = this.getYAxisInputTransformation();
xIn = this.getXAxisTransformation();
yOut = this.getYAxisOutputTransformation();
textOut.append(" The "); // $NON-NLS-1$
if (textOut instanceof IComplexText) {
try (final IMath math = ((IComplexText) textOut).inlineMath()) {
this.yAxisMathRender(math, DefaultParameterRenderer.INSTANCE);
}
} else {
this.yAxisMathRender(textOut, DefaultParameterRenderer.INSTANCE);
}
textOut.append( //
" represents the fraction of runs which reach a value of "); //$NON-NLS-1$
yIn.printShortName(textOut, use);
textOut.append(' ');
textOut.append(this.m_criterion.toString());
textOut.append(' ');
if (this.m_useLongGoal) {
textOut.append(this.m_goalValueLong);
} else {
textOut.append(this.m_goalValueDouble);
}
textOut.append(" for a given ellapsed runtime measured in "); // $NON-NLS-1$
xIn.getDimension().printShortName(textOut, use);
if (yOut.isIdentityTransformation()) {
textOut.append(". The "); // $NON-NLS-1$
this.printShortName(textOut, use);
} else {
textOut.append( //
". We do not use these fractions directly, but instead compute "); //$NON-NLS-1$
if (textOut instanceof IComplexText) {
try (final IMath math = ((IComplexText) textOut).inlineMath()) {
this.yAxisMathRender(math, DefaultParameterRenderer.INSTANCE);
}
} else {
this.yAxisMathRender(textOut, DefaultParameterRenderer.INSTANCE);
}
textOut.append(". The result of this formula"); // $NON-NLS-1$
}
textOut.append( //
" is always computed over the runs of an experiment for a given benchmark instance. If runs for multiple instances are available, we aggregate the results by computing their "); //$NON-NLS-1$
this.m_aggregate.printLongName(textOut, use);
textOut.append('.');
if (!(xIn.isIdentityTransformation())) {
textOut.append(" The x-axis does not represent the values of "); // $NON-NLS-1$
xIn.getDimension().printShortName(textOut, use);
textOut.append(" directly, but instead "); // $NON-NLS-1$
if (textOut instanceof IComplexText) {
try (final IMath math = ((IComplexText) textOut).inlineMath()) {
xIn.mathRender(math, DefaultParameterRenderer.INSTANCE);
}
} else {
xIn.mathRender(textOut, DefaultParameterRenderer.INSTANCE);
}
textOut.append('.');
}
if (yOut.isIdentityTransformation()) {
textOut.append(" The "); // $NON-NLS-1$
this.printShortName(textOut, use);
textOut.append(" is always between "); // $NON-NLS-1$
textOut.append(0);
textOut.append(" and "); // $NON-NLS-1$
textOut.append(1);
textOut.append(" \u2012 and the higher it is, the better."); // $NON-NLS-1$
}
return use;
}
/**
* convert a {@code double} goal value to a {@code long}
*
* @param d the goal value
* @param criterion the goal criterion
* @return the converted value
*/
private static final long __doubleToLong(final double d, final EComparison criterion) {
if (d <= Long.MIN_VALUE) {
return Long.MIN_VALUE;
}
if (d >= Long.MAX_VALUE) {
return Long.MAX_VALUE;
}
switch (criterion) {
case LESS:
case LESS_OR_EQUAL:
{
if (d != d) {
return Long.MIN_VALUE;
}
return ((long) (0.5d + Math.floor(d)));
}
case EQUAL:
case NOT_EQUAL:
{
if (d != d) {
throw new IllegalArgumentException( //
"Cannot transform "
+ d
+ //$NON-NLS-1$
" to a long under "
+ criterion); //$NON-NLS-1$
}
return ((long) d);
}
case GREATER:
case GREATER_OR_EQUAL:
{
if (d != d) {
return Long.MAX_VALUE;
}
return ((long) (0.5d + Math.ceil(d)));
}
default:
{
throw new IllegalArgumentException( //
"Illegal comparison operator: " //$NON-NLS-1$
+ criterion);
}
}
}
/** {@inheritDoc} */
@Override
protected final int calcHashCode() {
return HashUtils.combineHashes( //
HashUtils.combineHashes( //
super.calcHashCode(), //
(this.m_useLongGoal
? //
HashUtils.hashCode(this.m_goalValueLong)
: HashUtils.hashCode(this.m_goalValueDouble)) //
),
HashUtils.combineHashes( //
HashUtils.hashCode(this.m_criterion), //
HashUtils.hashCode(this.m_aggregate) //
));
}
/** {@inheritDoc} */
@Override
protected final boolean isEqual(final FunctionAttribute<IElementSet> other) {
final ECDF ecdf;
ecdf = ((ECDF) other);
return ((this.m_useLongGoal == ecdf.m_useLongGoal) //
&& //
(this.m_useLongGoal
? //
(this.m_goalValueLong == ecdf.m_goalValueLong)
: //
(EComparison.EQUAL.compare(this.m_goalValueDouble, ecdf.m_goalValueDouble)))
&& //
this.m_criterion.equals(ecdf.m_criterion)
&& //
this.m_aggregate.equals(ecdf.m_aggregate));
}
/**
* Is the goal value a {@code long} value?
*
* @return {@code true} if the goal value is a {@code long} value, {@code false} if it is a {@code
* double}
* @see #getGoalValueDouble()
* @see #getGoalValueLong()
*/
public final boolean isGoalValueLong() {
return this.m_useLongGoal;
}
/**
* Get the goal value as a {@code long}. The result of this method will only be accurate if {@link
* #isGoalValueLong()} returned {@code true}. Otherwise, {@link #getGoalValueDouble()} should be
* used.
*
* @return the goal value as a {@code long}
* @see #isGoalValueLong()
* @see #getGoalValueDouble()
*/
public final long getGoalValueLong() {
return this.m_goalValueLong;
}
/**
* Get the goal value as a {@code double}. The result of this method will only be accurate if
* {@link #isGoalValueLong()} returned {@code false}. Otherwise, {@link #getGoalValueLong()}
* should be used.
*
* @return the goal value as a {@code double}
* @see #isGoalValueLong()
* @see #getGoalValueLong()
*/
public final double getGoalValueDouble() {
return this.m_goalValueDouble;
}
/** {@inheritDoc} */
@Override
public final String getShortName() {
return "ECDF"; //$NON-NLS-1$
}
/** {@inheritDoc} */
@Override
public final String getLongName() {
return "estimated cumulative distribution function"; //$NON-NLS-1$
}
/**
* Compute the raw matrix for an instance run set
*
* @param data the data the instance runs
* @return the raw matrix
*/
private final IMatrix __computeInstanceRuns(final IInstanceRuns data) {
final _List list;
final DimensionTransformation xIn, yIn;
final Transformation yOut;
final IDimension timeDim, goalDim;
xIn = this.getXAxisTransformation();
try (final TransformationFunction xInFunc = xIn.use(data)) {
yIn = this.getYAxisInputTransformation();
try (final TransformationFunction yInFunc = yIn.use(data)) {
yOut = this.getYAxisOutputTransformation();
try (final TransformationFunction yOutFunc = yOut.use(data)) {
timeDim = xIn.getDimension();
goalDim = yIn.getDimension();
if (goalDim.getDataType().isInteger() && yInFunc.isLongArithmeticAccurate()) {
if (timeDim.getDataType().isInteger()) {
list =
new _LongTimeLongGoal(
timeDim, goalDim, this.m_criterion, yInFunc, this.m_goalValueLong);
} else {
list =
new _DoubleTimeLongGoal(
timeDim, goalDim, this.m_criterion, yInFunc, this.m_goalValueLong);
}
} else {
if (timeDim.getDataType().isInteger()) {
list =
new _LongTimeDoubleGoal(
timeDim, goalDim, this.m_criterion, yInFunc, this.m_goalValueDouble);
} else {
list =
new _DoubleTimeDoubleGoal(
timeDim, goalDim, this.m_criterion, yInFunc, this.m_goalValueDouble);
}
}
for (final IRun run : data.getData()) {
list._addRun(run);
}
return list._toMatrix(xInFunc, yOutFunc);
}
}
}
}
/**
* Compute the aggregate per experiment
*
* @param data the data
* @return the aggregated data
*/
private final IMatrix __computeExperiment(final IExperiment data) {
final IMatrix[] matrices;
final ArrayListView<? extends IInstanceRuns> runs;
int i;
runs = data.getData();
i = runs.size();
matrices = new IMatrix[i];
for (; (--i) >= 0; ) {
matrices[i] = this.__computeInstanceRuns(runs.get(i));
}
return this.m_aggregate.aggregate2D(matrices, 0, 1, Identity.INSTANCE);
}
/**
* Compute the aggregate per experiment set
*
* @param data the data
* @return the aggregated data
*/
private final IMatrix __computeExperimentSet(final IExperimentSet data) {
final ArrayList<IMatrix> matrices;
matrices = new ArrayList<>();
for (final IExperiment exp : data.getData()) {
for (final IInstanceRuns irs : exp.getData()) {
matrices.add(this.__computeInstanceRuns(irs));
}
}
return this.m_aggregate.aggregate2D(matrices, 0, 1, Identity.INSTANCE);
}
/** {@inheritDoc} */
@Override
protected final IMatrix compute(final IElementSet data) {
if (data instanceof IInstanceRuns) {
return this.__computeInstanceRuns((IInstanceRuns) data);
}
if (data instanceof IExperiment) {
return this.__computeExperiment((IExperiment) data);
}
if (data instanceof IExperimentSet) {
return this.__computeExperimentSet((IExperimentSet) data);
}
throw new IllegalArgumentException(
"Cannot compute ECDF over " //$NON-NLS-1$
+ data);
}
/**
* Create an instance of {@link ECDF} based on an experiment set and a configuration
*
* @param data the data (experiment set)
* @param config the configuration
* @return the instance of the aggregate
*/
public static final ECDF create(final IExperimentSet data, final Configuration config) {
DimensionTransformationParser dimParser;
final DimensionTransformation xIn, yIn;
final Transformation yOut;
final StatisticalParameter aggregate;
final IDimension goalDim;
EComparison compare;
Number goal;
dimParser = new DimensionTransformationParser(data);
xIn =
config.get( //
FunctionAttribute.X_AXIS_PARAM, dimParser, null);
if (xIn == null) { //
throw new IllegalArgumentException(
"Must specify an x-dimension via parameter '" //$NON-NLS-1$
+ FunctionAttribute.X_AXIS_PARAM
+ '\'');
}
yIn = config.get(FunctionAttribute.Y_INPUT_AXIS_PARAM, dimParser, null);
if (yIn == null) { //
throw new IllegalArgumentException(
"Must specify an input dimension for the y-axis via parameter '" //$NON-NLS-1$
+ FunctionAttribute.Y_INPUT_AXIS_PARAM
+ '\'');
}
dimParser = null;
yOut =
config.get(
FunctionAttribute.Y_AXIS_OUTPUT_PARAM,
new NamedParameterTransformationParser(data),
new Transformation());
aggregate =
config.get(
ECDF.AGGREGATE_PARAM,
StatisticalParameterParser.getInstance(),
ArithmeticMean.INSTANCE);
goalDim = yIn.getDimension();
if (goalDim.getDirection().isIncreasing()) {
compare = EComparison.GREATER_OR_EQUAL;
if (goalDim.getDataType().isInteger()) {
goal = Long.valueOf(goalDim.getParser().getUpperBoundLong());
} else {
goal = Double.valueOf(goalDim.getParser().getUpperBoundDouble());
}
} else {
compare = EComparison.LESS_OR_EQUAL;
if (goalDim.getDataType().isInteger()) {
goal = Long.valueOf(goalDim.getParser().getLowerBoundLong());
} else {
goal = Double.valueOf(goalDim.getParser().getLowerBoundDouble());
}
}
goal = config.get(ECDF.GOAL_PARAM, AnyNumberParser.INSTANCE, goal);
compare = config.get(ECDF.CRITERION_PARAM, ComparisonParser.getInstance(), compare);
return new ECDF(xIn, yIn, yOut, goal, compare, aggregate);
}
/**
* build the references to be used when plotting an ECDF
*
* @return the bibliography of references
*/
private static final Bibliography __buildReferences() {
final BibAuthor hoos;
hoos =
new BibAuthor( //
"Holger H.", //$NON-NLS-1$
"Hoos"); //$NON-NLS-1$
try (final BibliographyBuilder bibBuilder = new BibliographyBuilder()) {
try (final BibInProceedingsBuilder inProc = //
bibBuilder.inProceedings()) {
try (final BibAuthorsBuilder authors = inProc.setAuthors()) {
authors.addAuthor(hoos);
try (final BibAuthorBuilder author = authors.author()) {
author.setPersonalName("Thomas"); // $NON-NLS-1$
author.setFamilyName("St\u00fctzle"); // $NON-NLS-1$
}
}
inProc.setTitle( //
"Evaluating Las Vegas Algorithms \u2012 Pitfalls and Remedies"); //$NON-NLS-1$
try (final BibProceedingsBuilder proc = inProc.proceedings()) {
proc.setTitle( //
"Proceedings of the 14th Conference on Uncertainty in Artificial Intelligence (UAI'98)"); //$NON-NLS-1$
try (final BibDateBuilder date = proc.startDate()) {
date.setYear(1998);
date.setMonth(EBibMonth.JULY);
date.setDay(24);
}
try (final BibDateBuilder date = proc.endDate()) {
date.setYear(1998);
date.setMonth(EBibMonth.JULY);
date.setDay(26);
}
try (final BibAuthorsBuilder editors = proc.setEditors()) {
try (final BibAuthorBuilder editor = editors.author()) {
editor.setPersonalName("Gregory F."); // $NON-NLS-1$
editor.setFamilyName("Cooper"); // $NON-NLS-1$
}
try (final BibAuthorBuilder editor = editors.author()) {
editor.setPersonalName("Serafin"); // $NON-NLS-1$
editor.setFamilyName("Moral"); // $NON-NLS-1$
}
}
try (final BibOrganizationBuilder loc = proc.location()) {
loc.setAddress("Madison, WI, USA"); // $NON-NLS-1$
}
try (final BibOrganizationBuilder pub = proc.publisher()) {
pub.setAddress("San Francisco, CA, USA"); // $NON-NLS-1$
pub.setName("Morgan Kaufmann Publishers Inc."); // $NON-NLS-1$
}
}
inProc.setStartPage("238"); // $NON-NLS-1$
inProc.setEndPage("245"); // $NON-NLS-1$
inProc.setURL( //
"http://www.intellektik.informatik.tu-darmstadt.de/TR/1998/98-02.ps.Z"); //$NON-NLS-1$
}
try (final BibInProceedingsBuilder inProc = //
bibBuilder.inProceedings()) {
try (final BibAuthorsBuilder authors = inProc.setAuthors()) {
try (final BibAuthorBuilder author = authors.author()) {
author.setPersonalName("Dave Andrew Douglas"); // $NON-NLS-1$
author.setFamilyName("Tompkins"); // $NON-NLS-1$
}
authors.addAuthor(hoos);
}
inProc.setTitle( //
"UBCSAT: An Implementation and Experimentation Environment for SLS Algorithms for SAT and MAX-SAT"); //$NON-NLS-1$
try (final BibProceedingsBuilder proc = inProc.proceedings()) {
proc.setTitle( //
"Revised Selected Papers from the Seventh International Conference on Theory and Applications of Satisfiability Testing (SAT'04)"); //$NON-NLS-1$
try (final BibDateBuilder date = proc.startDate()) {
date.setYear(2004);
date.setMonth(EBibMonth.MAY);
date.setDay(10);
}
try (final BibDateBuilder date = proc.endDate()) {
date.setYear(2004);
date.setMonth(EBibMonth.MAY);
date.setDay(13);
}
try (final BibAuthorsBuilder editors = proc.setEditors()) {
editors.addAuthor(hoos);
try (final BibAuthorBuilder editor = editors.author()) {
editor.setPersonalName("David G."); // $NON-NLS-1$
editor.setFamilyName("Mitchell"); // $NON-NLS-1$
}
}
try (final BibOrganizationBuilder loc = proc.location()) {
loc.setAddress("Vancouver, BC, Canada"); // $NON-NLS-1$
}
try (final BibOrganizationBuilder pub = proc.publisher()) {
pub.setAddress("Berlin, Germany"); // $NON-NLS-1$
pub.setName("Springer-Verlag GmbH"); // $NON-NLS-1$
}
proc.setSeries("Lecture Notes in Computer Science (LNCS)"); // $NON-NLS-1$
proc.setVolume("3542"); // $NON-NLS-1$
}
inProc.setDOI("10.1007/11527695"); // $NON-NLS-1$
inProc.setStartPage("306"); // $NON-NLS-1$
inProc.setEndPage("320"); // $NON-NLS-1$
inProc.setURL( //
"http://ubcsat.dtompkins.com/downloads/sat04proc-ubcsat.pdf?attredirects=0"); //$NON-NLS-1$
}
try (BibTechReportBuilder report = bibBuilder.techReport()) {
try (final BibAuthorsBuilder authors = report.setAuthors()) {
try (final BibAuthorBuilder author = authors.author()) {
author.setPersonalName("Nikolaus"); // $NON-NLS-1$
author.setFamilyName("Hansen"); // $NON-NLS-1$
}
try (final BibAuthorBuilder author = authors.author()) {
author.setPersonalName("Anne"); // $NON-NLS-1$
author.setFamilyName("Auger"); // $NON-NLS-1$
}
try (final BibAuthorBuilder author = authors.author()) {
author.setPersonalName("Steffen"); // $NON-NLS-1$
author.setFamilyName("Finck"); // $NON-NLS-1$
}
try (final BibAuthorBuilder author = authors.author()) {
author.setPersonalName("Raymond"); // $NON-NLS-1$
author.setFamilyName("Ros"); // $NON-NLS-1$
}
}
report.setTitle( //
"Real-Parameter Black-Box Optimization Benchmarking: Experimental Setup"); //$NON-NLS-1$
try (final BibOrganizationBuilder pub = report.publisher()) {
pub.setAddress("Orsay, France"); // $NON-NLS-1$
pub.setName( //
"Universit\u00e9 Paris Sud, Institut National de Recherche en Informatique et en Automatique (INRIA) Futurs, \u00c9quipe TAO"); //$NON-NLS-1$
}
try (final BibDateBuilder date = report.date()) {
date.setYear(2012);
date.setMonth(EBibMonth.MARCH);
date.setDay(24);
}
report.setURL( //
"http://coco.lri.fr/BBOB-downloads/download11.05/bbobdocexperiment.pdf"); //$NON-NLS-1$
}
return bibBuilder.getResult();
}
}
}