http://sumowiki.intec.ugent.be/index.php?title=General_guidelines&feed=atom&action=historyGeneral guidelines - Revision history2024-03-28T15:22:16ZRevision history for this page on the wikiMediaWiki 1.35.4http://sumowiki.intec.ugent.be/index.php?title=General_guidelines&diff=5861&oldid=prevJavdrher: /* Adaptive Model Builders */2014-02-27T15:30:16Z<p><span dir="auto"><span class="autocomment">Adaptive Model Builders</span></span></p>
<table class="diff diff-contentalign-left diff-editfont-monospace" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 15:30, 27 February 2014</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l25" >Line 25:</td>
<td colspan="2" class="diff-lineno">Line 25:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>When using the [[Config:SequentialDesign#error|error-based sample selector]] separately, it is always a good idea to combine it with the [[Config:SequentialDesign#voronoi|voronoi]], to combat stability/robustness issues the error-based sample selector often causes. It is a good idea to select about 60% of the samples with error, and 40% with the voronoi. This will ensure that at least the entire design space is covered to a certain degree. This additional sample selector is NOT necessary when using LOLA-Voronoi. To combine sample selectors, create a CombinedSampleSelector. See the [[Config:SequentialDesign#default|default sample selector]] for an example.</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>When using the [[Config:SequentialDesign#error|error-based sample selector]] separately, it is always a good idea to combine it with the [[Config:SequentialDesign#voronoi|voronoi]], to combat stability/robustness issues the error-based sample selector often causes. It is a good idea to select about 60% of the samples with error, and 40% with the voronoi. This will ensure that at least the entire design space is covered to a certain degree. This additional sample selector is NOT necessary when using LOLA-Voronoi. To combine sample selectors, create a CombinedSampleSelector. See the [[Config:SequentialDesign#default|default sample selector]] for an example.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'>−</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>== <del class="diffchange diffchange-inline">Adaptive </del>Model Builders ==</div></td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>== Model Builders ==</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>The question that always gets asked is ''Which model type should I use for my data?'' Unfortunately there is no straightforward since it all depends on your problem: how many dimensions, how many points, is your function rugged, smooth, or both, is there noise, etc, etc. Based on this knowledge it is possible to say which model types are more likely to do well but it remains a heuristic. Best is to try a few and see what happens, or use the ''heterogenetic'' model builder to try multiple model types in parallel and automatically try to determine the best type.</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>The question that always gets asked is ''Which model type should I use for my data?'' Unfortunately there is no straightforward since it all depends on your problem: how many dimensions, how many points, is your function rugged, smooth, or both, is there noise, etc, etc. Based on this knowledge it is possible to say which model types are more likely to do well but it remains a heuristic. Best is to try a few and see what happens, or use the ''heterogenetic'' model builder to try multiple model types in parallel and automatically try to determine the best type.</div></td></tr>
</table>Javdrherhttp://sumowiki.intec.ugent.be/index.php?title=General_guidelines&diff=5860&oldid=prevJavdrher: /* Sample Selectors */2014-02-27T15:30:04Z<p><span dir="auto"><span class="autocomment">Sample Selectors</span></span></p>
<table class="diff diff-contentalign-left diff-editfont-monospace" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 15:30, 27 February 2014</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l15" >Line 15:</td>
<td colspan="2" class="diff-lineno">Line 15:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>Selecting a good Measure '''is a very important''' part of the modeling process! It is CRUCIAL that you think well about this. Make sure you also read [[Multi-Objective Modeling]].</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>Selecting a good Measure '''is a very important''' part of the modeling process! It is CRUCIAL that you think well about this. Make sure you also read [[Multi-Objective Modeling]].</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'>−</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>== <del class="diffchange diffchange-inline">Sample Selectors </del>==</div></td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>== <ins class="diffchange diffchange-inline">Sequential Design </ins>==</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>The default [[Config:SequentialDesign|Sequential Design]] is the [[Config:SequentialDesign#lola-voronoi|LOLA-Voronoi sample selector]] combined with the [[Config:SequentialDesign#error|error-based sample selector]], with a weight of 0.7 for LOLA and 0.3 for error. This is a very robust sample selector, capable of dealing with most situations. There are, however, some cases in which it is advisable to choose a different one:</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>The default [[Config:SequentialDesign|Sequential Design]] is the [[Config:SequentialDesign#lola-voronoi|LOLA-Voronoi sample selector]] combined with the [[Config:SequentialDesign#error|error-based sample selector]], with a weight of 0.7 for LOLA and 0.3 for error. This is a very robust sample selector, capable of dealing with most situations. There are, however, some cases in which it is advisable to choose a different one:</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'>−</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>* '''Large-scale problems (1000+ samples):''' LOLA-Voronoi's time complexity is O(n²) to the number of samples n, so for large-scale experiments in which many samples are taken, LOLA-Voronoibecomes quite slow. Depending on the time it takes to perform one simulation, this may or may not be a problem. If it takes a long time to perform one simulation, the cost for selecting new samples with LOLA-Voronoi might still be negligible. If, however, you need a quicker sample selector, it is advized to use [[Config:<del class="diffchange diffchange-inline">SampleSelector</del>#voronoi|voronoi]] or [[Config:<del class="diffchange diffchange-inline">SampleSelector</del>#error|error]] instead.</div></td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>* '''Large-scale problems (1000+ samples):''' LOLA-Voronoi's time complexity is O(n²) to the number of samples n, so for large-scale experiments in which many samples are taken, LOLA-Voronoibecomes quite slow. Depending on the time it takes to perform one simulation, this may or may not be a problem. If it takes a long time to perform one simulation, the cost for selecting new samples with LOLA-Voronoi might still be negligible. If, however, you need a quicker sample selector, it is advized to use [[Config:<ins class="diffchange diffchange-inline">SequentialDesign</ins>#voronoi|voronoi]] or [[Config:<ins class="diffchange diffchange-inline">SequentialDesign</ins>#error|error]] instead.</div></td></tr>
<tr><td class='diff-marker'>−</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>* '''Rational modeler:''' Benchmarks have shown that the gain of LOLA-Voronoi over the [[Config:<del class="diffchange diffchange-inline">SampleSelector</del>#error|error-based sample selector]] when using global approximation methods (mainly rational/polynomial) is pretty much zero. It is therefore advisable to use the (much faster) [[Config:SampleSelector#error|error-based sample selector]] when using the Rational modeler. This can be done by changing the weights in default.xml to 1.0 for error and 0.0 for LOLA.</div></td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>* '''Rational modeler:''' Benchmarks have shown that the gain of LOLA-Voronoi over the [[Config:<ins class="diffchange diffchange-inline">SequentialDesign</ins>#error|error-based sample selector]] when using global approximation methods (mainly rational/polynomial) is pretty much zero. It is therefore advisable to use the (much faster) [[Config:SampleSelector#error|error-based sample selector]] when using the Rational modeler. This can be done by changing the weights in default.xml to 1.0 for error and 0.0 for LOLA.</div></td></tr>
<tr><td class='diff-marker'>−</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>* If you need to sample multiple outputs at once, with one sample selector, or you need an auto-sampled input (for example: a frequency input), you should use [[Config:<del class="diffchange diffchange-inline">SampleSelector</del>#lola-voronoi|LOLA-Voronoi]]. It is the only sample selector with fully integrated and optimized support for these features.</div></td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>* If you need to sample multiple outputs at once, with one sample selector, or you need an auto-sampled input (for example: a frequency input), you should use [[Config:<ins class="diffchange diffchange-inline">SequentialDesign</ins>#lola-voronoi|LOLA-Voronoi]]. It is the only sample selector with fully integrated and optimized support for these features.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'>−</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>When using the [[Config:<del class="diffchange diffchange-inline">SampleSelector</del>#error|error-based sample selector]] separately, it is always a good idea to combine it with the [[Config:<del class="diffchange diffchange-inline">SampleSelector</del>#voronoi|voronoi]], to combat stability/robustness issues the error-based sample selector often causes. It is a good idea to select about 60% of the samples with error, and 40% with the voronoi. This will ensure that at least the entire design space is covered to a certain degree. This additional sample selector is NOT necessary when using LOLA-Voronoi. To combine sample selectors, create a CombinedSampleSelector. See the [[Config:<del class="diffchange diffchange-inline">SampleSelector</del>#default|default sample selector]] for an example.</div></td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>When using the [[Config:<ins class="diffchange diffchange-inline">SequentialDesign</ins>#error|error-based sample selector]] separately, it is always a good idea to combine it with the [[Config:<ins class="diffchange diffchange-inline">SequentialDesign</ins>#voronoi|voronoi]], to combat stability/robustness issues the error-based sample selector often causes. It is a good idea to select about 60% of the samples with error, and 40% with the voronoi. This will ensure that at least the entire design space is covered to a certain degree. This additional sample selector is NOT necessary when using LOLA-Voronoi. To combine sample selectors, create a CombinedSampleSelector. See the [[Config:<ins class="diffchange diffchange-inline">SequentialDesign</ins>#default|default sample selector]] for an example.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Adaptive Model Builders ==</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Adaptive Model Builders ==</div></td></tr>
</table>Javdrherhttp://sumowiki.intec.ugent.be/index.php?title=General_guidelines&diff=5859&oldid=prevJavdrher: /* Sample Selectors */2014-02-27T15:28:40Z<p><span dir="auto"><span class="autocomment">Sample Selectors</span></span></p>
<table class="diff diff-contentalign-left diff-editfont-monospace" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 15:28, 27 February 2014</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l17" >Line 17:</td>
<td colspan="2" class="diff-lineno">Line 17:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Sample Selectors ==</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Sample Selectors ==</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'>−</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>The default [[Config:<del class="diffchange diffchange-inline">SampleSelector</del>|<del class="diffchange diffchange-inline">SampleSelector</del>]] is the [[Config:<del class="diffchange diffchange-inline">SampleSelector</del>#lola-voronoi|LOLA-Voronoi sample selector]] combined with the [[Config:<del class="diffchange diffchange-inline">SampleSelector</del>#error|error-based sample selector]], with a weight of 0.7 for LOLA and 0.3 for error. This is a very robust sample selector, capable of dealing with most situations. There are, however, some cases in which it is advisable to choose a different one:</div></td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>The default [[Config:<ins class="diffchange diffchange-inline">SequentialDesign</ins>|<ins class="diffchange diffchange-inline">Sequential Design</ins>]] is the [[Config:<ins class="diffchange diffchange-inline">SequentialDesign</ins>#lola-voronoi|LOLA-Voronoi sample selector]] combined with the [[Config:<ins class="diffchange diffchange-inline">SequentialDesign</ins>#error|error-based sample selector]], with a weight of 0.7 for LOLA and 0.3 for error. This is a very robust sample selector, capable of dealing with most situations. There are, however, some cases in which it is advisable to choose a different one:</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''Large-scale problems (1000+ samples):''' LOLA-Voronoi's time complexity is O(n²) to the number of samples n, so for large-scale experiments in which many samples are taken, LOLA-Voronoibecomes quite slow. Depending on the time it takes to perform one simulation, this may or may not be a problem. If it takes a long time to perform one simulation, the cost for selecting new samples with LOLA-Voronoi might still be negligible. If, however, you need a quicker sample selector, it is advized to use [[Config:SampleSelector#voronoi|voronoi]] or [[Config:SampleSelector#error|error]] instead.</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''Large-scale problems (1000+ samples):''' LOLA-Voronoi's time complexity is O(n²) to the number of samples n, so for large-scale experiments in which many samples are taken, LOLA-Voronoibecomes quite slow. Depending on the time it takes to perform one simulation, this may or may not be a problem. If it takes a long time to perform one simulation, the cost for selecting new samples with LOLA-Voronoi might still be negligible. If, however, you need a quicker sample selector, it is advized to use [[Config:SampleSelector#voronoi|voronoi]] or [[Config:SampleSelector#error|error]] instead.</div></td></tr>
</table>Javdrherhttp://sumowiki.intec.ugent.be/index.php?title=General_guidelines&diff=4995&oldid=prevAdmin: /* Sample Selectors */2009-10-02T14:50:23Z<p><span dir="auto"><span class="autocomment">Sample Selectors</span></span></p>
<table class="diff diff-contentalign-left diff-editfont-monospace" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 14:50, 2 October 2009</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l20" >Line 20:</td>
<td colspan="2" class="diff-lineno">Line 20:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''Large-scale problems (1000+ samples):''' LOLA-Voronoi's time complexity is O(n²) to the number of samples n, so for large-scale experiments in which many samples are taken, LOLA-Voronoibecomes quite slow. Depending on the time it takes to perform one simulation, this may or may not be a problem. If it takes a long time to perform one simulation, the cost for selecting new samples with LOLA-Voronoi might still be negligible. If, however, you need a quicker sample selector, it is advized to use [[Config:SampleSelector#voronoi|voronoi]] or [[Config:SampleSelector#error|error]] instead.</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''Large-scale problems (1000+ samples):''' LOLA-Voronoi's time complexity is O(n²) to the number of samples n, so for large-scale experiments in which many samples are taken, LOLA-Voronoibecomes quite slow. Depending on the time it takes to perform one simulation, this may or may not be a problem. If it takes a long time to perform one simulation, the cost for selecting new samples with LOLA-Voronoi might still be negligible. If, however, you need a quicker sample selector, it is advized to use [[Config:SampleSelector#voronoi|voronoi]] or [[Config:SampleSelector#error|error]] instead.</div></td></tr>
<tr><td class='diff-marker'>−</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>* '''Rational modeler:''' Benchmarks have shown that the gain of LOLA-Voronoi over the [[SampleSelector#error|error-based sample selector]] when using global approximation methods (mainly rational/polynomial) is pretty much zero. It is therefore advisable to use the (much faster) [[Config:SampleSelector#error|error-based sample selector]] when using the Rational modeler. This can be done by changing the weights in default.xml to 1.0 for error and 0.0 for LOLA.</div></td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>* '''Rational modeler:''' Benchmarks have shown that the gain of LOLA-Voronoi over the [[<ins class="diffchange diffchange-inline">Config:</ins>SampleSelector#error|error-based sample selector]] when using global approximation methods (mainly rational/polynomial) is pretty much zero. It is therefore advisable to use the (much faster) [[Config:SampleSelector#error|error-based sample selector]] when using the Rational modeler. This can be done by changing the weights in default.xml to 1.0 for error and 0.0 for LOLA.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* If you need to sample multiple outputs at once, with one sample selector, or you need an auto-sampled input (for example: a frequency input), you should use [[Config:SampleSelector#lola-voronoi|LOLA-Voronoi]]. It is the only sample selector with fully integrated and optimized support for these features.</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* If you need to sample multiple outputs at once, with one sample selector, or you need an auto-sampled input (for example: a frequency input), you should use [[Config:SampleSelector#lola-voronoi|LOLA-Voronoi]]. It is the only sample selector with fully integrated and optimized support for these features.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
</table>Adminhttp://sumowiki.intec.ugent.be/index.php?title=General_guidelines&diff=4994&oldid=prevAdmin: /* Sample Selectors */2009-10-02T14:49:59Z<p><span dir="auto"><span class="autocomment">Sample Selectors</span></span></p>
<table class="diff diff-contentalign-left diff-editfont-monospace" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 14:49, 2 October 2009</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l20" >Line 20:</td>
<td colspan="2" class="diff-lineno">Line 20:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''Large-scale problems (1000+ samples):''' LOLA-Voronoi's time complexity is O(n²) to the number of samples n, so for large-scale experiments in which many samples are taken, LOLA-Voronoibecomes quite slow. Depending on the time it takes to perform one simulation, this may or may not be a problem. If it takes a long time to perform one simulation, the cost for selecting new samples with LOLA-Voronoi might still be negligible. If, however, you need a quicker sample selector, it is advized to use [[Config:SampleSelector#voronoi|voronoi]] or [[Config:SampleSelector#error|error]] instead.</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''Large-scale problems (1000+ samples):''' LOLA-Voronoi's time complexity is O(n²) to the number of samples n, so for large-scale experiments in which many samples are taken, LOLA-Voronoibecomes quite slow. Depending on the time it takes to perform one simulation, this may or may not be a problem. If it takes a long time to perform one simulation, the cost for selecting new samples with LOLA-Voronoi might still be negligible. If, however, you need a quicker sample selector, it is advized to use [[Config:SampleSelector#voronoi|voronoi]] or [[Config:SampleSelector#error|error]] instead.</div></td></tr>
<tr><td class='diff-marker'>−</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>* '''Rational modeler:''' Benchmarks have shown that the gain of LOLA-Voronoi over the [[SampleSelector#<del class="diffchange diffchange-inline">ErrorSampleSelector</del>|<del class="diffchange diffchange-inline">ErrorSampleSelector</del>]] when using global approximation methods (mainly rational/polynomial) is pretty much zero. It is therefore advisable to use the (much faster) [[Config:SampleSelector#error|error-based sample selector]] when using the Rational modeler. This can be done by changing the weights in default.xml to 1.0 for error and 0.0 for LOLA.</div></td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>* '''Rational modeler:''' Benchmarks have shown that the gain of LOLA-Voronoi over the [[SampleSelector#<ins class="diffchange diffchange-inline">error</ins>|<ins class="diffchange diffchange-inline">error-based sample selector</ins>]] when using global approximation methods (mainly rational/polynomial) is pretty much zero. It is therefore advisable to use the (much faster) [[Config:SampleSelector#error|error-based sample selector]] when using the Rational modeler. This can be done by changing the weights in default.xml to 1.0 for error and 0.0 for LOLA.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* If you need to sample multiple outputs at once, with one sample selector, or you need an auto-sampled input (for example: a frequency input), you should use [[Config:SampleSelector#lola-voronoi|LOLA-Voronoi]]. It is the only sample selector with fully integrated and optimized support for these features.</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* If you need to sample multiple outputs at once, with one sample selector, or you need an auto-sampled input (for example: a frequency input), you should use [[Config:SampleSelector#lola-voronoi|LOLA-Voronoi]]. It is the only sample selector with fully integrated and optimized support for these features.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
</table>Adminhttp://sumowiki.intec.ugent.be/index.php?title=General_guidelines&diff=4993&oldid=prevAdmin: /* Sample Selectors */2009-10-02T14:49:30Z<p><span dir="auto"><span class="autocomment">Sample Selectors</span></span></p>
<table class="diff diff-contentalign-left diff-editfont-monospace" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 14:49, 2 October 2009</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l17" >Line 17:</td>
<td colspan="2" class="diff-lineno">Line 17:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Sample Selectors ==</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Sample Selectors ==</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'>−</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>The default [[SampleSelector|SampleSelector]] is the [[SampleSelector#<del class="diffchange diffchange-inline">LOLASampleSelector</del>|<del class="diffchange diffchange-inline">LOLASampleSelector</del>]] combined with the [[SampleSelector#<del class="diffchange diffchange-inline">ErrorSampleSelector</del>|<del class="diffchange diffchange-inline">ErrorSampleSelector</del>]], with a weight of 0.<del class="diffchange diffchange-inline">8 </del>for LOLA and 0.<del class="diffchange diffchange-inline">2 </del>for error. This is a very robust sample selector, capable of dealing with most situations. There are, however, some cases in which it is advisable to choose a different one:</div></td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>The default [[<ins class="diffchange diffchange-inline">Config:</ins>SampleSelector|SampleSelector]] is the [[<ins class="diffchange diffchange-inline">Config:</ins>SampleSelector#<ins class="diffchange diffchange-inline">lola-voronoi</ins>|<ins class="diffchange diffchange-inline">LOLA-Voronoi sample selector</ins>]] combined with the [[<ins class="diffchange diffchange-inline">Config:</ins>SampleSelector#<ins class="diffchange diffchange-inline">error</ins>|<ins class="diffchange diffchange-inline">error-based sample selector</ins>]], with a weight of 0.<ins class="diffchange diffchange-inline">7 </ins>for LOLA and 0.<ins class="diffchange diffchange-inline">3 </ins>for error. This is a very robust sample selector, capable of dealing with most situations. There are, however, some cases in which it is advisable to choose a different one:</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'>−</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>* '''Large-scale problems (1000+ samples):''' <del class="diffchange diffchange-inline">The GradientSampleSelector</del>'s time complexity is O(n²) to the number of samples n, so for large-scale experiments in which many samples are taken, <del class="diffchange diffchange-inline">the GradientSampleSelector becomes </del>quite slow. Depending on the time it takes to perform one simulation, this may or may not be a problem. If it takes a long time to perform one simulation, the cost for selecting new samples with <del class="diffchange diffchange-inline">the GradientSampleSelector </del>might still be negligible.</div></td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>* '''Large-scale problems (1000+ samples):''' <ins class="diffchange diffchange-inline">LOLA-Voronoi</ins>'s time complexity is O(n²) to the number of samples n, so for large-scale experiments in which many samples are taken, <ins class="diffchange diffchange-inline">LOLA-Voronoibecomes </ins>quite slow. Depending on the time it takes to perform one simulation, this may or may not be a problem. If it takes a long time to perform one simulation, the cost for selecting new samples with <ins class="diffchange diffchange-inline">LOLA-Voronoi </ins>might still be negligible<ins class="diffchange diffchange-inline">. If, however, you need a quicker sample selector, it is advized to use [[Config:SampleSelector#voronoi|voronoi]] or [[Config:SampleSelector#error|error]] instead</ins>.</div></td></tr>
<tr><td class='diff-marker'>−</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>* '''Rational modeler:''' Benchmarks have shown that the gain of <del class="diffchange diffchange-inline">using the LOLASampleSelector </del>over the [[SampleSelector#ErrorSampleSelector|ErrorSampleSelector]] when using global approximation methods (mainly rational/polynomial) is pretty much zero. It is therefore advisable to use the (much faster) <del class="diffchange diffchange-inline">ErrorSampleSelector </del>when using the Rational modeler. This can be done by changing the weights in default.xml to 1.0 for error and 0.0 for LOLA.</div></td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>* '''Rational modeler:''' Benchmarks have shown that the gain of <ins class="diffchange diffchange-inline">LOLA-Voronoi </ins>over the [[SampleSelector#ErrorSampleSelector|ErrorSampleSelector]] when using global approximation methods (mainly rational/polynomial) is pretty much zero. It is therefore advisable to use the (much faster) <ins class="diffchange diffchange-inline">[[Config:SampleSelector#error|error-based sample selector]] </ins>when using the Rational modeler. This can be done by changing the weights in default.xml to 1.0 for error and 0.0 for LOLA<ins class="diffchange diffchange-inline">.</ins></div></td></tr>
<tr><td colspan="2"> </td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins class="diffchange diffchange-inline">* If you need to sample multiple outputs at once, with one sample selector, or you need an auto-sampled input (for example: a frequency input), you should use [[Config:SampleSelector#lola-voronoi|LOLA-Voronoi]]. It is the only sample selector with fully integrated and optimized support for these features</ins>.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'>−</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>When using the <del class="diffchange diffchange-inline">ErrorSampleSelector instead of the LOLASampleSelector</del>, it is always a good idea to combine it with the [[SampleSelector#<del class="diffchange diffchange-inline">DensitySampleSelector</del>|<del class="diffchange diffchange-inline">DensitySampleSelector</del>]], to combat stability/robustness issues the <del class="diffchange diffchange-inline">ErrorSampleSelector </del>often causes. It is a good idea to select about 60% of the samples with <del class="diffchange diffchange-inline">the ErrorSampleSelector</del>, and 40% with the <del class="diffchange diffchange-inline">DensitySampleSelector</del>. This will ensure that at least the entire design space is covered to a certain degree. This additional sample selector is NOT necessary when using the <del class="diffchange diffchange-inline">LOLASampleSelector</del>.</div></td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>When using the <ins class="diffchange diffchange-inline">[[Config:SampleSelector#error|error-based sample selector]] separately</ins>, it is always a good idea to combine it with the [[<ins class="diffchange diffchange-inline">Config:</ins>SampleSelector#<ins class="diffchange diffchange-inline">voronoi</ins>|<ins class="diffchange diffchange-inline">voronoi</ins>]], to combat stability/robustness issues the <ins class="diffchange diffchange-inline">error-based sample selector </ins>often causes. It is a good idea to select about 60% of the samples with <ins class="diffchange diffchange-inline">error</ins>, and 40% with the <ins class="diffchange diffchange-inline">voronoi</ins>. This will ensure that at least the entire design space is covered to a certain degree. This additional sample selector is NOT necessary when using <ins class="diffchange diffchange-inline">LOLA-Voronoi. To combine sample selectors, create a CombinedSampleSelector. See </ins>the <ins class="diffchange diffchange-inline">[[Config:SampleSelector#default|default sample selector]] for an example</ins>.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Adaptive Model Builders ==</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Adaptive Model Builders ==</div></td></tr>
</table>Adminhttp://sumowiki.intec.ugent.be/index.php?title=General_guidelines&diff=4758&oldid=prevDgorissen: /* Adaptive Model Builders */2009-03-26T15:23:32Z<p><span dir="auto"><span class="autocomment">Adaptive Model Builders</span></span></p>
<table class="diff diff-contentalign-left diff-editfont-monospace" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 15:23, 26 March 2009</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l40" >Line 40:</td>
<td colspan="2" class="diff-lineno">Line 40:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div># The ANN models generally perfom very well across all problems but are very slow to use. Also if the function is uniformly rugged the Kriging/RBF/... models will give a better fit with much less points (eg. ackley function).</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div># The ANN models generally perfom very well across all problems but are very slow to use. Also if the function is uniformly rugged the Kriging/RBF/... models will give a better fit with much less points (eg. ackley function).</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div># The FANN and NANN models are much faster than the ANN models, but usually the accuracy of the ANN models is much better</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div># The FANN and NANN models are much faster than the ANN models, but usually the accuracy of the ANN models is much better</div></td></tr>
<tr><td colspan="2"> </td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;"></ins></div></td></tr>
<tr><td colspan="2"> </td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">Finally, a related question is, which model builder variant should I use (e.g., svmsim, svmga, svmps, svnoptim, etc). The best optimization algorithm to use will usually depend on how many model parameters you have. For example, since SVM models only have 2 or 3 parameters most algorithms do well and you wont see that much difference. On the other hand, if you are fitting a 5D Kriging model (thus you have at least 5 model parameters to optimize) you will most likely see better performance using the GA or PSO versions over for example the pattern search or gradient descent versions.</ins></div></td></tr>
<tr><td colspan="2"> </td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;"></ins></div></td></tr>
<tr><td colspan="2"> </td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">However, our general experience is that it does not make that much of a difference (outside the obvious extremes like gradient descent vs GA). Only if data is really expensive and you want to be sure of the best model with least samples should you really start worrying about this.</ins></div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>'''Note this is just some very rough intuition gained from our experience with different datasets, your mileage may vary! If you have any suggestions [[Contact|let us know]]'''</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>'''Note this is just some very rough intuition gained from our experience with different datasets, your mileage may vary! If you have any suggestions [[Contact|let us know]]'''</div></td></tr>
</table>Dgorissenhttp://sumowiki.intec.ugent.be/index.php?title=General_guidelines&diff=4711&oldid=prevDgorissen: /* Measures */2009-03-10T14:59:54Z<p><span dir="auto"><span class="autocomment">Measures</span></span></p>
<table class="diff diff-contentalign-left diff-editfont-monospace" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 14:59, 10 March 2009</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l13" >Line 13:</td>
<td colspan="2" class="diff-lineno">Line 13:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''Rational modeler:''' When using Rational modeler, you might want to manually add a [[Measures#MinMax| MinMax]] measure (if you got a rough estimate of the minimum and maximum values for your outputs) and use it together with CrossValidation. By adding the MinMax measure, you eliminate models which have poles in the design space, because these poles always break the minimum and maximum bounds. This usually results in better models and quicker convergence.</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''Rational modeler:''' When using Rational modeler, you might want to manually add a [[Measures#MinMax| MinMax]] measure (if you got a rough estimate of the minimum and maximum values for your outputs) and use it together with CrossValidation. By adding the MinMax measure, you eliminate models which have poles in the design space, because these poles always break the minimum and maximum bounds. This usually results in better models and quicker convergence.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'>−</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>Selecting a good Measure '''is a very important''' part of the modeling process! <del class="diffchange diffchange-inline"> </del>Make sure you also read [[Multi-Objective Modeling]].</div></td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>Selecting a good Measure '''is a very important''' part of the modeling process! <ins class="diffchange diffchange-inline">It is CRUCIAL that you think well about this. </ins>Make sure you also read [[Multi-Objective Modeling]].</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Sample Selectors ==</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Sample Selectors ==</div></td></tr>
</table>Dgorissenhttp://sumowiki.intec.ugent.be/index.php?title=General_guidelines&diff=4710&oldid=prevDgorissen: /* Measures */2009-03-10T14:59:26Z<p><span dir="auto"><span class="autocomment">Measures</span></span></p>
<table class="diff diff-contentalign-left diff-editfont-monospace" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 14:59, 10 March 2009</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l9" >Line 9:</td>
<td colspan="2" class="diff-lineno">Line 9:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>The default [[Measures| Measure]] is [[Measures#CrossValidation| CrossValidation]]. Even though this is a very good, accurate, overall measure, there are some considerations to make in the following cases:</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>The default [[Measures| Measure]] is [[Measures#CrossValidation| CrossValidation]]. Even though this is a very good, accurate, overall measure, there are some considerations to make in the following cases:</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'>−</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #ffe49c; vertical-align: top; white-space: pre-wrap;"><div>* '''Expensive modelers (ann):''' If it is relatively expensive to train a model (for example, with neural networks), CrossValidation is also very slow, because it has to train a model for each fold (which is 5 by default). If modeling takes too long, you might want to use a faster alternative, such as [[Measures#ValidationSet|ValidationSet]].</div></td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div>* '''Expensive modelers (ann):''' If it is relatively expensive to train a model (for example, with neural networks), CrossValidation is also very slow, because it has to train a model for each fold (which is 5 by default). If modeling takes too long, you might want to use a faster alternative, such as [[Measures#ValidationSet|ValidationSet<ins class="diffchange diffchange-inline">]] or a combination of [[Measures#SampleError|SampleError]] and [[Measures#LRMMeasure|LRMMeasure</ins>]].</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''ErrorSampleSelector:''' CrossValidation might give a biased result when combined with the [[SampleSelector#ErrorSampleSelector|ErrorSampleSelector]]. This is because the ErrorSampleSelector tends to cluster samples around one point, which will result in very accurate surrogate models for all the points in this cluster (and thus good results with CrossValidation ). So when using CrossValidation and ErrorSampleSelector together, keep in mind that the real accuracy might be slightly lower than the estimated one.</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''ErrorSampleSelector:''' CrossValidation might give a biased result when combined with the [[SampleSelector#ErrorSampleSelector|ErrorSampleSelector]]. This is because the ErrorSampleSelector tends to cluster samples around one point, which will result in very accurate surrogate models for all the points in this cluster (and thus good results with CrossValidation ). So when using CrossValidation and ErrorSampleSelector together, keep in mind that the real accuracy might be slightly lower than the estimated one.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''Rational modeler:''' When using Rational modeler, you might want to manually add a [[Measures#MinMax| MinMax]] measure (if you got a rough estimate of the minimum and maximum values for your outputs) and use it together with CrossValidation. By adding the MinMax measure, you eliminate models which have poles in the design space, because these poles always break the minimum and maximum bounds. This usually results in better models and quicker convergence.</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''Rational modeler:''' When using Rational modeler, you might want to manually add a [[Measures#MinMax| MinMax]] measure (if you got a rough estimate of the minimum and maximum values for your outputs) and use it together with CrossValidation. By adding the MinMax measure, you eliminate models which have poles in the design space, because these poles always break the minimum and maximum bounds. This usually results in better models and quicker convergence.</div></td></tr>
</table>Dgorissenhttp://sumowiki.intec.ugent.be/index.php?title=General_guidelines&diff=4709&oldid=prevDgorissen: /* Measures */2009-03-10T14:58:19Z<p><span dir="auto"><span class="autocomment">Measures</span></span></p>
<table class="diff diff-contentalign-left diff-editfont-monospace" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #202122; text-align: center;">Revision as of 14:58, 10 March 2009</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l12" >Line 12:</td>
<td colspan="2" class="diff-lineno">Line 12:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''ErrorSampleSelector:''' CrossValidation might give a biased result when combined with the [[SampleSelector#ErrorSampleSelector|ErrorSampleSelector]]. This is because the ErrorSampleSelector tends to cluster samples around one point, which will result in very accurate surrogate models for all the points in this cluster (and thus good results with CrossValidation ). So when using CrossValidation and ErrorSampleSelector together, keep in mind that the real accuracy might be slightly lower than the estimated one.</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''ErrorSampleSelector:''' CrossValidation might give a biased result when combined with the [[SampleSelector#ErrorSampleSelector|ErrorSampleSelector]]. This is because the ErrorSampleSelector tends to cluster samples around one point, which will result in very accurate surrogate models for all the points in this cluster (and thus good results with CrossValidation ). So when using CrossValidation and ErrorSampleSelector together, keep in mind that the real accuracy might be slightly lower than the estimated one.</div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''Rational modeler:''' When using Rational modeler, you might want to manually add a [[Measures#MinMax| MinMax]] measure (if you got a rough estimate of the minimum and maximum values for your outputs) and use it together with CrossValidation. By adding the MinMax measure, you eliminate models which have poles in the design space, because these poles always break the minimum and maximum bounds. This usually results in better models and quicker convergence.</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>* '''Rational modeler:''' When using Rational modeler, you might want to manually add a [[Measures#MinMax| MinMax]] measure (if you got a rough estimate of the minimum and maximum values for your outputs) and use it together with CrossValidation. By adding the MinMax measure, you eliminate models which have poles in the design space, because these poles always break the minimum and maximum bounds. This usually results in better models and quicker convergence.</div></td></tr>
<tr><td colspan="2"> </td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;"></ins></div></td></tr>
<tr><td colspan="2"> </td><td class='diff-marker'>+</td><td style="color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">Selecting a good Measure '''is a very important''' part of the modeling process! Make sure you also read [[Multi-Objective Modeling]].</ins></div></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Sample Selectors ==</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #202122; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>== Sample Selectors ==</div></td></tr>
</table>Dgorissen