saberbud/q3DMASC
1
0
Fork 0
mirror of https://github.com/dgirardeau/q3DMASC.git synced 2026-08-30 00:50:49 +08:00
Code Issues Packages Projects Releases Wiki Activity

Option to evaluation a classifier with a point cloud

Browse Source
This commit is contained in:
Daniel Girardeau-Montaut
2019-03-24 23:16:55 +01:00
parent 35bfc92dbf
commit 4907e68ae2
4 changed files with 60 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
q3DMASC.cpp
+39 -6
View File
@@ -158,6 +158,13 @@ void q3DMASCPlugin::doClassifyAction()
return;
}
if (clouds.contains("TEST"))
{
//remove the test cloud (if any)
delete clouds["TEST"];
clouds.remove("TEST");
}
//the 'main cloud' is the cloud that should be classified
masc::CorePoints corePoints;
corePoints.origin = corePoints.cloud = clouds[mainCloudLabel];
@@ -317,6 +324,7 @@ void q3DMASCPlugin::doTrainAction()
delete group;
return;
}
if (corePoints.cloud != corePoints.origin)
{
//auto-hide the other clouds
@@ -342,6 +350,7 @@ void q3DMASCPlugin::doTrainAction()
corePoints.cloud->setName(QString("Core points (%1)").arg(corePointsName));
group->addChild(corePoints.cloud);
}
if (group->getChildrenNumber() != 0)
{
m_app->addToDB(group);
@@ -353,10 +362,19 @@ void q3DMASCPlugin::doTrainAction()
group = nullptr;
}
//test role
ccPointCloud* testCloud = nullptr;
if (loadedClouds.contains("TEST"))
{
testCloud = loadedClouds["TEST"];
loadedClouds.remove("TEST");
}
//train / test subsets
QScopedPointer<CCLib::ReferenceCloud> trainSubset(new CCLib::ReferenceCloud(corePoints.cloud));
QScopedPointer<CCLib::ReferenceCloud> testSubset(new CCLib::ReferenceCloud(corePoints.cloud));
float previousTrainSubsetRatio = -1.0f;
float previousTestSubsetRatio = -1.0f;
SFCollector generatedScalarFields;
@@ -418,15 +436,30 @@ void q3DMASCPlugin::doTrainAction()
s_params.rt.maxTreeCount = trainDlg.maxTreeCountSpinBox->value();
s_params.rt.activeVarCount = trainDlg.activeVarCountSpinBox->value();
s_params.rt.minSampleCount = trainDlg.minSampleCountSpinBox->value();
s_params.testDataRatio = trainDlg.testDataRatioSpinBox->value() / 100.0f;
if (s_params.testDataRatio < 0 || s_params.testDataRatio > 0.99f)
float testDataRatio = s_params.testDataRatio = trainDlg.testDataRatioSpinBox->value() / 100.0f;
QScopedPointer<CCLib::ReferenceCloud> testSubset2;
if (testCloud)
{
m_app->dispToConsole("Test data cloud provided (ignoring test data ratio)", ccMainAppInterface::WRN_CONSOLE_MESSAGE);
testDataRatio = 0.0f;
testSubset2.reset(new CCLib::ReferenceCloud(testCloud));
if (!testSubset2->reserve(testCloud->size()))
{
m_app->dispToConsole("Not enough memory to evaluate the classifier", ccMainAppInterface::ERR_CONSOLE_MESSAGE);
generatedScalarFields.releaseAllSFs();
return;
}
testSubset2->addPointIndex(0, testCloud->size());
}
if (s_params.testDataRatio < 0.0f || s_params.testDataRatio > 0.99f)
{
assert(false);
m_app->dispToConsole("Invalid test data ratio", ccMainAppInterface::ERR_CONSOLE_MESSAGE);
}
else
{
if (previousTrainSubsetRatio != s_params.testDataRatio)
if (previousTestSubsetRatio != s_params.testDataRatio)
{
//randomly select the training points
testSubset->clear();
@@ -437,7 +470,7 @@ void q3DMASCPlugin::doTrainAction()
generatedScalarFields.releaseAllSFs();
return;
}
previousTrainSubsetRatio = s_params.testDataRatio;
previousTestSubsetRatio = s_params.testDataRatio;
}
//train the classifier
@@ -457,7 +490,7 @@ void q3DMASCPlugin::doTrainAction()
{
masc::Classifier::AccuracyMetrics metrics;
QString errorMessage;
if (!classifier.evaluate(features, testSubset.data(), metrics, errorMessage, m_app->getMainWindow()))
if (!classifier.evaluate(features, testSubset2 ? testSubset2.data() : testSubset.data(), metrics, errorMessage, m_app->getMainWindow()))
{
m_app->dispToConsole(errorMessage, ccMainAppInterface::ERR_CONSOLE_MESSAGE);
generatedScalarFields.releaseAllSFs();