void ClassifySvmSharedCommand::trainSharedAndDesignData(vector<SharedRAbundVector*> lookup) {
try {
LabeledObservationVector labeledObservationVector;
FeatureVector featureVector;
readSharedRAbundVectors(lookup, designMap, labeledObservationVector, featureVector);
SvmDataset svmDataset(labeledObservationVector, featureVector);
int evaluationFoldCount = 3;
int trainFoldCount = 5;
OutputFilter outputFilter(2);
OneVsOneMultiClassSvmTrainer t(svmDataset, evaluationFoldCount, trainFoldCount, *this, outputFilter);
KernelParameterRangeMap kernelParameterRangeMap;
getDefaultKernelParameterRangeMap(kernelParameterRangeMap);
t.train(kernelParameterRangeMap);
std::cout << "done training" << std::endl;
map<string, string> variables;
variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(sharedfile));
variables["[distance]"] = lookup[0]->getLabel();
string filename = getOutputFileName("summary", variables);
outputNames.push_back(filename);
outputTypes["summary"].push_back(filename);
m->mothurOutEndLine();
std::cout << "leaving processSharedAndDesignData" << std::endl;
}
catch (exception& e) {
m->errorOut(e, "ClassifySvmSharedCommand", "trainSharedAndDesignData");
exit(1);
}
}
void MSBuildProvider::createFiltersFile(const BuildSetup &setup, const std::string &name) {
// No filters => no need to create a filter file
if (_filters.empty())
return;
// Sort all list alphabetically
_filters.sort();
_compileFiles.sort();
_includeFiles.sort();
_otherFiles.sort();
_resourceFiles.sort();
_asmFiles.sort();
const std::string filtersFile = setup.outputDir + '/' + name + getProjectExtension() + ".filters";
std::ofstream filters(filtersFile.c_str());
if (!filters)
error("Could not open \"" + filtersFile + "\" for writing");
filters << "<?xml version=\"1.0\" encoding=\"utf-8\"?>\n"
"<Project ToolsVersion=\"" << (_version >= 12 ? _version : 4) << ".0\" xmlns=\"http://schemas.microsoft.com/developer/msbuild/2003\">\n";
// Output the list of filters
filters << "\t<ItemGroup>\n";
for (std::list<std::string>::iterator filter = _filters.begin(); filter != _filters.end(); ++filter) {
filters << "\t\t<Filter Include=\"" << *filter << "\">\n"
"\t\t\t<UniqueIdentifier>" << createUUID() << "</UniqueIdentifier>\n"
"\t\t</Filter>\n";
}
filters << "\t</ItemGroup>\n";
// Output files
outputFilter(filters, _compileFiles, "ClCompile");
outputFilter(filters, _includeFiles, "ClInclude");
outputFilter(filters, _otherFiles, "None");
outputFilter(filters, _resourceFiles, "ResourceCompile");
outputFilter(filters, _asmFiles, "CustomBuild");
filters << "</Project>";
}
void Column::Private::setColumnMode(SciDAVis::ColumnMode mode, AbstractFilter *filter)
{
if (mode == d_column_mode) return;
void * old_data = d_data;
// remark: the deletion of the old data will be done in the dtor of a command
AbstractSimpleFilter *new_in_filter, *new_out_filter;
bool filter_is_temporary; // it can also become outputFilter(), which we may not delete here
Column* temp_col = 0;
if (filter)
filter_is_temporary = false;
emit d_owner->modeAboutToChange(d_owner);
// determine the conversion filter and allocate the new data vector
switch(d_column_mode)
{
case SciDAVis::Numeric:
disconnect(static_cast<Double2StringFilter *>(d_output_filter), SIGNAL(formatChanged()),
d_owner, SLOT(notifyDisplayChange()));
switch(mode)
{
case SciDAVis::Numeric:
break;
case SciDAVis::Text:
if (!filter) {
filter = outputFilter();
filter_is_temporary = false;
}
temp_col = new Column("temp_col", *(static_cast< QVector<qreal>* >(old_data)), d_validity);
d_data = new QStringList();
d_data_type = SciDAVis::TypeQString;
break;
case SciDAVis::DateTime:
if (!filter) {
filter = new Double2DateTimeFilter();
filter_is_temporary = true;
}
temp_col = new Column("temp_col", *(static_cast< QVector<qreal>* >(old_data)), d_validity);
d_data = new QList<QDateTime>();
d_data_type = SciDAVis::TypeQDateTime;
break;
case SciDAVis::Month:
if (!filter) {
filter = new Double2MonthFilter();
filter_is_temporary = true;
}
temp_col = new Column("temp_col", *(static_cast< QVector<qreal>* >(old_data)), d_validity);
d_data = new QList<QDateTime>();
d_data_type = SciDAVis::TypeQDateTime;
break;
case SciDAVis::Day:
if (!filter) {
filter = new Double2DayOfWeekFilter();
filter_is_temporary = true;
}
temp_col = new Column("temp_col", *(static_cast< QVector<qreal>* >(old_data)), d_validity);
d_data = new QList<QDateTime>();
d_data_type = SciDAVis::TypeQDateTime;
break;
} // switch(mode)
break;
case SciDAVis::Text:
switch(mode)
{
case SciDAVis::Text:
break;
case SciDAVis::Numeric:
if (!filter) {
filter = new String2DoubleFilter();
filter_is_temporary = true;
}
temp_col = new Column("temp_col", *(static_cast< QStringList* >(old_data)), d_validity);
d_data = new QVector<double>();
d_data_type = SciDAVis::TypeDouble;
break;
case SciDAVis::DateTime:
if (!filter) {
filter = new String2DateTimeFilter();
filter_is_temporary = true;
}
temp_col = new Column("temp_col", *(static_cast< QStringList* >(old_data)), d_validity);
d_data = new QList<QDateTime>();
d_data_type = SciDAVis::TypeQDateTime;
break;
case SciDAVis::Month:
if (!filter) {
filter = new String2MonthFilter();
filter_is_temporary = true;
}
temp_col = new Column("temp_col", *(static_cast< QStringList* >(old_data)), d_validity);
d_data = new QList<QDateTime>();
d_data_type = SciDAVis::TypeQDateTime;
break;
case SciDAVis::Day:
if (!filter) {
filter = new String2DayOfWeekFilter();
filter_is_temporary = true;
}
temp_col = new Column("temp_col", *(static_cast< QStringList* >(old_data)), d_validity);
d_data = new QList<QDateTime>();
d_data_type = SciDAVis::TypeQDateTime;
break;
} // switch(mode)
break;
case SciDAVis::DateTime:
case SciDAVis::Month:
case SciDAVis::Day:
disconnect(static_cast<DateTime2StringFilter *>(d_output_filter), SIGNAL(formatChanged()),
d_owner, SLOT(notifyDisplayChange()));
switch(mode)
{
case SciDAVis::DateTime:
break;
case SciDAVis::Text:
if (!filter) {
filter = outputFilter();
filter_is_temporary = false;
}
temp_col = new Column("temp_col", *(static_cast< QList<QDateTime>* >(old_data)), d_validity);
d_data = new QStringList();
d_data_type = SciDAVis::TypeQString;
break;
case SciDAVis::Numeric:
if (!filter) {
if (d_column_mode == SciDAVis::Month)
filter = new Month2DoubleFilter();
else if (d_column_mode == SciDAVis::Day)
filter = new DayOfWeek2DoubleFilter();
else
filter = new DateTime2DoubleFilter();
filter_is_temporary = true;
}
temp_col = new Column("temp_col", *(static_cast< QList<QDateTime>* >(old_data)), d_validity);
d_data = new QVector<double>();
d_data_type = SciDAVis::TypeDouble;
break;
case SciDAVis::Month:
case SciDAVis::Day:
break;
} // switch(mode)
break;
}
// determine the new input and output filters
switch(mode)
{
case SciDAVis::Numeric:
new_in_filter = new String2DoubleFilter();
new_out_filter = new Double2StringFilter();
#ifdef LEGACY_CODE_0_2_x // TODO: in a later version this must use the new global setting method
{
#ifdef Q_OS_MAC // Mac
QSettings settings(QSettings::IniFormat,QSettings::UserScope, "SciDAVis", "SciDAVis");
#else
QSettings settings(QSettings::NativeFormat,QSettings::UserScope, "SciDAVis", "SciDAVis");
#endif
settings.beginGroup("/General");
static_cast<Double2StringFilter *>(new_out_filter)->setNumDigits(settings.value("/DecimalDigits", 14).toInt());
static_cast<Double2StringFilter *>(new_out_filter)->setNumericFormat(settings.value("/DefaultNumericFormat", 'f').toChar().toAscii());
}
#endif
connect(static_cast<Double2StringFilter *>(new_out_filter), SIGNAL(formatChanged()),
d_owner, SLOT(notifyDisplayChange()));
break;
case SciDAVis::Text:
new_in_filter = new SimpleCopyThroughFilter();
new_out_filter = new SimpleCopyThroughFilter();
break;
case SciDAVis::DateTime:
new_in_filter = new String2DateTimeFilter();
new_out_filter = new DateTime2StringFilter();
connect(static_cast<DateTime2StringFilter *>(new_out_filter), SIGNAL(formatChanged()),
d_owner, SLOT(notifyDisplayChange()));
break;
case SciDAVis::Month:
new_in_filter = new String2MonthFilter();
new_out_filter = new DateTime2StringFilter();
static_cast<DateTime2StringFilter *>(new_out_filter)->setFormat("MMMM");
connect(static_cast<DateTime2StringFilter *>(new_out_filter), SIGNAL(formatChanged()),
d_owner, SLOT(notifyDisplayChange()));
break;
case SciDAVis::Day:
new_in_filter = new String2DayOfWeekFilter();
new_out_filter = new DateTime2StringFilter();
static_cast<DateTime2StringFilter *>(new_out_filter)->setFormat("dddd");
connect(static_cast<DateTime2StringFilter *>(new_out_filter), SIGNAL(formatChanged()),
d_owner, SLOT(notifyDisplayChange()));
break;
} // switch(mode)
d_column_mode = mode;
new_in_filter->setName("InputFilter");
new_out_filter->setName("OutputFilter");
d_input_filter = new_in_filter;
d_output_filter = new_out_filter;
d_input_filter->input(0, d_owner->d_string_io);
d_output_filter->input(0, d_owner);
if (temp_col) // if temp_col == 0, only the input/output filters need to be changed
{
// copy the filtered, i.e. converted, column
filter->input(0, temp_col);
copy(filter->output(0));
delete temp_col;
if (filter_is_temporary) delete filter;
}
emit d_owner->modeChanged(d_owner);
}
void ClassifySvmSharedCommand::processSharedAndDesignData(vector<SharedRAbundVector*> lookup) {
try {
OutputFilter outputFilter(verbosity);
LabeledObservationVector labeledObservationVector;
FeatureVector featureVector;
readSharedRAbundVectors(lookup, designMap, labeledObservationVector, featureVector);
// optionally remove features with low standard deviation
if ( stdthreshold > 0.0 ) {
FeatureVector removedFeatureVector = applyStdThreshold(stdthreshold, labeledObservationVector, featureVector);
if (removedFeatureVector.size() > 0) {
std::cout << removedFeatureVector.size() << " OTUs were below the stdthreshold of " << stdthreshold << " and were removed" << std::endl;
if ( outputFilter.debug() ) {
std::cout << "the following OTUs were below the standard deviation threshold of " << stdthreshold << std::endl;
for (FeatureVector::iterator i = removedFeatureVector.begin(); i != removedFeatureVector.end(); i++) {
std::cout << " " << i->getFeatureLabel() << std::endl;
}
}
}
}
// apply [0,1] standardization
if ( transformName == "zeroone") {
std::cout << "transforming data to lie within range [0,1]" << std::endl;
transformZeroOne(labeledObservationVector);
}
else {
std::cout << "transforming data to have zero mean and unit variance" << std::endl;
transformZeroMeanUnitVariance(labeledObservationVector);
}
SvmDataset svmDataset(labeledObservationVector, featureVector);
OneVsOneMultiClassSvmTrainer trainer(svmDataset, evaluationFoldCount, trainingFoldCount, *this, outputFilter);
if ( mode == "rfe" ) {
SvmRfe svmRfe;
ParameterRange& linearKernelConstantRange = kernelParameterRangeMap["linear"]["constant"];
ParameterRange& linearKernelSmoCRange = kernelParameterRangeMap["linear"]["smoc"];
RankedFeatureList rankedFeatureList = svmRfe.getOrderedFeatureList(svmDataset, trainer, linearKernelConstantRange, linearKernelSmoCRange);
map<string, string> variables;
variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(sharedfile));
variables["[distance]"] = lookup[0]->getLabel();
string filename = getOutputFileName("summary", variables);
outputNames.push_back(filename);
outputTypes["summary"].push_back(filename);
m->mothurOutEndLine();
std::ofstream outputFile(filename.c_str());
int n = 0;
int rfeRoundCount = rankedFeatureList.front().getRank();
std::cout << "ordered features:" << std::endl;
std::cout << setw(5) << "index"
<< setw(12) << "OTU"
<< setw(5) << "rank"
<< std::endl;
outputFile << setw(5) << "index"
<< setw(12) << "OTU"
<< setw(5) << "rank"
<< std::endl;
for (RankedFeatureList::iterator i = rankedFeatureList.begin(); i != rankedFeatureList.end(); i++) {
n++;
int rank = rfeRoundCount - i->getRank() + 1;
outputFile << setw(5) << n
<< setw(12) << i->getFeature().getFeatureLabel()
<< setw(5) << rank
<< std::endl;
if ( n <= 20 ) {
std::cout << setw(5) << n
<< setw(12) << i->getFeature().getFeatureLabel()
<< setw(5) << rank
<< std::endl;
}
}
outputFile.close();
}
else {
MultiClassSVM* mcsvm = trainer.train(kernelParameterRangeMap);
map<string, string> variables;
variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(sharedfile));
variables["[distance]"] = lookup[0]->getLabel();
string filename = getOutputFileName("summary", variables);
outputNames.push_back(filename);
outputTypes["summary"].push_back(filename);
m->mothurOutEndLine();
std::ofstream outputFile(filename.c_str());
printPerformanceSummary(mcsvm, std::cout);
printPerformanceSummary(mcsvm, outputFile);
outputFile << "actual predicted" << std::endl;
for ( LabeledObservationVector::const_iterator i = labeledObservationVector.begin(); i != labeledObservationVector.end(); i++ ) {
Label actualLabel = i->getLabel();
outputFile << i->getDatasetIndex() << " " << actualLabel << " ";
try {
Label predictedLabel = mcsvm->classify(*(i->getObservation()));
outputFile << predictedLabel << std::endl;
}
catch ( MultiClassSvmClassificationTie& m ) {
outputFile << "tie" << std::endl;
std::cout << "classification tie for observation " << i->datasetIndex << " with label " << i->first << std::endl;
}
}
outputFile.close();
delete mcsvm;
}
}
catch (exception& e) {
m->errorOut(e, "ClassifySvmSharedCommand", "processSharedAndDesignData");
exit(1);
}
}
