#include "daal.h"
#include "service.h"
using namespace std;
using namespace daal;
using namespace daal::algorithms;
using namespace daal::algorithms::decision_forest::classification;
const string testDatasetFileName = "../data/batch/df_classification_model_builder_test.csv";
const size_t categoricalFeaturesIndices[] = { 2 };
const size_t nFeatures = 3;
const size_t nTrees = 3;
const size_t nClasses = 5;
void testModel(decision_forest::classification::ModelPtr& model);
decision_forest::classification::ModelPtr buildModel();
void loadData(const std::string& fileName, NumericTablePtr& pData, NumericTablePtr& pDependentVar);
int main(int argc, char *argv[])
{
checkArguments(argc, argv, 1, &testDatasetFileName);
decision_forest::classification::ModelPtr model = buildModel();
testModel(model);
return 0;
}
decision_forest::classification::ModelPtr buildModel()
{
const size_t nNodes = 3;
ModelBuilder modelBuilder(nClasses, nTrees);
ModelBuilder::TreeId tree1 = modelBuilder.createTree(nNodes);
ModelBuilder::NodeId root1 = modelBuilder.addSplitNode(tree1, ModelBuilder::noParent, 0, 0, 0.174108);
ModelBuilder::NodeId child12 = modelBuilder.addLeafNode(tree1, root1, 1, 4);
ModelBuilder::NodeId child11 = modelBuilder.addLeafNode(tree1, root1, 0, 0);
ModelBuilder::TreeId tree2 = modelBuilder.createTree(nNodes);
ModelBuilder::NodeId root2 = modelBuilder.addSplitNode(tree2, ModelBuilder::noParent, 0, 1, 0.571184);
ModelBuilder::NodeId child22 = modelBuilder.addLeafNode(tree2, root2, 1, 4);
ModelBuilder::NodeId child21 = modelBuilder.addLeafNode(tree2, root2, 0, 2);
ModelBuilder::TreeId tree3 = modelBuilder.createTree(nNodes);
ModelBuilder::NodeId root3 = modelBuilder.addSplitNode(tree3, ModelBuilder::noParent, 0, 0, 0.303995);
ModelBuilder::NodeId child32 = modelBuilder.addLeafNode(tree3, root3, 1, 4);
ModelBuilder::NodeId child31 = modelBuilder.addLeafNode(tree3, root3, 0, 2);
return modelBuilder.getModel();
}
void testModel(decision_forest::classification::ModelPtr& model)
{
NumericTablePtr testData;
NumericTablePtr testGroundTruth;
loadData(testDatasetFileName, testData, testGroundTruth);
prediction::Batch<> algorithm(nClasses);
algorithm.input.set(classifier::prediction::data, testData);
algorithm.input.set(classifier::prediction::model, model);
algorithm.compute();
classifier::prediction::ResultPtr predictionResult = algorithm.getResult();
printNumericTable(predictionResult->get(classifier::prediction::prediction),
"Decision forest prediction results (first 10 rows):", 10);
printNumericTable(testGroundTruth, "Ground truth (first 10 rows):", 10);
}
void loadData(const std::string& fileName, NumericTablePtr& pData, NumericTablePtr& pDependentVar)
{
FileDataSource<CSVFeatureManager> trainDataSource(fileName,
DataSource::notAllocateNumericTable,
DataSource::doDictionaryFromContext);
pData.reset(new HomogenNumericTable<>(nFeatures, 0, NumericTable::notAllocate));
pDependentVar.reset(new HomogenNumericTable<>(1, 0, NumericTable::notAllocate));
NumericTablePtr mergedData(new MergedNumericTable(pData, pDependentVar));
trainDataSource.loadDataBlock(mergedData.get());
NumericTableDictionaryPtr pDictionary = pData->getDictionarySharedPtr();
for(size_t i = 0, n = sizeof(categoricalFeaturesIndices) / sizeof(categoricalFeaturesIndices[0]); i < n; ++i)
(*pDictionary)[categoricalFeaturesIndices[i]].featureType = data_feature_utils::DAAL_CATEGORICAL;
}