#include "daal.h"
#include "service.h"
using namespace std;
using namespace daal;
using namespace daal::algorithms;
using namespace daal::algorithms::multinomial_naive_bayes;
const string trainDatasetFileNames[4] =
{
"../data/distributed/naivebayes_train_dense_1.csv", "../data/distributed/naivebayes_train_dense_2.csv",
"../data/distributed/naivebayes_train_dense_3.csv", "../data/distributed/naivebayes_train_dense_4.csv"
};
string testDatasetFileName = "../data/distributed/naivebayes_test_dense.csv";
const size_t nFeatures = 20;
const size_t nClasses = 20;
const size_t nBlocks = 4;
void trainModel();
void testModel();
void printResults();
training::ResultPtr trainingResult;
classifier::prediction::ResultPtr predictionResult;
NumericTablePtr testGroundTruth;
int main(int argc, char *argv[])
{
checkArguments(argc, argv, 5,
&trainDatasetFileNames[0], &trainDatasetFileNames[1],
&trainDatasetFileNames[2], &trainDatasetFileNames[3],
&testDatasetFileName);
trainModel();
testModel();
printResults();
return 0;
}
void trainModel()
{
training::Distributed<step2Master> masterAlgorithm(nClasses);
for(size_t i = 0; i < nBlocks; i++)
{
FileDataSource<CSVFeatureManager> trainDataSource(trainDatasetFileNames[i],
DataSource::notAllocateNumericTable,
DataSource::doDictionaryFromContext);
NumericTablePtr trainData(new HomogenNumericTable<>(nFeatures, 0, NumericTable::doNotAllocate));
NumericTablePtr trainGroundTruth(new HomogenNumericTable<>(1, 0, NumericTable::doNotAllocate));
NumericTablePtr mergedData(new MergedNumericTable(trainData, trainGroundTruth));
trainDataSource.loadDataBlock(mergedData.get());
training::Distributed<step1Local> localAlgorithm(nClasses);
localAlgorithm.input.set(classifier::training::data, trainData);
localAlgorithm.input.set(classifier::training::labels, trainGroundTruth);
localAlgorithm.compute();
masterAlgorithm.input.add(training::partialModels, localAlgorithm.getPartialResult());
}
masterAlgorithm.compute();
masterAlgorithm.finalizeCompute();
trainingResult = masterAlgorithm.getResult();
}
void testModel()
{
FileDataSource<CSVFeatureManager> testDataSource(testDatasetFileName,
DataSource::notAllocateNumericTable,
DataSource::doDictionaryFromContext);
NumericTablePtr testData(new HomogenNumericTable<>(nFeatures, 0, NumericTable::doNotAllocate));
testGroundTruth = NumericTablePtr(new HomogenNumericTable<>(1, 0, NumericTable::doNotAllocate));
NumericTablePtr mergedData(new MergedNumericTable(testData, testGroundTruth));
testDataSource.loadDataBlock(mergedData.get());
prediction::Batch<> algorithm(nClasses);
algorithm.input.set(classifier::prediction::data, NumericTablePtr(testData));
algorithm.input.set(classifier::prediction::model, trainingResult->get(classifier::training::model));
algorithm.compute();
predictionResult = algorithm.getResult();
}
void printResults()
{
printNumericTables<int, int>(testGroundTruth,
predictionResult->get(classifier::prediction::prediction),
"Ground truth", "Classification results",
"NaiveBayes classification results (first 20 observations):", 20);
}