int main(int argc, char** argv)
{
Col<uint32_t> a;
a<<1<<2<<4<<3<<1<<6<<7<<2<<4;
cout<<a.t()<<endl;
cout<<(a==1).t()<<endl;
cout<<sum(a==1)<<endl;
mat A;
A.randn(10,2);
A.rows(0,4) += 4;
cout<<A<<endl;
Col<uint32_t> b=a;
b.resize(b.n_elem+1);
b(b.n_elem-1)=99;
cout<<a.t()<<endl;
cout<<b.t()<<endl;
colvec c(10);
c.zeros();
c(4)=math::inf();
c(5)=math::nan();
cout<<c<<endl;
cout<<is_finite(c(3))<<endl;
cout<<is_finite(c(4))<<endl;
cout<<is_finite(c(5))<<endl;
cout<<is_finite(c)<<endl;
colvec d(2);
d.ones();
c.insert_rows(c.n_elem,d);
cout<<c<<endl;
colvec l(2);
l << -8.0 << -16.0;
cout<<l.t()<<endl;
cout<<exp(l).t()<<endl;
return 0;
}
inline
bool
hess
(
Mat<typename T1::elem_type>& U,
Mat<typename T1::elem_type>& H,
const Base<typename T1::elem_type,T1>& X,
const typename arma_blas_type_only<typename T1::elem_type>::result* junk = 0
)
{
arma_extra_debug_sigprint();
arma_ignore(junk);
arma_debug_check( void_ptr(&U) == void_ptr(&H), "hess(): 'U' is an alias of 'H'" );
typedef typename T1::elem_type eT;
Col<eT> tao;
const bool status = auxlib::hess(H, X.get_ref(), tao);
if(H.n_rows == 0)
{
U.reset();
}
else
if(H.n_rows == 1)
{
U.ones(1, 1);
}
else
if(H.n_rows == 2)
{
U.eye(2, 2);
}
else
{
U.eye(size(H));
Col<eT> v;
for(uword i=0; i < H.n_rows-2; ++i)
{
// TODO: generate v in a more efficient manner;
// TODO: the .ones() operation is an overkill, as most of v is overwritten afterwards
v.ones(H.n_rows-i-1);
v(span(1, H.n_rows-i-2)) = H(span(i+2, H.n_rows-1), i);
U(span::all, span(i+1, H.n_rows-1)) -= tao(i) * (U(span::all, span(i+1, H.n_rows-1)) * v * v.t());
}
U(span::all, H.n_rows-1) = U(span::all, H.n_rows-1) * (eT(1) - tao(H.n_rows-2));
for(uword i=0; i < H.n_rows-2; ++i)
{
H(span(i+2, H.n_rows-1), i).zeros();
}
}
if(status == false)
{
U.soft_reset();
H.soft_reset();
arma_debug_warn("hess(): decomposition failed");
}
return status;
}
int main(int argc, char** argv)
{
CLI::ParseCommandLine(argc, argv);
// Get reference dataset filename.
const string trainingDataFilename = CLI::GetParam<string>("train_file");
mat trainingData;
data::Load(trainingDataFilename, trainingData, true);
const string labelsFilename = CLI::GetParam<string>("labels_file");
// Load labels.
mat labelsIn;
// Did the user pass in labels?
if (CLI::HasParam("labels_file"))
{
// Load labels.
const string labelsFilename = CLI::GetParam<string>("labels_file");
data::Load(labelsFilename, labelsIn, true);
}
else
{
// Use the last row of the training data as the labels.
Log::Info << "Using the last dimension of training set as labels." << endl;
labelsIn = trainingData.row(trainingData.n_rows - 1).t();
trainingData.shed_row(trainingData.n_rows - 1);
}
// Do the labels need to be transposed?
if (labelsIn.n_rows == 1)
{
labelsIn = labelsIn.t();
}
// Normalize the labels.
Col<size_t> labels;
vec mappings;
data::NormalizeLabels(labelsIn.unsafe_col(0), labels, mappings);
// Load test dataset.
const string testingDataFilename = CLI::GetParam<string>("test_file");
mat testingData;
data::Load(testingDataFilename, testingData, true);
if (testingData.n_rows != trainingData.n_rows)
{
Log::Fatal << "Test data dimensionality (" << testingData.n_rows << ") "
<< "must be the same as training data (" << trainingData.n_rows - 1
<< ")!" << std::endl;
}
int iterations = CLI::GetParam<int>("iterations");
// Create and train the classifier.
Timer::Start("Training");
Perceptron<> p(trainingData, labels.t(), iterations);
Timer::Stop("Training");
// Time the running of the Perceptron Classifier.
Row<size_t> predictedLabels(testingData.n_cols);
Timer::Start("Testing");
p.Classify(testingData, predictedLabels);
Timer::Stop("Testing");
// Un-normalize labels to prepare output.
vec results;
data::RevertLabels(predictedLabels.t(), mappings, results);
// saving the predictedLabels in the transposed manner in output
const string outputFilename = CLI::GetParam<string>("output");
data::Save(outputFilename, results, true, false);
}
int main(int argc, char* argv[])
{
CLI::ParseCommandLine(argc, argv);
// Check input parameters.
const string trainingDataFilename = CLI::GetParam<string>("train_file");
mat trainingData;
data::Load(trainingDataFilename, trainingData, true);
// Normalize labels.
Col<size_t> labels;
Row<size_t> labelst;
vec mappings;
// Did the user pass in labels?
const string labelsFilename = CLI::GetParam<string>("labels_file");
if (labelsFilename != "")
{
// Load labels.
mat rawLabels;
data::Load(labelsFilename, rawLabels, true, false);
// Do the labels need to be transposed?
if (rawLabels.n_rows == 1)
rawLabels = rawLabels.t();
data::NormalizeLabels(rawLabels.unsafe_col(0), labels, mappings);
}
else
{
// Use the last row of the training data as the labels.
Log::Info << "Using last dimension of training data as training labels."
<< std::endl;
vec rawLabels = trans(trainingData.row(trainingData.n_rows - 1));
data::NormalizeLabels(rawLabels, labels, mappings);
// Remove the label row.
trainingData.shed_row(trainingData.n_rows - 1);
}
labelst = labels.t();
const string testingDataFilename = CLI::GetParam<std::string>("test_file");
mat testingData;
data::Load(testingDataFilename, testingData, true);
if (testingData.n_rows != trainingData.n_rows)
Log::Fatal << "Test data dimensionality (" << testingData.n_rows << ") "
<< "must be the same as training data (" << trainingData.n_rows
<< ")!" << std::endl;
const bool incrementalVariance = CLI::HasParam("incremental_variance");
// Create and train the classifier.
Timer::Start("training");
NaiveBayesClassifier<> nbc(trainingData, labelst, mappings.n_elem,
incrementalVariance);
Timer::Stop("training");
// Time the running of the Naive Bayes Classifier.
Row<size_t> results;
Timer::Start("testing");
nbc.Classify(testingData, results);
Timer::Stop("testing");
// Un-normalize labels to prepare output.
vec rawResults;
data::RevertLabels(results.t(), mappings, rawResults);
// Output results. Transpose: one result per line.
const string outputFilename = CLI::GetParam<string>("output");
data::Save(outputFilename, rawResults, true, false);
}
