void split(std::vector<DataSetView>& groups, DataSet& dataset,
int numberOfGroups, bool shuffling)
{
OPENANN_CHECK(numberOfGroups > 1);
std::vector<int> indices;
indices.reserve(dataset.samples());
groups.reserve(numberOfGroups);
for(int i = 0; i < dataset.samples(); ++i)
indices.push_back(i);
int samplesPerGroup = std::floor(dataset.samples() / numberOfGroups + 0.5);
if(shuffling)
std::random_shuffle(indices.begin(), indices.end());
for(int i = 0; i < numberOfGroups; ++i)
{
std::vector<int>::iterator it = indices.begin() + i * samplesPerGroup;
if(i < numberOfGroups - 1)
groups.push_back(DataSetView(dataset, it, it + samplesPerGroup));
else
groups.push_back(DataSetView(dataset, it, indices.end()));
}
}
void split(std::vector<DataSetView>& groups, DataSet& dataset, double ratio,
bool shuffling)
{
OPENANN_CHECK_WITHIN(ratio, 0.0, 1.0);
std::vector<int> indices;
indices.reserve(dataset.samples());
groups.reserve(2);
for(int i = 0; i < dataset.samples(); ++i)
indices.push_back(i);
int samples = std::ceil(ratio * dataset.samples());
if(shuffling)
std::random_shuffle(indices.begin(), indices.end());
groups.push_back(DataSetView(dataset, indices.begin(), indices.begin() + samples));
groups.push_back(DataSetView(dataset, indices.begin() + samples, indices.end()));
}
DataSetView sample(DataSet& dataSet, double fraction, bool replacement)
{
std::vector<int> indices;
int samples = std::ceil(dataSet.samples() * fraction);
indices.reserve(samples);
RandomNumberGenerator rng;
if(replacement)
for(int n = 0; n < samples; n++)
indices[n] = rng.generateIndex(dataSet.samples());
else
rng.generateIndices(dataSet.samples(), indices, false);
return DataSetView(dataSet, indices.begin(), indices.begin() + samples);
}
