Exp::Exp()
{
STANDARD_CONSTRUCTOR()
setMean(10.0);
std::cout << "Exp::Exp() called\n";
}
void descriptorExtractor::loadNet(const String& model_file, const String& trained_file, const String& mean_file)
{
if (net_set)
{
/* Load the network. */
convnet = new Net<float>(model_file, TEST);
convnet->CopyTrainedLayersFrom(trained_file);
if (convnet->num_inputs() != 1)
std::cout << "Network should have exactly one input." << std::endl;
if (convnet->num_outputs() != 1)
std::cout << "Network should have exactly one output." << std::endl;
Blob<float>* input_layer = convnet->input_blobs()[0];
num_channels = input_layer->channels();
if (num_channels != 3 && num_channels != 1)
std::cout << "Input layer should have 1 or 3 channels." << std::endl;
input_geometry = cv::Size(input_layer->width(), input_layer->height());
/* Load the binaryproto mean file. */
if (!mean_file.empty())
{
setMean(mean_file);
net_ready = 2;
}
else
{
net_ready = 1;
}
}
else
{
std::cout << "Error: Net is not set properly in advance using construtor." << std::endl;
}
};
NormalDistribution::NormalDistribution(double mean, double standardDeviation)
: UncertaintyDescription(boost::shared_ptr<detail::UncertaintyDescription_Impl>(
new detail::UncertaintyDescription_Impl(NormalDistribution::type())))
{
setMean(mean);
setStandardDeviation(standardDeviation);
}
void CybGaussianNaiveBayes::initData()
{
for(int i=0; i < this->getVariablesNumber();i++) //X, Y, Z
{
float mean = 0; //mean variable
for(int j=0; j < this->getData()->size(); j++) //from 0 to size do
mean += this->getData()->pos(j)->operator[](i); //add value to mean
setMean(mean/this->getData()->size(), i); //divide mean by its size
}
for(int i=0; i < this->getVariablesNumber();i++) //X, Y, Z
{
float variance = 0; //variance variable
for(int j=0; j<this->getData()->size(); j++) //from 0 to size do
{
//variancia = variancia + pow((float_aux[j] - media), 2);
variance += pow((this->getData()->pos(j)->operator[](i) - this->getMean(i)), 2); //add to variance
}
setVariance(variance/this->getData()->size(), i); //divide variance by its size
}
}
Moment& Moment::operator=(const Moment& source)
{
if (this != &source) // protect against invalid self-assignment
{
m_numStates = source.m_numStates;
setMean(source.mean());
setCovariance(source.covariance());
}
// by convention, always return *this
return *this;
}
void Gaussian::configure(const std::string& parameters) {
if (parameters.empty()) return;
std::vector<std::string> values = Op::split(parameters, " ");
std::size_t required = 2;
if (values.size() < required) {
std::ostringstream ex;
ex << "[configuration error] term <" << className() << ">"
<< " requires <" << required << "> parameters";
throw fl::Exception(ex.str(), FL_AT);
}
setMean(Op::toScalar(values.at(0)));
setStandardDeviation(Op::toScalar(values.at(1)));
if (values.size() > required)
setHeight(Op::toScalar(values.at(required)));
}
bool Exp::setSlotMean(const oe::base::Number* const mean)
{
if (mean != nullptr)
setMean(mean->getDouble());
return true;
}
