The Dlib library uses the serialization API for decision_function and neural network type objects. Let's learn how to use it by implementing a real example.
First, we define the types for the neural network, regression kernel, and training sample:
using namespace dlib; using NetworkType = loss_mean_squared<fc<1, input<matrix<double>>>>; using SampleType = matrix<double, 1, 1>; using KernelType = linear_kernel<SampleType>;
Then, we generate the training data with the following code:
size_t n = 1000; std::vector<matrix<double>> x(n); std::vector<float> y(n); std::random_device rd; std::mt19937 re(rd()); std::uniform_real_distribution<float> dist(-1.5, 1.5); // generate data for (size_t i = 0; i < n; ++i) { ...
