added oversammpling method, reverted random device

EnvTest/src/Modell/rejection_sampling/impl.cpp

@@ -18,7 +18,7 @@ void rejection_sampler::setModell(imperfect_virtualage_likelihood modell)
 
 std::vector<double> rejection_sampler::solve(simulation_params param)
 {
-	std::random_device gen;
+	std::default_random_engine gen;
 	auto U = std::bind(std::uniform_real_distribution(0., 1.),gen);
 	auto i = param.sample_size;
 	L_.set_data(failures_);
@@ -74,3 +74,12 @@ std::vector<double> rejection_sampler::estimate_result_()
 	std::for_each(avg.begin(), avg.end(), [accepted_number](double& param) {param = param / double(accepted_number); });
 	return avg;
 }
+
+void rejection_sampler::oversample(const unsigned int rate)
+{
+	for (auto outer_incrementer =  raw_result_.size()-1;outer_incrementer --> 0;)
+	{
+		for (auto i = rate; i-- > 0;)
+			raw_result_.push_back(raw_result_[outer_incrementer]);
+	}
+}

EnvTest/src/Modell/rejection_sampling/impl.h

@@ -30,4 +30,5 @@ class rejection_sampler
 	//implement modell here
 	std::vector<sample_result> raw_result_;
 	std::vector<double> estimate_result_();
+	void oversample(const unsigned int rate);
 };