ADD: adjusted and added new learning methods

src/examples/simulator.cpp

@@ -12,21 +12,22 @@
 #include <assert.h>
 
 #include "4neuro.h"
-#include "../LearningMethods/KallmanFilter.h"
+#include "../LearningMethods/KalmanFilter.h"
+#include "../LearningMethods/RandomSolution.h"
 
 int main(int argc, char** argv) {
 
     bool normalize_data = true;
     double prec = 1e-9;
-    double prec_lm = 1e-15;
+    double prec_lm = 1e-9;
     int restart_interval = 500;
     int max_n_iters_gradient = 10000;
     int max_n_iters_gradient_lm = 10000;
 
     int max_n_iters_swarm = 20;
     int n_particles_swarm = 200;
-    int batch_size = 0;
-    int max_number_of_cycles = 5;
+    unsigned long batch_size = 0;
+    int max_number_of_cycles = 1;
     try {
         /* PHASE 1 - TRAINING DATA LOADING, NETWORK ASSEMBLY AND PARTICLE SWARM OPTIMIZATION */
         l4n::CSVReader reader1("/home/fluffymoo/Dropbox/data_BACK_RH_1.csv", ";", true);  // File, separator, skip 1st line
@@ -43,7 +44,7 @@ int main(int argc, char** argv) {
         }
 
         /* Numbers of neurons in layers (including input and output layers) */
-        std::vector<unsigned int> neuron_numbers_in_layers = { 1, 6, 6, 1 };
+        std::vector<unsigned int> neuron_numbers_in_layers = { 1, 2, 2, 1 };
 
         /* Fully connected feed-forward network with linear activation functions for input and output */
         /* layers and the specified activation fns for the hidden ones (each entry = layer)*/
@@ -84,17 +85,19 @@ int main(int argc, char** argv) {
         // 1) Threshold for the successful ending of the optimization - deviation from minima
         // 2) Number of iterations to reset step size to tolerance/10.0
         // 3) Maximal number of iterations - optimization will stop after that, even if not converged
+        l4n::RandomSolution rnd;
         l4n::GradientDescent gs_(prec, restart_interval, max_n_iters_gradient, batch_size);
         l4n::GradientDescentBB gs(prec, restart_interval, max_n_iters_gradient, batch_size);
         l4n::GradientDescentSingleItem gs_si(prec, 0, 5000);//TODO needs improvement
-        l4n::LevenbergMarquardt leven(max_n_iters_gradient_lm, prec_lm);
+        l4n::LevenbergMarquardt leven(max_n_iters_gradient_lm, batch_size, prec_lm );
+        l4n::KalmanFilter kalman(prec, max_n_iters_gradient_lm, batch_size);
         l4n::LearningSequence learning_sequence( 1e-6, max_number_of_cycles );
-        l4n::KallmanFilter kallman(prec, max_n_iters_gradient_lm);
 
+        learning_sequence.add_learning_method( &rnd );
 //        learning_sequence.add_learning_method( &ps );
 //        learning_sequence.add_learning_method( &gs );
-//        learning_sequence.add_learning_method( &leven );
-        learning_sequence.add_learning_method( &kallman );
+        learning_sequence.add_learning_method( &leven );
+//        learning_sequence.add_learning_method( &kalman );
 //        learning_sequence.add_learning_method( &gs_ );
 //        learning_sequence.add_learning_method( &gs_si );
 //        learning_sequence.add_learning_method( &gs );