diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
index 71271162ad7fc6fee1e3cb274eb142d517459e01..9526f91ffb6c8ab0aeed995306a14d8363005d3a 100644
--- a/src/CMakeLists.txt
+++ b/src/CMakeLists.txt
@@ -28,7 +28,7 @@ target_link_libraries(test_cases 4neuro)
add_executable(neuron_serialization_example neuron_serialization_example.cpp)
target_link_libraries(neuron_serialization_example 4neuro)
-add_executable(net_test_1 net_test_1.cpp)
+add_executable(net_test_1 net_test_1.cpp settings.h)
target_link_libraries(net_test_1 4neuro)
add_executable(net_test_2 net_test_2.cpp)
diff --git a/src/LearningMethods/ParticleSwarm.cpp b/src/LearningMethods/ParticleSwarm.cpp
index 77bdff10562285c8be25c2c2d147bff79c7f4c46..20c79490e3f7fcf25a60224a1bf0d961bb318101 100644
--- a/src/LearningMethods/ParticleSwarm.cpp
+++ b/src/LearningMethods/ParticleSwarm.cpp
@@ -42,7 +42,7 @@ Particle::Particle(ErrorFunction* ef, double *domain_bounds) {
this->domain_bounds = domain_bounds;
for(unsigned int i = 0; i < this->coordinate_dim; ++i){
- this->coordinate[i] = (rand() % 100001) / (double)100000 * (domain_bounds[2 * i + 1] - domain_bounds[2 * i]) + domain_bounds[2 * i];
+ this->coordinate[i] = (rand() % 100001) / (double)100000 + domain_bounds[2 * i] / (domain_bounds[2 * i + 1] - domain_bounds[2 * i]);
this->optimal_coordinate[i] = this->coordinate[i];
}
@@ -192,7 +192,7 @@ void ParticleSwarm::optimize( double gamma, double epsilon, double delta) {
unsigned int outer_it = 0;
Particle *particle;
double *p_min_glob = new double[this->func_dim];
- double optimal_value;
+ double optimal_value = 0.0;
std::set<Particle*> cluster; //!< Particles in a cluster
double* coords;
diff --git a/src/NetConnection/Connection.cpp b/src/NetConnection/Connection.cpp
index cd6b07918e0900f18d4658e96ab34c82b572746b..25eb8c90e42d7e32acbb0ac0f7eeaf2c47135ca4 100644
--- a/src/NetConnection/Connection.cpp
+++ b/src/NetConnection/Connection.cpp
@@ -45,7 +45,9 @@ Neuron* Connection::get_neuron_out() {
//}
void Connection::pass_signal() {
-// printf(" passing signal between neurons %d -> %d, value: %f * %f\n", this->neuron_in, this->neuron_out, this->neuron_in->get_state(), this->con->eval());
+#ifdef VERBOSE_NN_EVAL
+ printf(" passing signal between neurons %d -> %d, value: %f * %f\n", this->neuron_in, this->neuron_out, this->neuron_in->get_state(), this->con->eval());
+#endif
this->neuron_out->adjust_potential(this->neuron_in->get_state() * this->con->eval());
}
diff --git a/src/NetConnection/Connection.h b/src/NetConnection/Connection.h
index 293754cb24039fb784630f3aa284a643f96b395c..b9895f32ff29223c804c07ab886b08d06df7b6f0 100644
--- a/src/NetConnection/Connection.h
+++ b/src/NetConnection/Connection.h
@@ -10,6 +10,7 @@
#include "../Neuron/Neuron.h"
#include "ConnectionWeight.h"
+#include "../settings.h"
class Neuron;
class ConnectionWeight;
diff --git a/src/Network/NeuralNetwork.cpp b/src/Network/NeuralNetwork.cpp
index c5331f20986521a791a5d9f75e484316d2a5aad6..66246c83fd46a6e3d11e525f10c8f9b1fc5a86a6 100644
--- a/src/Network/NeuralNetwork.cpp
+++ b/src/Network/NeuralNetwork.cpp
@@ -472,8 +472,9 @@ void NeuralNetwork::eval_single(std::vector<double> &input, std::vector<double>
neuron->set_potential(input[i]);
-// printf("INPUT NEURON %d, POTENTIAL: %f\n", neuron, input[i]);
-
+#ifdef VERBOSE_NN_EVAL
+ printf("INPUT NEURON %d, POTENTIAL: %f\n", neuron, input[i]);
+#endif
++i;
}
Neuron* active_neuron;
@@ -485,8 +486,9 @@ void NeuralNetwork::eval_single(std::vector<double> &input, std::vector<double>
for(i = 0; i < active_set_size[idx1]; ++i){
active_neuron = this->active_eval_set->at(i + n * idx1);
active_neuron->activate();//computes the activation function on its potential
-// printf(" active neuron %d, potential: %f, state: %f\n", active_neuron, active_neuron->get_potential(), active_neuron->get_state());
-
+#ifdef VERBOSE_NN_EVAL
+ printf(" active neuron %d, potential: %f, state: %f\n", active_neuron, active_neuron->get_potential(), active_neuron->get_state());
+#endif
for(Connection* connection: *(active_neuron->get_connections_out())){
connection->pass_signal();
@@ -499,7 +501,9 @@ void NeuralNetwork::eval_single(std::vector<double> &input, std::vector<double>
}
}
}
-// printf("-----------\n");
+#ifdef VERBOSE_NN_EVAL
+ printf("-----------\n");
+#endif
idx1 = idx2;
idx2 = (idx1 + 1) % 2;
@@ -512,7 +516,9 @@ void NeuralNetwork::eval_single(std::vector<double> &input, std::vector<double>
output[i] = neuron->get_state();
-// printf("OUTPUT NEURON %d, VALUE: %f\n", neuron, output[i]);
+#ifdef VERBOSE_NN_EVAL
+ printf("OUTPUT NEURON %d, VALUE: %f\n", neuron, output[i]);
+#endif
++i;
}
diff --git a/src/Network/NeuralNetwork.h b/src/Network/NeuralNetwork.h
index 1bfa39a5c47e8bd1110958e1a0291ac85a42e1fe..68b558fd5751c8b0a8e1c20ae34b243492253327 100644
--- a/src/Network/NeuralNetwork.h
+++ b/src/Network/NeuralNetwork.h
@@ -13,6 +13,7 @@
#include <vector>
#include "../Neuron/Neuron.h"
+#include "../settings.h"
enum NET_TYPE{GENERAL};
diff --git a/src/Solvers/DESolver.cpp b/src/Solvers/DESolver.cpp
index cda2606bebf231e5b3304abf371451a8e7cac025..5c37a97084d926db5b3ab5067f47d5fb2e22be30 100644
--- a/src/Solvers/DESolver.cpp
+++ b/src/Solvers/DESolver.cpp
@@ -64,7 +64,7 @@ DESolver::DESolver( unsigned int n_equations, unsigned int n_inputs, unsigned in
}
printf("Differential Equation Solver with %d equations\n--------------------------------------------------------------------------\n", n_equations);
- printf("Constructing NN structure representing the solution [%d input neurons][%d inner neurons][%d output neurons]...", n_inputs, m, n_outputs);
+ printf("Constructing NN structure representing the solution [%d input neurons][%d inner neurons][%d output neurons]...\n", n_inputs, m, n_outputs);
this->dim_i = n_inputs;
this->dim_o = n_outputs;
@@ -102,17 +102,20 @@ DESolver::DESolver( unsigned int n_equations, unsigned int n_inputs, unsigned in
}
/* connections between input neurons and inner neurons */
+ size_t weight_idx;
unsigned int dim_shift = this->dim_i + this->dim_o;
for(unsigned int i = 0; i < this->dim_i; ++i){
for(unsigned int j = 0; j < this->dim_inn; ++j){
- this->solution->add_connection_simple(i, dim_shift + j);
+ weight_idx = this->solution->add_connection_simple(i, dim_shift + j);
+ printf(" adding a connection between input neuron %2d and inner neuron %2d, weight index %d\n", i, j, weight_idx);
}
}
/* connections between inner neurons and output neurons */
for(unsigned int i = 0; i < this->dim_inn; ++i){
for(unsigned int j = 0; j < this->dim_o; ++j){
- this->solution->add_connection_simple(dim_shift + i, this->dim_i + j);
+ weight_idx = this->solution->add_connection_simple(dim_shift + i, this->dim_i + j);
+ printf(" adding a connection between inner neuron %2d and output neuron %2d, weight index %d\n", i, j, weight_idx);
}
}
@@ -129,7 +132,7 @@ DESolver::DESolver( unsigned int n_equations, unsigned int n_inputs, unsigned in
this->errors_functions_types = new std::vector<ErrorFunctionType >(this->n_equations);
this->errors_functions_data_sets = new std::vector<DataSet*>(this->n_equations);
- printf(" done\n");
+ printf("done\n\n");
}
@@ -248,7 +251,7 @@ void DESolver::add_to_differential_equation( unsigned int equation_idx, MultiInd
size_t weight_idx = 0;
for(unsigned int i = 0; i < this->dim_i; ++i){
for(unsigned int j = 0; j < this->dim_inn; ++j){
-// printf(" adding a connection between input neuron %2d and inner neuron %2d, weight index %d\n", i, j, weight_idx);
+ printf(" adding a connection between input neuron %2d and inner neuron %2d, weight index %d\n", i, j, weight_idx);
new_net->add_connection_simple(i, dim_shift + j, weight_idx);
weight_idx++;
}
@@ -260,7 +263,7 @@ void DESolver::add_to_differential_equation( unsigned int equation_idx, MultiInd
// size_t weight_idx_mem = weight_idx;
for(unsigned int i = 0; i < this->dim_inn; ++i){
for(unsigned int j = 0; j < this->dim_o; ++j){
-// printf(" adding a connection between inner neuron %2d and glue neuron %2d, weight index %d\n", i, i * this->dim_o + j, weight_idx);
+ printf(" adding a connection between inner neuron %2d and glue neuron %2d, weight index %d\n", i, i * this->dim_o + j, weight_idx);
new_net->add_connection_simple(dim_shift + i, glue_shift + i * this->dim_o + j, weight_idx );
weight_idx++;
}
@@ -274,7 +277,7 @@ void DESolver::add_to_differential_equation( unsigned int equation_idx, MultiInd
unsigned int local_weight_idx = pi * this->dim_inn + i;
for(unsigned int j = 0; j < this->dim_o; ++j){
-// printf(" adding a connection between glue neuron %2d and output neuron %2d, weight index %d\n", i * this->dim_o + j, j, local_weight_idx);
+ printf(" adding a connection between glue neuron %2d and output neuron %2d, weight index %d\n", i * this->dim_o + j, j, local_weight_idx);
new_net->add_connection_simple(glue_shift + i * this->dim_o + j, j + this->dim_i, local_weight_idx );
}
}
@@ -288,7 +291,7 @@ void DESolver::add_to_differential_equation( unsigned int equation_idx, MultiInd
unsigned int local_weight_idx = pi * this->dim_inn + i;
for(unsigned int j = 0; j < this->dim_o; ++j){
-// printf(" adding a connection between glue neuron[1] %2d and glue neuron[2] %2d, weight index %d\n", i * this->dim_o + j, i * this->dim_o + j, local_weight_idx);
+ printf(" adding a connection between glue neuron[1] %2d and glue neuron[2] %2d, weight index %d\n", i * this->dim_o + j, i * this->dim_o + j, local_weight_idx);
new_net->add_connection_simple(glue_shift + i * this->dim_o + j, glue_shift_2 + i * this->dim_o + j, local_weight_idx );
}
}
@@ -300,8 +303,8 @@ void DESolver::add_to_differential_equation( unsigned int equation_idx, MultiInd
unsigned int local_weight_idx = pi * this->dim_inn + i;
for(unsigned int j = 0; j < this->dim_o; ++j){
-// printf(" adding a connection between glue neuron[2] %2d and output neuron %2d, weight index %d\n", i * this->dim_o + j, j, local_weight_idx);
- new_net->add_connection_simple(glue_shift_2 + i * this->dim_o + j, j, local_weight_idx );
+ printf(" adding a connection between glue neuron[2] %2d and output neuron %2d, weight index %d\n", i * this->dim_o + j, j, local_weight_idx);
+ new_net->add_connection_simple(glue_shift_2 + i * this->dim_o + j, this->dim_i + j, local_weight_idx );
}
}
}
@@ -349,10 +352,28 @@ void DESolver::solve_via_particle_swarm(double *domain_bounds, double c1, double
ParticleSwarm swarm_01(&total_error, domain_bounds, c1, c2, w, n_particles, max_iters);
+ this->solution->randomize_weights();
+
swarm_01.optimize(gamma, epsilon, delta);
}
NeuralNetwork* DESolver::get_solution() {
return this->solution;
+}
+
+void DESolver::eval_equation( unsigned int equation_idx, double *weight_values, std::vector<double> &input ) {
+ std::vector<double> output(this->dim_o);
+
+ this->differential_equations->at( equation_idx )->eval_single( input, output, weight_values );
+
+ printf("Input: ");
+ for( auto e: input ){
+ printf("%f, ", e);
+ }
+ printf("\nOutput: ");
+ for( auto e: output ){
+ printf("%f, ", e);
+ }
+ printf("\n");
}
\ No newline at end of file
diff --git a/src/Solvers/DESolver.h b/src/Solvers/DESolver.h
index 5165c5c30d48bcf70ef5648349fe71c5c87727eb..f3c3379778470720355f449ca9583211514cd2db 100644
--- a/src/Solvers/DESolver.h
+++ b/src/Solvers/DESolver.h
@@ -162,6 +162,11 @@ public:
* @return
*/
NeuralNetwork* get_solution( );
+
+ /**
+ * For testing purposes only
+ */
+ void eval_equation( unsigned int equation_idx, double * weight_values, std::vector<double> &input );
};
diff --git a/src/net_test_ode_1.cpp b/src/net_test_ode_1.cpp
index abafd1c8c221e7971da23374ec57e757f4b6ccf3..02be2df3369e9f9af7f2bcd18b244c536c1f9835 100644
--- a/src/net_test_ode_1.cpp
+++ b/src/net_test_ode_1.cpp
@@ -18,7 +18,7 @@
int main() {
unsigned int n_inputs = 1;
- unsigned int n_inner_neurons = 1;
+ unsigned int n_inner_neurons = 64;
unsigned int n_outputs = 1;
unsigned int n_equations = 3;
@@ -42,6 +42,10 @@ int main() {
/* neumann boundary condition */
solver_01.add_to_differential_equation( 2, alpha_1, 1.0 );
+// double weights[2] = {0.5, 1.0};
+// std::vector<double> inp = { 1.0 };
+// solver_01.eval_equation( 0, weights, inp );
+
/* SETUP OF THE TRAINING DATA */
std::vector<double> inp, out;
@@ -94,7 +98,7 @@ int main() {
/* PARTICLE SWARM TRAINING METHOD SETUP */
- unsigned int max_iters = 100;
+ unsigned int max_iters = 1000;
//must encapsulate each of the partial error functions
double *domain_bounds = new double[ 2 * n_inner_neurons * (n_inputs + n_outputs) ];
@@ -105,11 +109,11 @@ int main() {
double c1 = 0.5, c2 = 1.5, w = 0.8;
- unsigned int n_particles = 10;
+ unsigned int n_particles = 100;
double gamma = 0.5, epsilon = 0.02, delta = 0.9;
- solver_01.solve_via_particle_swarm( domain_bounds, c1, c2, w, n_particles, max_iters, gamma, epsilon, delta);
+ solver_01.solve_via_particle_swarm( domain_bounds, c1, c2, w, n_particles, max_iters, gamma, epsilon, delta );
NeuralNetwork *solution = solver_01.get_solution();
diff --git a/src/settings.h b/src/settings.h
new file mode 100644
index 0000000000000000000000000000000000000000..008004e49aa55e5bc18c8c9f64ab190c6a49d914
--- /dev/null
+++ b/src/settings.h
@@ -0,0 +1,17 @@
+/**
+ * DESCRIPTION OF THE FILE
+ *
+ * @author Michal KravÄŤenko
+ * @date 24.7.18 -
+ */
+
+#ifndef INC_4NEURO_SETTINGS_H
+#define INC_4NEURO_SETTINGS_H
+
+/**
+ * If defined, the NN feed-forward will print out whats happening
+ */
+//#define VERBOSE_NN_EVAL
+
+
+#endif //INC_4NEURO_SETTINGS_H