new doc and optical backprop

tests/CMakeLists.txt

@@ -10,11 +10,20 @@ target_link_libraries(basis NeuralNetwork)
 add_executable(backpropagation backpropagation.cpp)
 target_link_libraries(backpropagation NeuralNetwork)
 
+add_executable(backpropagation_perf backpropagation_perf.cpp)
+target_link_libraries(backpropagation_perf NeuralNetwork)
+
 add_executable(feedforward feedforward.cpp)
 target_link_libraries(feedforward NeuralNetwork)
 
 add_executable(feedforward_perf feedforward_perf.cpp)
 target_link_libraries(feedforward_perf NeuralNetwork)
 
+add_executable(optical_backpropagation optical_backpropagation.cpp)
+target_link_libraries(optical_backpropagation NeuralNetwork)
+
 add_executable(recurrent recurrent.cpp)
-target_link_libraries(recurrent NeuralNetwork)
+target_link_libraries(recurrent NeuralNetwork)
+
+add_executable(recurrent_perf recurrent_perf.cpp)
+target_link_libraries(recurrent_perf NeuralNetwork)

tests/backpropagation_perf.cpp

@@ -0,0 +1,26 @@
+#include <NeuralNetwork/FeedForward/Network.h>
+
+#include <cassert>
+#include <iostream>
+#include "../include/NeuralNetwork/Learning/BackPropagation.h"
+
+int main() {
+	{	// XOR problem
+		NeuralNetwork::FeedForward::Network n(2);
+		NeuralNetwork::ActivationFunction::Sigmoid a(-1);
+		n.appendLayer(200,a);
+		n.appendLayer(500,a);
+		n.appendLayer(900,a);
+		n.appendLayer(1,a);
+
+		n.randomizeWeights();
+
+		NeuralNetwork::Learning::BackPropagation prop;
+		for(int i=0;i<100;i++) {
+			prop.teach(n,{1,0},{1});
+			prop.teach(n,{1,1},{0});
+			prop.teach(n,{0,0},{0});
+			prop.teach(n,{0,1},{1});
+		}
+	}
+}

tests/feedforward_perf.cpp

@@ -7,8 +7,8 @@ int main() {
 	{	// XOR problem
 		NeuralNetwork::FeedForward::Network n(2);
 		NeuralNetwork::ActivationFunction::Sigmoid a(-1);
-		n.appendLayer(2000,a);
-		n.appendLayer(2000,a);
+		n.appendLayer(5000,a);
+		n.appendLayer(5000,a);
 		n.appendLayer(1,a);
 
 		for(int i=0;i<500;i++) {

tests/optical_backpropagation.cpp

@@ -0,0 +1,116 @@
+#include <NeuralNetwork/FeedForward/Network.h>
+
+#include <cassert>
+#include <iostream>
+#include "../include/NeuralNetwork/Learning/OpticalBackPropagation.h"
+
+int main() {
+	{	// XOR problem
+		NeuralNetwork::FeedForward::Network n(2);
+		NeuralNetwork::ActivationFunction::Sigmoid a(-1);
+		n.appendLayer(2,a);
+		n.appendLayer(1,a);
+
+		n.randomizeWeights();
+
+		NeuralNetwork::Learning::OpticalBackPropagation prop;
+		for(int i=0;i<10000;i++) {
+			prop.teach(n,{1,0},{1});
+			prop.teach(n,{1,1},{0});
+			prop.teach(n,{0,0},{0});
+			prop.teach(n,{0,1},{1});
+		}
+
+		{
+			std::vector<float> ret =n.computeOutput({1,1});
+			assert(ret[0] < 0.1);
+		}
+
+		{
+			std::vector<float> ret =n.computeOutput({0,1});
+			assert(ret[0] > 0.9);
+		}
+
+		{
+			std::vector<float> ret =n.computeOutput({1,0});
+			assert(ret[0] > 0.9);
+		}
+
+		{
+			std::vector<float> ret =n.computeOutput({0,0});
+			assert(ret[0] < 0.1);
+		}
+	}
+	{	// AND problem
+		NeuralNetwork::FeedForward::Network n(2);
+		NeuralNetwork::ActivationFunction::Sigmoid a(-1);
+		n.appendLayer(2,a);
+		n.appendLayer(1,a);
+
+		n.randomizeWeights();
+
+		NeuralNetwork::Learning::OpticalBackPropagation prop;
+		for(int i=0;i<10000;i++) {
+			prop.teach(n,{1,1},{1});
+			prop.teach(n,{0,0},{0});
+			prop.teach(n,{0,1},{0});
+			prop.teach(n,{1,0},{0});
+		}
+
+		{
+			std::vector<float> ret =n.computeOutput({1,1});
+			assert(ret[0] > 0.9);
+		}
+
+		{
+			std::vector<float> ret =n.computeOutput({0,1});
+			assert(ret[0] < 0.1);
+		}
+
+		{
+			std::vector<float> ret =n.computeOutput({1,0});
+			assert(ret[0] < 0.1);
+		}
+
+		{
+			std::vector<float> ret =n.computeOutput({0,0});
+			assert(ret[0] < 0.1);
+		}
+	}
+	{	// NOT AND problem
+		NeuralNetwork::FeedForward::Network n(2);
+		NeuralNetwork::ActivationFunction::Sigmoid a(-1);
+		n.appendLayer(2,a);
+		n.appendLayer(1,a);
+
+		n.randomizeWeights();
+
+		NeuralNetwork::Learning::OpticalBackPropagation prop;
+		for(int i=0;i<10000;i++) {
+			prop.teach(n,{1,1},{0});
+			prop.teach(n,{0,0},{1});
+			prop.teach(n,{0,1},{1});
+			prop.teach(n,{1,0},{1});
+		}
+
+		{
+			std::vector<float> ret =n.computeOutput({1,1});
+			assert(ret[0] < 0.1);
+		}
+
+		{
+			std::vector<float> ret =n.computeOutput({0,1});
+			assert(ret[0] > 0.9);
+		}
+
+		{
+			std::vector<float> ret =n.computeOutput({1,0});
+			assert(ret[0] > 0.9);
+		}
+
+		{
+			std::vector<float> ret =n.computeOutput({0,0});
+			assert(ret[0] > 0.9);
+		}
+	}
+}

tests/recurrent_perf.cpp

@@ -0,0 +1,12 @@
+#include <NeuralNetwork/Recurrent/Network.h>
+
+#include <assert.h>
+#include <iostream>
+
+int main() {
+	NeuralNetwork::Recurrent::Network a(2,1,1000);
+
+	for(size_t i=0;i<10000;i++) {
+		 a.computeOutput({1,0.7});
+	}
+}