public string Learn(string knowFile)
{
var network = new ActivationNetwork(new BipolarSigmoidFunction(), Constants.StoneCount, 1);
var teacher = new BackPropagationLearning(network); //new PerceptronLearning(network);
var data = LoadData(knowFile);
double error = int.MaxValue;
int index = 0;
while (error > 1.0 && index++ < 5000) {
error = teacher.RunEpoch(data.Item1, data.Item2);
}
var networkFile = knowFile + ".net";
network.Save(networkFile);
Console.WriteLine("Learn: {0}, Gen: {1}", knowFile, networkFile);
return networkFile;
}
private void Worker()
{
ExtendedIOHelpers.ShowAlert("Generate new samples before training? y/n ", () => GenerateSamples());
if (!LoadSamples())
{
return;
}
Console.WriteLine("Loaded {0} samples", _inputList.Count);
var network = new ActivationNetwork(new SigmoidFunction(1), InputCount, OutputCount);
var teacher = new BackPropagationLearning(network)
{
LearningRate = 0.01
};
Console.WriteLine("Start training the network.");
int iteration = 0;
const int iterations = 5000;
double error = 1.0;
var st = new Stopwatch();
st.Start();
while (iteration < iterations && error > 0.00005)
{
error = teacher.RunEpoch(_inputList.ToArray(), _outputList.ToArray()) / _inputList.Count;
iteration++;
}
var time = st.ElapsedMilliseconds / 1000.0;
st.Stop();
Console.WriteLine("Network successfully trained! Error = {0:0.######}, Iteration = {1}", error, iteration);
Console.WriteLine("Time = {0:0.000} s\n", time);
// Normalize weights and convert to string format.
var weights = network.Layers
.Select(layer => layer.Neurons
.Select(neuron => neuron.Weights
.Select(x => string.Format("{0,6}", Convert.ToInt32(x * 1000.0)))));
ExtendedIOHelpers.ShowAlert("Do you want to save network to file? y/n ", () =>
{
SaveWeights(weights);
network.Save(NetworkFile);
});
Console.ReadLine();
}
public static void Optimize(
FsdParser inputParser, string outputFileName,
int[][] layers,
double minThreshold, double maxThreshold, double thresholdStepSize,
int minTrades,
int iterations,
int randomize,
FsdParser[] validationParsers)
{
HashSet<int[]> layerCombinations;
getLayerCombinations(layers, out layerCombinations);
Network bestNetwork = null;
double bestThreshold = double.NaN;
double bestMinWinRate = double.NegativeInfinity;
int cursorLeft = Console.CursorLeft;
int cursorTop = Console.CursorTop;
int numTotalIterations = (int)(layerCombinations.Count * ((maxThreshold - minThreshold) / thresholdStepSize + 1) * iterations);
int numIterations = 0;
foreach (int[] layerCombination in layerCombinations)
{
ActivationNetwork network = new ActivationNetwork(
new SigmoidFunction(),
inputParser.InputVectors[0].Length,
layerCombination);
network.Randomize();
ParallelResilientBackpropagationLearning teacher = new ParallelResilientBackpropagationLearning(network);
for (double threshold = minThreshold; threshold <= maxThreshold; threshold += thresholdStepSize)
{
for (int iteration = 1; iteration <= iterations; iteration++)
{
numIterations++;
Console.CursorLeft = cursorLeft;
Console.CursorTop = cursorTop;
Console.Write(new string(' ', Console.BufferWidth * 10));
Console.CursorLeft = cursorLeft;
Console.CursorTop = cursorTop;
Console.Write("layerCombination[]: { ");
for (int layerComponentIdx = 0; layerComponentIdx < layerCombination.Length; layerComponentIdx++)
{
Console.Write(layerCombination[layerComponentIdx]);
if (layerComponentIdx < layerCombination.Length - 1)
Console.Write(", ");
}
Console.WriteLine(" }");
Console.WriteLine("threshold: {0:0.00}", threshold);
Console.WriteLine("iteration: {0}", iteration);
Console.WriteLine("bestMinWinRate: {0:0.00}%", bestMinWinRate);
Console.WriteLine("");
Console.WriteLine("Progress: {0:0.00}%", (double)numIterations / numTotalIterations * 100.0);
if (randomize > 0 && (iteration - 1) % randomize == 0)
network.Randomize();
teacher.RunEpoch(inputParser.InputVectors, inputParser.OutputVectors);
bool validData = true;
double minWinRate = double.PositiveInfinity;
foreach (FsdParser validationParser in validationParsers)
{
int numTradesWon, numTradesLost;
double tradeWinRate;
getStatistics(network, validationParser, threshold, out numTradesWon, out numTradesLost, out tradeWinRate);
if (numTradesWon + numTradesLost < minTrades)
{
validData = false;
break;
}
minWinRate = Math.Min(minWinRate, tradeWinRate);
if (minWinRate < bestMinWinRate)
{
validData = false;
break;
}
}
if (validData)
{
bestNetwork = network;
bestThreshold = threshold;
bestMinWinRate = minWinRate;
network.Save(outputFileName);
// Konfigurationsinformationen speichern
string configuration = "";
configuration += "layerCombination[]: { ";
for (int layerComponentIdx = 0; layerComponentIdx < layerCombination.Length; layerComponentIdx++)
{
configuration += layerCombination[layerComponentIdx];
if (layerComponentIdx < layerCombination.Length - 1)
configuration += ", ";
}
configuration += " }\r\n";
configuration += string.Format("threshold: {0:0.00}\r\n", threshold);
configuration += string.Format("iteration: {0}\r\n", iteration);
configuration += string.Format("bestMinWinRate: {0:0.00}%\r\n", bestMinWinRate);
File.WriteAllText(outputFileName + ".txt", configuration);
}
}
}
}
}
/// <summary>
/// Background worker for neural network learning
/// </summary>
/// <param name="sender"></param>
/// <param name="e"></param>
private void backgroundWorkerNeural_DoWork(object sender, DoWorkEventArgs e)
{
string connectionString = "Data Source=192.168.0.245;Initial Catalog=MyDB;Integrated Security=True; Connection Timeout=30000";// +
using (SqlConnection connection = new SqlConnection(connectionString))
{
try
{
connection.Open();
string queryString = "SELECT * FROM rates WHERE [i1]<>0 AND [i2]<>0 AND [i3]<>0 AND [i4]<>0 AND [i5]<>0 AND [i6]<>0 AND [i7]<>0 AND [i8]<>0" +
" AND [i9]<>0 AND [i10]<>0 AND [i11]<>0 AND [i12]<>0 AND [i13]<>0 AND [i14]<>0 AND [i15]<>0 AND [i16]<>0" +
" AND [i17]<>0 AND [i18]<>0 AND [i20]<>0 AND [i21]<>0 AND [i23]<>0";
SqlCommand command = new SqlCommand(queryString, connection);
SqlDataReader reader = command.ExecuteReader();
DataTable dt = new DataTable();
dt.Load(reader);
int value_count = dt.Rows.Count;
double[][] input_arr = new double[value_count][];
double[][] output_arr = new double[value_count][];
for (int j = 0; j < value_count; j++)
{
input_arr[j] = new double[21];
output_arr[j] = new double[1];
}
// Call Read before accessing data.
for (int i = 0; i < value_count; i++)
{
DataRow row = dt.Rows[i];
input_arr[i][0] = Double.Parse(row["i1"].ToString());
input_arr[i][1] = Double.Parse(row["i2"].ToString());
input_arr[i][2] = Double.Parse(row["i3"].ToString());
input_arr[i][3] = Double.Parse(row["i4"].ToString());
input_arr[i][4] = Double.Parse(row["i5"].ToString());
input_arr[i][5] = Double.Parse(row["i6"].ToString());
input_arr[i][6] = Double.Parse(row["i7"].ToString());
input_arr[i][7] = Double.Parse(row["i8"].ToString());
input_arr[i][8] = Double.Parse(row["i9"].ToString());
input_arr[i][9] = Double.Parse(row["i10"].ToString());
input_arr[i][10] = Double.Parse(row["i11"].ToString());
input_arr[i][11] = Double.Parse(row["i12"].ToString());
input_arr[i][12] = Double.Parse(row["i13"].ToString());
input_arr[i][13] = Double.Parse(row["i14"].ToString());
input_arr[i][14] = Double.Parse(row["i15"].ToString());
input_arr[i][15] = Double.Parse(row["i16"].ToString());
input_arr[i][16] = Double.Parse(row["i17"].ToString());
input_arr[i][17] = Double.Parse(row["i18"].ToString());
input_arr[i][18] = Double.Parse(row["i20"].ToString());
input_arr[i][19] = Double.Parse(row["i21"].ToString());
input_arr[i][20] = Double.Parse(row["i23"].ToString());
//output_arr[i][0] = Double.Parse(row["Difference"].ToString());
if (Double.Parse(row["Difference"].ToString()) > 0)
output_arr[i][0] = 1;
else if (Double.Parse(row["Difference"].ToString()) == 0)
output_arr[i][0] = 0;
else
output_arr[i][0] = -1;
}
int[] neurons = new int[5] { 21, 21, 21, 21, 1 };
AForge.Neuro.BipolarSigmoidFunction sigmoiddFunction = new AForge.Neuro.BipolarSigmoidFunction();
//AForge.Neuro.SigmoidFunction sigmoiddFunction = new AForge.Neuro.SigmoidFunction(2);
AForge.Neuro.ActivationNetwork network = new AForge.Neuro.ActivationNetwork(sigmoiddFunction, 21, 1);
AForge.Neuro.ActivationNetwork network1 = new AForge.Neuro.ActivationNetwork(sigmoiddFunction, 21, 1);//neurons);
//AForge.Neuro.Learning.DeltaRuleLearning teacher = new AForge.Neuro.Learning.DeltaRuleLearning(network) { LearningRate = 1};
AForge.Neuro.Learning.EvolutionaryLearning teacher = new AForge.Neuro.Learning.EvolutionaryLearning(network, 1000);
// AForge.Neuro.Learning.ResilientBackpropagationLearning teacher = new AForge.Neuro.Learning.ResilientBackpropagationLearning(network) { LearningRate = 1 };
//AForge.Neuro.Learning.PerceptronLearning teacherP = new PerceptronLearning(network1){ LearningRate =1};
//AForge.Neuro.Learning.BackPropagationLearning teacher = new AForge.Neuro.Learning.BackPropagationLearning(network) { LearningRate =1, Momentum = .2 };
// loop
bool noNeedToStop = false;
double error = 0;
//double error1 = 0;
double lastError = 0;
double learningRate = 1;
int k = 0;
sigmoiddFunction.Alpha = 0.01;
while (!noNeedToStop)
{
// run epoch of learning procedure
//error = teacher.RunEpoch(input_arr, output_arr);
//error = teacherP.RunEpoch(input_arr,output_arr);
error = teacher.RunEpoch(input_arr, output_arr);
double temp = Math.Abs(lastError - error);
if (error < 30)
noNeedToStop = true;
else if (temp < 0.0000001)
{
lastError = error;
k++;
if (k > 1000)
{
network.Randomize();
k = 0;
}
learningRate /= 2;
//if (learningRate < 0.001)
// {
// learningRate = 0.001;
//network.Randomize();
// noNeedToStop = true;
// }
}
else
lastError = error;
// teacherP.LearningRate = learningRate;
}
network.Save(@"E:\\neural");
}
catch (Exception ex)
{
message += " Exception: " + ex.Message;
}
finally
{
connection.Close();
}
}
}
public void Learn()
{
var network = new ActivationNetwork(new BipolarSigmoidFunction(), Constants.StoneCount, 1);
var teacher = new BackPropagationLearning(network);//new PerceptronLearning(network);
var data = LoadData("4-6-2012-04-24.know");
double error = 1.0;
int index = 0;
while (error > 0.001 && index < 100000) {
error = teacher.RunEpoch(data.Item1, data.Item2);
index++;
}
network.Save("4-6-2012-04-24.bp.net");
var text = "□○○○●○○□○●●□□●□□";
var i = ToDouble(text);//-2
var o = network.Compute(i);
var eval = o[0] * 2 * Constants.StoneCount - Constants.StoneCount;
Console.WriteLine("{0} {1}", text, eval);
}
public void LearnDemo()
{
ActivationNetwork network = new ActivationNetwork(new ThresholdFunction(), 2, 1);//Constants.StoneCount
PerceptronLearning teacher = new PerceptronLearning(network);
double[][] input = new double[4][];
double[][] output = new double[4][];
input[0] = new double[] { 0, 0 };
output[0] = new double[] { 0 };
input[1] = new double[] { 0, 1 };
output[1] = new double[] { 0 };
input[2] = new double[] { 1, 0 };
output[2] = new double[] { 0 };
input[3] = new double[] { 1, 1 };
output[3] = new double[] { 1 };
double error = 1.0;
while (error > 0.001) {
error = teacher.RunEpoch(input, output);
}
var k = network.Compute(new double[] { 0.9, 0.7 });
var o = network.Output;
network.Save("a.txt");
}
private void SearchSolution()
{
MemoryStream ms = new MemoryStream();
mNetwork = new ActivationNetwork(
new BipolarSigmoidFunction(sigmoidAlphaValue),
mInput[0].Length, mHiddenNeurons, mOutput[0].Length);
// create teacher
BackPropagationLearning teacher = new BackPropagationLearning(mNetwork);
// set learning rate and momentum
teacher.LearningRate = mLearningRate;
teacher.Momentum = mMomentum;
bool needToStop = false;
int iteration = 0;
while (!needToStop)
{
double error = teacher.RunEpoch(mInput, mOutput) / mInput.Length;
mErrors[iteration] = error;
double test_error = 0.0;
for (int i = 0; i < mTestInput.Length; i++)
{
double[] test_result = mNetwork.Compute(mTestInput[i]);
test_error += ( mTestOutput[i][0] - test_result[0])*( mTestOutput[i][0] - test_result[0]);
}
mTestErrors[iteration] = Math.Sqrt(test_error);
if (min_test_error > mTestErrors[iteration])
{
min_test_error = mTestErrors[iteration];
// mTestBestNetwork = mNetwork;
ms = new MemoryStream();
mNetwork.Save(this.Id + ".txt");
}
iteration++;
if (iteration >= mIterations) needToStop = true;
}
mTestBestNetwork = (ActivationNetwork)ActivationNetwork.Load(this.Id + ".txt");
}
private static void trainNetwork(ActivationNetwork neuralNet, ISupervisedLearning teacher,
double[][] input, double[][] output, double[][] crossValidationInput, char[] crossValidationDataLabels)
{
//Make the network learn the data
DefaultLog.Info("Training the neural network . . .");
double error;
//TODO: Store the previous NUM_ITERATIONS_EQUAL_IMPLIES_PLATEAU networks so in the event of over-learning, we can return to the best one
//Use the cross-validation data to notice if the network starts to over-learn the data.
//Store the previous network (before training) and check if the performance drops on the cross-validation data
MemoryStream prevNetworkStream = new MemoryStream();
uint prevNetworkNumMisclassified = uint.MaxValue;
Queue<uint> prevNetworksNumMisclassified = new Queue<uint>(NUM_ITERATIONS_EQUAL_IMPLIES_PLATEAU);
//Initialise the queue to be full of uint.MaxValue
for (int i = 0; i < NUM_ITERATIONS_EQUAL_IMPLIES_PLATEAU; i++)
{
prevNetworksNumMisclassified.Enqueue(prevNetworkNumMisclassified);
}
int iterNum = 1;
do
{
//Perform an iteration of training (calls teacher.Run() for each item in the array of inputs/outputs provided)
error = teacher.RunEpoch(input, output);
//Progress update
if (iterNum % ITERATIONS_PER_PROGRESS_UPDATE == 0)
{
DefaultLog.Debug(String.Format("Learned for {0} iterations. Error: {1}", iterNum, error));
}
//Evaluate this network on the cross-validation data
NeuralNetworkEvaluator crossValidationEvaluator = new NeuralNetworkEvaluator(neuralNet);
crossValidationEvaluator.Evaluate(crossValidationInput, crossValidationDataLabels);
uint networkNumMisclassified = crossValidationEvaluator.ConfusionMatrix.NumMisclassifications;
DefaultLog.Debug(String.Format("Network misclassified {0} / {1} on the cross-validation data set", networkNumMisclassified,
crossValidationEvaluator.ConfusionMatrix.TotalClassifications));
//Check if we've overlearned the data and performance on the cross-valiadtion data has dropped off
if (networkNumMisclassified > Stats.Mean(prevNetworksNumMisclassified)) //Use the mean of the number of misclassification, as the actual number can move around a bit
{
//Cross-Validation performance has dropped, reinstate the previous network & break
DefaultLog.Debug(String.Format("Network has started to overlearn the training data on iteration {0}. Using previous classifier.", iterNum));
prevNetworkStream.Position = 0; //Set head to start of stream
neuralNet = ActivationNetwork.Load(prevNetworkStream) as ActivationNetwork; //Read in the network
break;
}
//Clear the Memory Stream storing the previous network
prevNetworkStream.SetLength(0);
//Store this network & the number of characters it misclassified on the cross-validation data
neuralNet.Save(prevNetworkStream);
//This is now the previous network, update the number it misclassified
prevNetworkNumMisclassified = networkNumMisclassified;
prevNetworksNumMisclassified.Dequeue();
prevNetworksNumMisclassified.Enqueue(prevNetworkNumMisclassified);
//Check if the performance has plateaued
if (prevNetworksNumMisclassified.Distinct().Count() == 1) //Allow for slight movement in performance here??
{
//Cross-Validation performance has plateaued, use this network as the final one & break
DefaultLog.Debug(String.Format("Network performance on cross-validation data has plateaued on iteration {0}.", iterNum));
break;
}
//Check if we've performed the max number of iterations
if (iterNum > MAX_LEARNING_ITERATIONS)
{
DefaultLog.Debug(String.Format("Reached the maximum number of learning iterations ({0}), with error {1}", MAX_LEARNING_ITERATIONS, error));
break;
}
iterNum++;
}
while (error > LEARNED_AT_ERROR);
DefaultLog.Info("Data learned to an error of {0}", error);
}
//Train the network many times, with different initial values, evaluate them on the cross valiadtion data and select the best one
private static ActivationNetwork trainNetworksCompeteOnCrossValidation(ActivationNetwork neuralNet, ISupervisedLearning teacher,
double[][] input, double[][] output, double[][] crossValidationInput, char[] crossValidationDataLabels)
{
DefaultLog.Info("Training {0} neural networks & picking the one that performs best on the cross-validation data . . .",
NUM_NETWORKS_TO_TRAIN_FOR_CROSS_VALIDATION_COMPETITION);
MemoryStream bestNetworkStream = new MemoryStream();
uint bestNetworkNumMisclassified = uint.MaxValue;
for (int i = 0; i < NUM_NETWORKS_TO_TRAIN_FOR_CROSS_VALIDATION_COMPETITION; i++)
{
DefaultLog.Info("Training network {0}/{1}", (i + 1), NUM_NETWORKS_TO_TRAIN_FOR_CROSS_VALIDATION_COMPETITION);
//Train a new network
neuralNet.Randomize(); //Reset the weights to random values
trainNetwork(neuralNet, teacher, input, output, crossValidationInput, crossValidationDataLabels);
//Compare this new networks performance to our current best network
NeuralNetworkEvaluator evaluator = new NeuralNetworkEvaluator(neuralNet);
evaluator.Evaluate(crossValidationInput, crossValidationDataLabels);
uint numMisclassified = evaluator.ConfusionMatrix.NumMisclassifications;
if (numMisclassified < bestNetworkNumMisclassified)
{
//This network performed better than out current best network, make this the new best
//Clear the Memory Stream storing the current best network
bestNetworkStream.SetLength(0);
//Save the network & update the best numMisclassified
neuralNet.Save(bestNetworkStream);
bestNetworkNumMisclassified = numMisclassified;
}
}
DefaultLog.Info("Trained all networks and selected the best one");
//Load up the network that performed best
bestNetworkStream.Position = 0; //Read from the start of the stream
ActivationNetwork bestNetwork = ActivationNetwork.Load(bestNetworkStream) as ActivationNetwork;
return bestNetwork;
}
private static NeuralNetworkEvaluator evaluateSingleLayerActivationNetworkWithSigmoidFunctionBackPropagationLearning(
double[][] input, double[][] output, double[][] crossValidationInput, char[] crossValidationDataLabels,
double[][] evaluationInput, char[] evaluationDataLabels, double learningRate, string networkName)
{
//Create the neural Network
BipolarSigmoidFunction sigmoidFunction = new BipolarSigmoidFunction(2.0f);
ActivationNetwork neuralNet = new ActivationNetwork(sigmoidFunction, input[0].Length, ClassifierHelpers.NUM_CHAR_CLASSES);
//Randomise the networks initial weights
neuralNet.Randomize();
//Create teacher that the network will use to learn the data (Back Propogation Learning technique used here)
BackPropagationLearning teacher = new BackPropagationLearning(neuralNet);
teacher.LearningRate = LEARNING_RATE;
//Train the Network
//trainNetwork(neuralNet, teacher, input, output, crossValidationInput, crossValidationDataLabels);
//Train multiple networks, pick the one that performs best on the Cross-Validation data
neuralNet = trainNetworksCompeteOnCrossValidation(neuralNet, teacher,
input, output, crossValidationInput, crossValidationDataLabels);
//Evaluate the network returned on the cross-validation data so it can be compared to the current best
NeuralNetworkEvaluator crossValEvaluator = new NeuralNetworkEvaluator(neuralNet);
crossValEvaluator.Evaluate(crossValidationInput, crossValidationDataLabels);
//See if this network is better than the current best network of it's type
//Try and load a previous network of this type
string previousNetworkPath = Program.NEURAL_NETWORKS_PATH + networkName + Program.NEURAL_NETWORK_FILE_EXTENSION;
string networkCMPath = Program.NEURAL_NETWORKS_PATH + networkName + ".csv";
bool newBest = false;
ActivationNetwork bestNetwork = neuralNet;
if(File.Exists(previousNetworkPath))
{
//Load the previous network & evaluate it
ActivationNetwork previous = ActivationNetwork.Load(previousNetworkPath) as ActivationNetwork;
NeuralNetworkEvaluator prevCrossValEval = new NeuralNetworkEvaluator(previous);
prevCrossValEval.Evaluate(crossValidationInput, crossValidationDataLabels);
//If this network is better than the previous best, write it out as the new best
if(prevCrossValEval.ConfusionMatrix.NumMisclassifications > crossValEvaluator.ConfusionMatrix.NumMisclassifications)
{
DefaultLog.Info("New best cross-validation score for network \"{0}\". Previous was {1}/{2}, new best is {3}/{2}",
networkName, prevCrossValEval.ConfusionMatrix.NumMisclassifications, prevCrossValEval.ConfusionMatrix.TotalClassifications,
crossValEvaluator.ConfusionMatrix.NumMisclassifications);
//Delete the old files
File.Delete(previousNetworkPath);
File.Delete(networkCMPath);
newBest = true;
}
else //The previous network is still the best
{
DefaultLog.Info("Existing \"{0}\" network performed better than new one. New network scored {1}/{2}, existing scored {3}/{2}",
networkName, crossValEvaluator.ConfusionMatrix.NumMisclassifications, crossValEvaluator.ConfusionMatrix.TotalClassifications,
prevCrossValEval.ConfusionMatrix.NumMisclassifications);
bestNetwork = previous;
}
}
else //Otherwise there isn't a previous best
{
DefaultLog.Info("No previous best record for network \"{0}\" . . .", networkName);
newBest = true;
}
//Evaluate the best system on the evaluation data
NeuralNetworkEvaluator evaluator = new NeuralNetworkEvaluator(bestNetwork);
evaluator.Evaluate(evaluationInput, evaluationDataLabels);
//If there is a new best to write out
if(newBest)
{
DefaultLog.Info("Writing out net best network of type\"{0}\"", networkName);
neuralNet.Save(previousNetworkPath);
//Write out the Confusion Matrix for the evaluation data, not cross-validation
evaluator.ConfusionMatrix.WriteToCsv(networkCMPath);
DefaultLog.Info("Finished writing out network \"{0}\"", networkName);
}
return evaluator;
}
public void startTrain()
{
int class_count = 0;
if ("NN".Equals(Constants.NN_SVM_SURF))
{
double sigmoidAlphaValue = 1.0;
double learningRate = 100;
int max_epoch = 50;
double min_err = 0.000000001;
List<FileInfo> trainingFiles = FileTools.getTrainingFiles(ref class_count);
int samples = trainingFiles.Count;
// prepare learning data
double[][] input = new double[samples][];
Dictionary<int, double[][]> outputs = new Dictionary<int, double[][]>();
for (int i = 0; i < samples; i++)
{
int currentImageClass = Int32.Parse(trainingFiles[i].Directory.Name);
Bitmap bmp = (Bitmap)Bitmap.FromFile(trainingFiles[i].FullName, false);
int com_x = 0, com_y = 0;
ByteTools.imageCoM(bmp, ref com_x, ref com_y);
input[i] = new double[numOfinputs];
List<Ipoint> featureList = fillFeatures(bmp, com_x, com_y, input[i]);
if (!outputs.ContainsKey(currentImageClass))
{
outputs.Add(currentImageClass, new double[samples][]);
for (int j = 0; j < samples; j++)
{
outputs[currentImageClass][j] = new double[] { 0d };
}
}
outputs[currentImageClass][i][0] = 1d;
}
Dictionary<int, ActivationNetwork> networks = new Dictionary<int, ActivationNetwork>();
int[] availSigns = outputs.Keys.ToArray();
foreach (int sign in availSigns)
{
ActivationNetwork network = new ActivationNetwork(new SigmoidFunction(sigmoidAlphaValue),
numOfinputs, new int[] { Constants.NUM_OF_NN_HIDDEN_LAYER_NODES, 1 });
Accord.Neuro.Learning.LevenbergMarquardtLearning teacher = new Accord.Neuro.Learning.LevenbergMarquardtLearning(network);
teacher.LearningRate = learningRate;
int epoch = 0;
double error;
while (true)
{
// run epoch of learning procedure
error = teacher.RunEpoch(input, outputs[sign]) / samples;
Console.WriteLine("Epoch:" + epoch + " Error:" + error);
if (epoch++ > max_epoch || error < min_err)
break;
}
networks.Add(sign, network);
network.Save(Constants.base_folder + "nn_12x12_" + sign + ".dat");
Logger.log("Error: " + error + " Epoch:" + epoch);
}
}
else if ("NN_SURF".Equals(Constants.NN_SVM_SURF) || "NN_12SIMPLE".Equals(Constants.NN_SVM_SURF))
{
double sigmoidAlphaValue = 1.0;
if ("NN_SURF".Equals(Constants.NN_SVM_SURF)) {
if ("triangle".Equals(Constants.CIRCLE_TRIANGLE))
sigmoidAlphaValue = 6.0;
if ("circle".Equals(Constants.CIRCLE_TRIANGLE))
sigmoidAlphaValue = 6.0;
}
else if ("NN_12SIMPLE".Equals(Constants.NN_SVM_SURF))
{
if ("triangle".Equals(Constants.CIRCLE_TRIANGLE))
sigmoidAlphaValue = 1.0;
if ("circle".Equals(Constants.CIRCLE_TRIANGLE))
sigmoidAlphaValue = 1.0;
}
double learningRate = 1.00;
int max_epoch = 3000;
double min_err = 0.000001;
ActivationNetwork network = new ActivationNetwork(new SigmoidFunction(sigmoidAlphaValue),
numOfinputs, Constants.NUM_OF_SIGN_TYPES);
DeltaRuleLearning teacher = new DeltaRuleLearning(network);
teacher.LearningRate = learningRate;
/*
ActivationNetwork network = new ActivationNetwork(new SigmoidFunction(sigmoidAlphaValue),
numOfinputs, new int[] { (numOfinputs + Constants.NUM_OF_SIGN_TYPES)/2, Constants.NUM_OF_SIGN_TYPES });
BackPropagationLearning teacher = new BackPropagationLearning(network);
teacher.LearningRate = learningRate;
//teacher.Momentum = momentum;
*/
List<FileInfo> trainingFiles = new List<FileInfo>(1000);
DirectoryInfo di = new DirectoryInfo(Constants.base_folder + "train_" + Constants.CIRCLE_TRIANGLE);
DirectoryInfo[] dirs = di.GetDirectories("*");
foreach (DirectoryInfo dir in dirs)
{
int i = 0;
FileInfo[] files = dir.GetFiles("*.bmp");
foreach (FileInfo fi in files)
{
trainingFiles.Add(fi);
if (i++ > Constants.MAX_TRAIN_SAMPLE)
break;
}
}
// List<FileInfo> trainingFiles = FileTools.getTrainingFiles(ref class_count);
int samples = trainingFiles.Count;
// prepare learning data
double[][] input = new double[samples][];
double[][] output = new double[samples][];
for (int i = 0; i < samples; i++)
{
Bitmap bmp = (Bitmap)Bitmap.FromFile(trainingFiles[i].FullName, false);
int com_x = 0, com_y = 0;
ByteTools.imageCoM(bmp, ref com_x, ref com_y);
input[i] = new double[numOfinputs];
output[i] = new double[Constants.NUM_OF_SIGN_TYPES];
bmp.Tag = trainingFiles[i].Directory.Name + "_" + trainingFiles[i].Name;
fillFeatures_SURF(bmp, com_x, com_y, input[i]);
output[i][Int32.Parse(trainingFiles[i].Directory.Name) - 1] = 1d;
}
int epoch = 0;
double error = 0;
while (true)
{
// run epoch of learning procedure
error = teacher.RunEpoch(input, output) / samples;
Console.WriteLine("Epoch:" + epoch + " Error:" + error);
if (epoch++ > max_epoch || error < min_err)
break;
}
network.Save(Constants.base_folder + Constants.NN_SVM_SURF + "_" + Constants.CIRCLE_TRIANGLE + ".dat");
Logger.log("NNTrain [" + error + "]: " + Constants.NN_SVM_SURF + ", " + Constants.CIRCLE_TRIANGLE + ", " + learningRate + ", " + sigmoidAlphaValue);
}
}
