private void recognizeButton_Click(object sender, EventArgs e)
{
var bitmap = new Bitmap(charPictureBox.Image);
var inputVector = new double[IMG_WIDHT * IMG_HEIGHT];
for (var x = 0; x < bitmap.Width; x++)
{
for (var y = 0; y < bitmap.Height; y++)
{
var pixel = bitmap.GetPixel(x, y);
inputVector[y + x * IMG_HEIGHT] = Convert.ToDouble(pixel.R == 0 && pixel.G == 0 && pixel.B == 0);
}
}
NumberRecognizerNeuralNetwork numberRecognizerNeuralNetwork = new NumberRecognizerNeuralNetwork(this.neuralNetwork);
int result = numberRecognizerNeuralNetwork.GetNumberFromImgVector(inputVector);
answerListBox.Items.Clear();
foreach (var neuron in numberRecognizerNeuralNetwork.GetNeuralNetwork().GetLastLayer().GetListNeurons())
{
answerListBox.Items.Add(Math.Round(neuron.GetOutputData(), 5).ToString());
}
if (result == -1)
{
MessageBox.Show("Can not recognize char.");
}
else
{
MessageBox.Show($"Char is: {result}.");
}
}
private void educateNetworkButton_Click(object sender, EventArgs e)
{
UncertaintyPropagationMethod uncertaintyPropagationMethod = new UncertaintyPropagationMethod(LEARNING_RATE);
NumberRecognizerNeuralNetwork numberRecognizerNeuralNetwork = new NumberRecognizerNeuralNetwork(this.neuralNetwork);
MachineLearning.NeuralNetwork.Report.Manager reportManager = new MachineLearning.NeuralNetwork.Report.Manager();
NeuralNetworkObj neuralNetworkObj = numberRecognizerNeuralNetwork.GetNeuralNetwork();
var prepareData = this.GetPrepareData(neuralNetworkObj.GetLastLayer().GetCountNeurons());
int countEpoch = Convert.ToInt16(educateNetworkNumericUpDown.Value);
for (int epoch = 0; epoch < countEpoch; epoch++)
{
reportManager.AddDataBeforeEducate(neuralNetworkObj, prepareData);
for (int iteration = 0; iteration < COUNT_ITERATION; iteration++)
{
foreach (var entity in prepareData)
{
double[] inputVector = entity.Key;
double[] expectedResultVector = entity.Value;
neuralNetworkObj = uncertaintyPropagationMethod.GetTaughtNeuralNetwork(
numberRecognizerNeuralNetwork.GetNeuralNetwork(),
inputVector,
expectedResultVector
);
}
}
educateNetworkProgressBar.Value = Convert.ToInt16((epoch + 1.0) / countEpoch * 100);
reportManager.AddDataAfterEducate(neuralNetworkObj, prepareData);
reportManager.SaveReport(neuralNetworkObj);
neuralNetworkObj.EpochPassed();
}
numberRecognizerNeuralNetwork.UpdateNeuralNetwork(neuralNetworkObj);
}
private void generateButton_Click(object sender, EventArgs e)
{
//todo validation
string name = newNetworkNameTextBox.Text;
int countLayers = Convert.ToInt16(layersCountTextBox.Text);
int[] neuronsInLayer = new int[countLayers];
for (int layerId = 0; layerId < countLayers; layerId++)
{
neuronsInLayer[layerId] = Convert.ToInt16(
Interaction.InputBox($"Input count neuron in layer {layerId}?", "Configuration layers", "")
);
}
NumberRecognizerNeuralNetwork numberRecognizerNeuralNetwork = new NumberRecognizerNeuralNetwork(name, neuronsInLayer, IMG_HEIGHT, IMG_WIDHT);
numberRecognizerNeuralNetwork.GenerateRandom();
this.LoadNetworks();
}
