private void deserializeNetwork_Click(object sender, EventArgs e)
{
switch (KlasyfikatorComboBox.SelectedIndex)
{
case 1:
ENNetwork = EncogNeuralNetwork.DeSerialize();
break;
case 3:
ENNetworkQ = EncogNeuralNetworkQuick.DeSerialize(OpenGlController.divisionCountX, OpenGlController.divisionCountY);
break;
case 4:
ENNetworkS = EncogNeuralNetworkSlow.DeSerialize();
break;
}
}
public string Teach(AlgorithmDataModel teachingData)
{
DataSet dataSet;
var files = teachingData.Files.Select(i => teachingData.Directory + i).ToArray();
var forgeries = teachingData.Forgeries?.Select(i => teachingData.ForgeriesDirectory + i).ToArray();
switch (teachingData.FeatureAlgorithmName)
{
case "RadonTransformFeatureExtractor":
_extractor = new RadonTransformFeatureExtractor(int.Parse(teachingData.FeatureAlgorithmParameters[0]));
dataSet = _extractor.GetTeachingDataset(files, forgeries);
break;
default:
throw new ArgumentException("FeatureAlgorithmName");
}
switch (teachingData.ComputingAlgorithmName)
{
case "NeuralNetworkBp":
_engine = new NeuralNetworkBp(double.Parse(teachingData.ComputingAlgorithmParameters[0].Replace(".", ",")),
double.Parse(teachingData.ComputingAlgorithmParameters[1].Replace(".", ",")));
_engine.Train(dataSet);
break;
case "EncogNeuralNetwork":
_engine = new EncogNeuralNetwork(double.Parse(teachingData.ComputingAlgorithmParameters[0].Replace(".", ",")),
double.Parse(teachingData.ComputingAlgorithmParameters[1].Replace(".", ",")));
_engine.Train(dataSet);
break;
default:
throw new ArgumentException("ComputingAlgorithmName");
}
return("OK");
}
private async void TrainButtonClick(object sender, EventArgs e)
{
if (Controller.File == null)
{
BrakPlikuMessageBox();
return;
}
switch (KlasyfikatorComboBox.SelectedIndex)
{
//2plik = 0.056 4plik = 0.066 [300k points]
case 0:
DisableChangeButtons();
await Task.Run(() =>
{
KNNClassifier = new KnnAccordClassifier(3);
if (!KNNClassifier.CanClassify)
{
var pointsToTrain = Utills.GenerateClassArray(Controller.File.LasPointDataRecords, 0.066);
var inputOutputArrays = KnnAccordClassifier.MakeInputOutputs(pointsToTrain, Controller.File);
KNNClassifier.Teach(inputOutputArrays.Item1, inputOutputArrays.Item2);
}
});
EnableChangeButtons();
break;
case 1:
DisableChangeButtons();
await Task.Run(() =>
{
ENNetwork = new EncogNeuralNetwork(Controller.File);
});
EnableChangeButtons();
break;
case 2:
DisableChangeButtons();
await Task.Run(() =>
{
KNNClassifierQuick = new KnnAccordClassifierQuick(3);
if (!KNNClassifierQuick.CanClassify)
{
var inputOutputArrays =
KnnAccordClassifierQuick.MakeInputOutputs(Utills.GroupPointsList(
Controller.File, OpenGlController.divisionCountX, OpenGlController.divisionCountY));
KNNClassifierQuick.TeachGroups(inputOutputArrays.Item1, inputOutputArrays.Item2);
}
});
EnableChangeButtons();
break;
case 3:
DisableChangeButtons();
await Task.Run(() =>
{
ENNetworkQ = new EncogNeuralNetworkQuick(Controller.File, OpenGlController.divisionCountX, OpenGlController.divisionCountY);
});
EnableChangeButtons();
break;
case 4:
DisableChangeButtons();
await Task.Run(() =>
{
KNNClassifierSlow = new KnnAccordClassifierSlow(3);
if (!KNNClassifierSlow.CanClassify)
{
var pointsToTrain = Utills.GenerateClassArray(Controller.File.LasPointDataRecords, 0.10);
var inputOutputArrays = KnnAccordClassifierSlow.MakeInputOutputs(pointsToTrain, Controller.File);
KNNClassifierSlow.Teach(inputOutputArrays.Item1, inputOutputArrays.Item2);
}
});
EnableChangeButtons();
break;
case 5:
DisableChangeButtons();
await Task.Run(() =>
{
ENNetworkS = new EncogNeuralNetworkSlow(Controller.File);
});
EnableChangeButtons();
break;
}
KoniecTreninguMessageBox();
}
