public static IMLDataSet GenerateTrainingSet(double[][] inputData, double[][] outputData, int leadWindowSize, int lagWindowSize)
{
var temporalDataSet = new TemporalMLDataSet(lagWindowSize, leadWindowSize);
int inputCount = inputData[0].Length;
int outputCount = outputData[0].Length;
for (int i = 0; i < inputCount; i++)
{
var desc = new TemporalDataDescription(TemporalDataDescription.Type.Raw, true, false);
temporalDataSet.AddDescription(desc);
}
for (int i = 0; i < outputCount; i++)
{
var desc = new TemporalDataDescription(TemporalDataDescription.Type.Raw, false, true);
temporalDataSet.AddDescription(desc);
}
for (int i = 0; i < inputData.Length; i++)
{
var point = temporalDataSet.CreatePoint(i);
point.Data = inputData[i].Concat(outputData[i]).ToArray();
}
temporalDataSet.Generate();
return(temporalDataSet);
}
/// <summary>
/// Generates a temporal data set with a given double serie.
/// uses Type percent change.
/// </summary>
/// <param name="inputserie">The inputserie.</param>
/// <param name="windowsize">The windowsize.</param>
/// <param name="predictsize">The predictsize.</param>
/// <returns></returns>
public static TemporalMLDataSet GenerateTrainingWithPercentChangeOnSerie(double[] inputserie, int windowsize, int predictsize)
{
TemporalMLDataSet result = new TemporalMLDataSet(windowsize, predictsize);
TemporalDataDescription desc = new TemporalDataDescription(TemporalDataDescription.Type.PercentChange, true, true);
result.AddDescription(desc);
for (int index = 0; index < inputserie.Length - 1; index++)
{
TemporalPoint point = new TemporalPoint(1);
point.Sequence = index;
point.Data[0] = inputserie[index];
result.Points.Add(point);
}
result.Generate();
return(result);
}
// ReSharper disable UnusedMember.Local
private static void CreateEvaluationSet(string @fileName)
// ReSharper restore UnusedMember.Local
{
List <double> Opens = QuickCSVUtils.QuickParseCSV(fileName, "Open", 1200, 1200);
List <double> High = QuickCSVUtils.QuickParseCSV(fileName, "High", 1200, 1200);
List <double> Low = QuickCSVUtils.QuickParseCSV(fileName, "Low", 1200, 1200);
List <double> Close = QuickCSVUtils.QuickParseCSV(fileName, "Close", 1200, 1200);
List <double> Volume = QuickCSVUtils.QuickParseCSV(fileName, 5, 1200, 1200);
double[] Ranges = NetworkUtility.CalculateRanges(Opens.ToArray(), Close.ToArray());
TemporalMLDataSet superTemportal = TrainerHelper.GenerateTrainingWithPercentChangeOnSerie(100, 1, Opens.ToArray(),
Close.ToArray(), High.ToArray(), Low.ToArray(), Volume.ToArray());
IMLDataPair aPairInput = TrainerHelper.ProcessPairs(NetworkUtility.CalculatePercents(Opens.ToArray()), NetworkUtility.CalculatePercents(Opens.ToArray()), 100, 1);
IMLDataPair aPairInput3 = TrainerHelper.ProcessPairs(NetworkUtility.CalculatePercents(Close.ToArray()), NetworkUtility.CalculatePercents(Close.ToArray()), 100, 1);
IMLDataPair aPairInput2 = TrainerHelper.ProcessPairs(NetworkUtility.CalculatePercents(High.ToArray()), NetworkUtility.CalculatePercents(High.ToArray()), 100, 1);
IMLDataPair aPairInput4 = TrainerHelper.ProcessPairs(NetworkUtility.CalculatePercents(Volume.ToArray()), NetworkUtility.CalculatePercents(Volume.ToArray()), 100, 1);
IMLDataPair aPairInput5 = TrainerHelper.ProcessPairs(NetworkUtility.CalculatePercents(Ranges.ToArray()), NetworkUtility.CalculatePercents(Ranges.ToArray()), 100, 1);
List <IMLDataPair> listData = new List <IMLDataPair>();
listData.Add(aPairInput);
listData.Add(aPairInput2);
listData.Add(aPairInput3);
listData.Add(aPairInput4);
listData.Add((aPairInput5));
var minitrainning = new BasicMLDataSet(listData);
var network = (BasicNetwork)CreateElmanNetwork(100, 1);
double normalCorrectRate = EvaluateNetworks(network, minitrainning);
double temporalErrorRate = EvaluateNetworks(network, superTemportal);
Console.WriteLine(@"Percent Correct with normal Data Set:" + normalCorrectRate + @" Percent Correct with temporal Dataset:" +
temporalErrorRate);
Console.WriteLine(@"Paused , Press a key to continue to evaluation");
Console.ReadKey();
}
public IMLDataSet GenerateTraining()
{
var result = new TemporalMLDataSet(WindowSize, 1);
var desc = new TemporalDataDescription(TemporalDataDescription.Type.Raw, true, true);
result.AddDescription(desc);
for (int i = WindowSize; i < _normalizedTrainingData.Length; i++)
{
var point = new TemporalPoint(1)
{
Sequence = i
};
point.Data[0] = _normalizedTrainingData[i];
result.Points.Add(point);
}
result.Generate();
return(result);
}
public IMLDataSet GenerateTraining()
{
var result = new TemporalMLDataSet(WindowSize, 1);
var desc = new TemporalDataDescription(TemporalDataDescription.Type.Raw, true, true);
result.AddDescription(desc);
for (var year = TrainStart; year < TrainEnd; year++)
{
var point = new TemporalPoint(1)
{
Sequence = year
};
point.Data[0] = _normalizedForexPair[year];
result.Points.Add(point);
}
result.Generate();
return(result);
}
public static IMLDataSet GenerateTraining()
{
TemporalMLDataSet result = new TemporalMLDataSet(WindowSize, 1);
TemporalDataDescription desc =
new TemporalDataDescription(TemporalDataDescription.Type.Raw, true, true);
result.AddDescription(desc);
for (int year = TrainStart; year < TrainEnd; year++)
{
TemporalPoint point = new TemporalPoint(1);
point.Sequence = year;
point.Data[0] = _normalizedSunspots[year];
result.Points.Add(point);
}
result.Generate();
return(result);
}
public static void TrainSVMNetwork(ref IExampleInterface app)
{
//BasicMLDataSet set = CreateEval.CreateEvaluationSetAndLoad(app.Args[1],1000,500,CONFIG.INPUT_WINDOW,CONFIG.PREDICT_WINDOW);
TemporalMLDataSet Tempo = CreateEval.GenerateATemporalSet(app.Args[1], 1000, 500, CONFIG.INPUT_WINDOW, CONFIG.PREDICT_WINDOW);
SupportVectorMachine machine = createNetwork();
//Train it..
double error = TrainNetworks(machine, Tempo);
Console.WriteLine(@"SVM NetWork Trained to :" + error);
SuperUtils.SaveTraining(CONFIG.DIRECTORY, CONFIG.SVMTRAINING_FILE, Tempo);
SuperUtils.SaveNetwork(CONFIG.DIRECTORY, CONFIG.SVMNETWORK_FILE, machine);
Console.WriteLine(@"Network Saved to :" + CONFIG.DIRECTORY + @" File Named :" +
CONFIG.SVMNETWORK_FILE);
Console.WriteLine(@"Training Saved to :" + CONFIG.DIRECTORY + @" File Named :" +
CONFIG.SVMTRAINING_FILE);
MakeAPause();
}
/// <summary>
/// Generates a temporal data set with a given double serie or a any number of double series , making your inputs.
/// uses Type percent change.
/// </summary>
/// <param name="windowsize">The windowsize.</param>
/// <param name="predictsize">The predictsize.</param>
/// <param name="inputserie">The inputserie.</param>
/// <returns></returns>
public static TemporalMLDataSet GenerateTrainingWithPercentChangeOnSerie(int windowsize, int predictsize, params double[][] inputserie)
{
TemporalMLDataSet result = new TemporalMLDataSet(windowsize, predictsize);
TemporalDataDescription desc = new TemporalDataDescription(TemporalDataDescription.Type.PercentChange, true,
true);
result.AddDescription(desc);
foreach (double[] t in inputserie)
{
for (int j = 0; j < t.Length; j++)
{
TemporalPoint point = new TemporalPoint(1);
point.Sequence = j;
point.Data[0] = t[j];
result.Points.Add(point);
}
result.Generate();
return(result);
}
return(null);
}
public IMLDataSet GenerateTraining()
{
var result = new TemporalMLDataSet(WindowSize, 1);
var desc = new TemporalDataDescription(
TemporalDataDescription.Type.Raw, true, true);
result.AddDescription(desc);
for (int day = TrainStart; day < TrainEnd; day++)
{
var point = new TemporalPoint(1)
{
Sequence = day,
Data = { [0] = _normalizedArray[day] }
};
result.Points.Add(point);
}
result.Generate();
return(result);
}
public static BasicMLDataSet CreateEvaluationSetAndLoad(string @fileName, int startLine, int HowMany, int WindowSize, int outputsize)
{
List <double> Opens = NetworkUtility.QuickParseCSV(fileName, "Open", startLine, HowMany);
List <double> High = NetworkUtility.QuickParseCSV(fileName, "High", startLine, HowMany);
List <double> Low = NetworkUtility.QuickParseCSV(fileName, "Low", startLine, HowMany);
List <double> Close = NetworkUtility.QuickParseCSV(fileName, "Close", startLine, HowMany);
List <double> Volume = NetworkUtility.QuickParseCSV(fileName, 5, startLine, HowMany);
TemporalMLDataSet superTemportal = new TemporalMLDataSet(WindowSize, outputsize);
double[] Ranges = NetworkUtility.CalculateRanges(Opens.ToArray(), Close.ToArray());
superTemportal = NetworkUtility.GenerateTrainingWithPercentChangeOnSerie(100, 1, Opens.ToArray(),
Close.ToArray(), High.ToArray(), Low.ToArray(), Volume.ToArray());
IMLDataPair aPairInput = SuperUtils.ProcessPair(NetworkUtility.CalculatePercents(Opens.ToArray()), NetworkUtility.CalculatePercents(Opens.ToArray()), WindowSize, outputsize);
IMLDataPair aPairInput3 = SuperUtils.ProcessPair(NetworkUtility.CalculatePercents(Close.ToArray()), NetworkUtility.CalculatePercents(Close.ToArray()), WindowSize, outputsize);
IMLDataPair aPairInput2 = SuperUtils.ProcessPair(NetworkUtility.CalculatePercents(High.ToArray()), NetworkUtility.CalculatePercents(High.ToArray()), WindowSize, outputsize);
IMLDataPair aPairInput4 = SuperUtils.ProcessPair(NetworkUtility.CalculatePercents(Volume.ToArray()), NetworkUtility.CalculatePercents(Volume.ToArray()), WindowSize, outputsize);
IMLDataPair aPairInput5 = SuperUtils.ProcessPair(NetworkUtility.CalculatePercents(Ranges.ToArray()), NetworkUtility.CalculatePercents(Ranges.ToArray()), WindowSize, outputsize);
List <IMLDataPair> listData = new List <IMLDataPair>();
listData.Add(aPairInput);
listData.Add(aPairInput2);
listData.Add(aPairInput3);
listData.Add(aPairInput4);
listData.Add((aPairInput5));
var minitrainning = new BasicMLDataSet(listData);
return(minitrainning);
}
public IMLDataSet GenerateTraining(double[] normalizedData)
{
var result = new TemporalMLDataSet(WindowSize, 1);
TemporalDataDescription desc = new TemporalDataDescription(TemporalDataDescription.Type.Raw, true, true);
result.AddDescription(desc);
int TrainStart = 0;
int TrainEnd = normalizedData.Length;
for (int index = TrainStart; index < TrainEnd; index++)
{
TemporalPoint point = new TemporalPoint(1)
{
Sequence = index
};
point.Data[0] = normalizedData[index];
result.Points.Add(point);
}
result.Generate();
return(result);
}
