public override void AddDescription(TemporalDataDescription desc)
{
if (!(desc is MarketDataDescription))
{
throw new MarketError("Only MarketDataDescription objects may be used with the MarketNeuralDataSet container.");
}
base.AddDescription(desc);
}
/// <summary>
/// Get data between two points in raw form.
/// </summary>
/// <param name="desc">The data description.</param>
/// <param name="index">The index to get data from.</param>
/// <returns>The requested data.</returns>
private double GetDataRaw(TemporalDataDescription desc,
int index)
{
TemporalPoint point = _points[index - 1];
return point[desc.Index];
}
/// <summary>
/// Get data between two points in percent form.
/// </summary>
/// <param name="desc">The data description.</param>
/// <param name="index">The index to get data from.</param>
/// <returns>The requested data.</returns>
private double GetDataPercentChange(TemporalDataDescription desc,
int index)
{
if (index == 0)
{
return 0.0;
}
TemporalPoint point = _points[index];
TemporalPoint previousPoint = _points[index - 1];
double currentValue = point[desc.Index];
double previousValue = previousPoint[desc.Index];
return (currentValue - previousValue)/previousValue;
}
/// <summary>
/// Get data between two points in delta form.
/// </summary>
/// <param name="desc">The data description.</param>
/// <param name="index">The index to get data from.</param>
/// <returns>The requested data.</returns>
private double GetDataDeltaChange(TemporalDataDescription desc,
int index)
{
if (index == 0)
{
return 0.0;
}
TemporalPoint point = _points[index];
TemporalPoint previousPoint = _points[index - 1];
return point[desc.Index]
- previousPoint[desc.Index];
}
/// <summary>
/// Format data according to the type specified in the description.
/// </summary>
/// <param name="desc">The data description.</param>
/// <param name="index">The index to format the data at.</param>
/// <returns>The formatted data.</returns>
private double FormatData(TemporalDataDescription desc,
int index)
{
var result = new double[1];
switch (desc.DescriptionType)
{
case TemporalDataDescription.Type.DeltaChange:
result[0] = GetDataDeltaChange(desc, index);
break;
case TemporalDataDescription.Type.PercentChange:
result[0] = GetDataPercentChange(desc, index);
break;
case TemporalDataDescription.Type.Raw:
result[0] = GetDataRaw(desc, index);
break;
default:
throw new TemporalError("Unsupported data type.");
}
if (desc.ActivationFunction != null)
{
desc.ActivationFunction.ActivationFunction(result, 0, 1);
}
return result[0];
}
/// <summary>
/// Add a data description.
/// </summary>
/// <param name="desc">The data description to add.</param>
public virtual void AddDescription(TemporalDataDescription desc)
{
if (_points.Count > 0)
{
throw new TemporalError(
"Can't add anymore descriptions, there are "
+ "already temporal points defined.");
}
int index = _descriptions.Count;
desc.Index = index;
_descriptions.Add(desc);
CalculateNeuronCounts();
}
/// <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 virtual void AddDescription(TemporalDataDescription desc)
{
if (this._x6fa2570084b2ad39.Count <= 0)
{
int count = this._x6849a3dfb0329317.Count;
if (((uint) count) <= uint.MaxValue)
{
desc.Index = count;
this._x6849a3dfb0329317.Add(desc);
this.CalculateNeuronCounts();
return;
}
}
throw new TemporalError("Can't add anymore descriptions, there are already temporal points defined.");
}
public MarketDataDescription(TickerSymbol ticker, MarketDataType dataType, TemporalDataDescription.Type type, bool input, bool predict)
: this(ticker, dataType, type, null, input, predict)
{
}
private double xea9b27496fd731f6(TemporalDataDescription x3253f43e16d3c2a3, int xc0c4c459c6ccbd00)
{
double num;
double num2;
if (xc0c4c459c6ccbd00 == 0)
{
return 0.0;
}
TemporalPoint point = this._x6fa2570084b2ad39[xc0c4c459c6ccbd00];
TemporalPoint point2 = this._x6fa2570084b2ad39[xc0c4c459c6ccbd00 - 1];
do
{
num = point[x3253f43e16d3c2a3.Index];
num2 = point2[x3253f43e16d3c2a3.Index];
}
while ((((uint) num) | 3) == 0);
return ((num - num2) / num2);
}
/// <summary>
/// Get data between two points in raw form.
/// </summary>
/// <param name="desc">The data description.</param>
/// <param name="index">The index to get data from.</param>
/// <returns>The requested data.</returns>
private double GetDataRaw(TemporalDataDescription desc,
int index)
{
// Note: The reason that we subtract 1 from the index is because we are always one ahead.
// This allows the DELTA change formatter to work. DELTA change requires two timeslices,
// so we have to be one ahead. RAW only requires one, so we shift backwards.
TemporalPoint point = _points[index-1];
return point[desc.Index];
}
private double xb45e1635c92024eb(TemporalDataDescription x3253f43e16d3c2a3, int xc0c4c459c6ccbd00)
{
if (xc0c4c459c6ccbd00 == 0)
{
return 0.0;
}
TemporalPoint point = this._x6fa2570084b2ad39[xc0c4c459c6ccbd00];
TemporalPoint point2 = this._x6fa2570084b2ad39[xc0c4c459c6ccbd00 - 1];
return (point[x3253f43e16d3c2a3.Index] - point2[x3253f43e16d3c2a3.Index]);
}
private double x8c7fc30b887213d1(TemporalDataDescription x3253f43e16d3c2a3, int xc0c4c459c6ccbd00)
{
double[] d = new double[1];
switch (x3253f43e16d3c2a3.DescriptionType)
{
case TemporalDataDescription.Type.Raw:
d[0] = this.x0f80f6735d6ae7a3(x3253f43e16d3c2a3, xc0c4c459c6ccbd00);
break;
case TemporalDataDescription.Type.PercentChange:
d[0] = this.xea9b27496fd731f6(x3253f43e16d3c2a3, xc0c4c459c6ccbd00);
if (0 == 0)
{
break;
}
goto Label_009C;
case TemporalDataDescription.Type.DeltaChange:
d[0] = this.xb45e1635c92024eb(x3253f43e16d3c2a3, xc0c4c459c6ccbd00);
break;
default:
goto Label_009C;
}
Label_001C:
if (x3253f43e16d3c2a3.ActivationFunction != null)
{
x3253f43e16d3c2a3.ActivationFunction.ActivationFunction(d, 0, 1);
if (((uint) xc0c4c459c6ccbd00) < 0)
{
goto Label_001C;
}
}
Label_009C:
return d[0];
}
private double x0f80f6735d6ae7a3(TemporalDataDescription x3253f43e16d3c2a3, int xc0c4c459c6ccbd00)
{
TemporalPoint point = this._x6fa2570084b2ad39[xc0c4c459c6ccbd00 - 1];
return point[x3253f43e16d3c2a3.Index];
}
public IMLDataSet GenerateTraining()
{
var result = new TemporalMLDataSet(WindowSize, 1);
var desc = new TemporalDataDescription(TemporalDataDescription.Type.Raw, true, true);
result.AddDescription(desc);
for (int year = TrainStart; year < TrainEnd; year++)
{
var point = new TemporalPoint(1) { Sequence = year };
point.Data[0] = _normalizedSunspots[year];
result.Points.Add(point);
}
result.Generate();
return result;
}
public MarketDataDescription(TickerSymbol ticker, MarketDataType dataType, TemporalDataDescription.Type type, IActivationFunction activationFunction, bool input, bool predict)
: base(activationFunction, type, input, predict)
{
this._x96e4701dec47675e = ticker;
this._xd2f3d36b96df4542 = dataType;
}
/// <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)
{
var result = new TemporalMLDataSet(windowsize, predictsize);
var desc = new TemporalDataDescription(TemporalDataDescription.Type.PercentChange, true, true);
result.AddDescription(desc);
for (int index = 0; index < inputserie.Length - 1; index++)
{
var point = new TemporalPoint(1) {Sequence = index};
point.Data[0] = inputserie[index];
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 (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;
}