public void DFTTestMethod3()
{
DiscreteFourierTransform DFT = new DiscreteFourierTransform();
// test case 2
List <float> Samples = new List <float> {
1, 3, 5, 7, 9, 11, 13, 15
};
DFT.InputTimeDomainSignal = new DSPAlgorithms.DataStructures.Signal(Samples, false);
DFT.InputSamplingFrequency = 4;
var FrequenciesAmplitudes = new List <float> {
64, 20.9050074380220f, 11.3137084989848f, 8.65913760233915f, 8, 8.65913760233915f, 11.3137084989848f, 20.9050074380220f
};
var FrequenciesPhaseShifts = new List <float> {
0, 1.96349540849362f, 2.35619449019235f, 2.74889357189107f, -3.14159265358979f, -2.74889357189107f, -2.35619449019235f, -1.96349540849362f
};
var Frequencies = new List <float> {
0, 1, 2, 3, 4, 5, 6, 7
};
DFT.Run();
Assert.IsTrue(UnitTestUtitlities.SignalsSamplesAreEqual(FrequenciesAmplitudes, DFT.OutputFreqDomainSignal.FrequenciesAmplitudes) &&
UnitTestUtitlities.SignalsPhaseShiftsAreEqual(FrequenciesPhaseShifts, DFT.OutputFreqDomainSignal.FrequenciesPhaseShifts));
}
public void NaiveTransformsRealSineCorrectly()
{
var samples = SignalGenerator.EquidistantPeriodic(w => new Complex(Math.Sin(w), 0), Constants.Pi2, 0, 16);
// real-odd transforms to imaginary odd
var dft = new DiscreteFourierTransform();
var spectrum = dft.NaiveForward(samples, FourierOptions.Matlab);
// all real components must be zero
foreach (var c in spectrum)
{
Assert.AreEqual(0, c.Real, 1e-12, "real");
}
// all imaginary components except second and last musth be zero
for (var i = 0; i < spectrum.Length; i++)
{
if (i == 1)
{
Assert.AreEqual(-8, spectrum[i].Imaginary, 1e-12, "imag second");
}
else if (i == spectrum.Length - 1)
{
Assert.AreEqual(8, spectrum[i].Imaginary, 1e-12, "imag last");
}
else
{
Assert.AreEqual(0, spectrum[i].Imaginary, 1e-12, "imag");
}
}
}
public void NaiveTransformsRealSineCorrectly()
{
var samples = SignalGenerator.EquidistantPeriodic(w => new Complex(Math.Sin(w), 0), Constants.Pi2, 0, 16);
// real-odd transforms to imaginary odd
var dft = new DiscreteFourierTransform();
var spectrum = dft.NaiveForward(samples, FourierOptions.Matlab);
// all real components must be zero
foreach (var c in spectrum)
{
Assert.AreEqual(0, c.Real, 1e-12, "real");
}
// all imaginary components except second and last musth be zero
for (var i = 0; i < spectrum.Length; i++)
{
if (i == 1)
{
Assert.AreEqual(-8, spectrum[i].Imaginary, 1e-12, "imag second");
}
else if (i == spectrum.Length - 1)
{
Assert.AreEqual(8, spectrum[i].Imaginary, 1e-12, "imag last");
}
else
{
Assert.AreEqual(0, spectrum[i].Imaginary, 1e-12, "imag");
}
}
}
public void FourierNaiveIsReversible(FourierOptions options)
{
var dft = new DiscreteFourierTransform();
VerifyIsReversibleComplex(
0x80,
1e-12,
s => dft.NaiveForward(s, options),
s => dft.NaiveInverse(s, options));
}
public void Radix2ThrowsWhenNotPowerOfTwo()
{
var samples = SignalGenerator.Random((u, v) => new Complex(u, v), GetUniform(1), 0x7F);
var dft = new DiscreteFourierTransform();
Assert.Throws(typeof(ArgumentException), () => dft.Radix2Forward(samples, FourierOptions.Default));
Assert.Throws(typeof(ArgumentException), () => dft.Radix2Inverse(samples, FourierOptions.Default));
Assert.Throws(typeof(ArgumentException), () => DiscreteFourierTransform.Radix2(samples, -1));
Assert.Throws(typeof(ArgumentException), () => DiscreteFourierTransform.Radix2Parallel(samples, -1));
}
public BandDetector()
{
this.source = new AudioFileDataSource(320, null);
RaisedCosineWindower raisedCosineWindower = new RaisedCosineWindower(0.97000002861022949, 25.625f, 10f);
DiscreteFourierTransform discreteFourierTransform = new DiscreteFourierTransform(512, false);
MelFrequencyFilterBank melFrequencyFilterBank = new MelFrequencyFilterBank(130.0, 6800.0, 40);
ArrayList arrayList = new ArrayList();
arrayList.add(this.source);
arrayList.add(raisedCosineWindower);
arrayList.add(discreteFourierTransform);
arrayList.add(melFrequencyFilterBank);
this.frontend = new FrontEnd(arrayList);
}
public BandDetector()
{
// standard frontend
_source = new AudioFileDataSource(320, null);
var windower = new RaisedCosineWindower(0.97f, 25.625f, 10.0f);
var fft = new DiscreteFourierTransform(512, false);
var filterbank = new MelFrequencyFilterBank(130.0, 6800.0, Bands);
var list = new List <IDataProcessor> {
_source, windower, fft, filterbank
};
_frontend = new FrontEnd(list);
}
public void FFT_TestMethod2()
{
DiscreteFourierTransform DFT = new DiscreteFourierTransform();
// test case 1 ..
var sig1 = UnitTestUtitlities.LoadSignal("TestingSignals/Signal2_FFT.ds");
var expectedOutput = UnitTestUtitlities.LoadSignal("TestingSignals/Signal2_FFT_Results.ds");
DFT.InputTimeDomainSignal = sig1;
DFT.InputSamplingFrequency = 360;
DFT.Run();
Assert.IsTrue(UnitTestUtitlities.SignalsSamplesAreEqual(expectedOutput.FrequenciesAmplitudes, DFT.OutputFreqDomainSignal.FrequenciesAmplitudes) &&
UnitTestUtitlities.SignalsPhaseShiftsAreEqual(expectedOutput.FrequenciesPhaseShifts, DFT.OutputFreqDomainSignal.FrequenciesPhaseShifts));
}
public void FourierRadix2IsReversible(FourierOptions options)
{
var dft = new DiscreteFourierTransform();
var samples = SignalGenerator.Random((u, v) => new Complex(u, v), GetUniform(1), 0x8000);
var work = new Complex[samples.Length];
samples.CopyTo(work, 0);
dft.Radix2Forward(work, options);
Assert.IsFalse(work.ListAlmostEqual(samples, 6));
dft.Radix2Inverse(work, options);
AssertHelpers.ListAlmostEqual(samples, work, 12);
}
private List <double[]> featureExtraction(List <double[]> sinyal)
{
var jumlahKanal = sinyal.Count();
var jumlahSample = sinyal.First().Length / 2;
var dft = new List <double[]>();
for (int i = 0; i < jumlahKanal; i++)
{
dft.Add(new double[jumlahSample]);
}
Parallel.For(0, jumlahKanal, i => {
dft[i] = new DiscreteFourierTransform().discreteFourierTransform(sinyal[i]).Take(jumlahSample).ToArray();
});
return(dft);
}
public void FourierBluesteinMatchesNaiveOnRandomNonPowerOfTwo(FourierOptions options)
{
var dft = new DiscreteFourierTransform();
var samples = SignalGenerator.Random((u, v) => new Complex(u, v), _uniform, 0x7F);
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveForward(s, options),
s => dft.BluesteinForward(s, options));
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveInverse(s, options),
s => dft.BluesteinInverse(s, options));
}
public void FourierBluesteinMatchesNaiveOnRandomNonPowerOfTwo(FourierOptions options)
{
var dft = new DiscreteFourierTransform();
var samples = SignalGenerator.Random((u, v) => new Complex(u, v), _uniform, 0x7F);
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveForward(s, options),
s => dft.BluesteinForward(s, options));
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveInverse(s, options),
s => dft.BluesteinInverse(s, options));
}
public void DiscreteFourierTransformTest()
{
DiscreteFourierTransform df = new DiscreteFourierTransform();
int i = 1;
FourierTestPair[] testToDo = new FourierTestPair[] {
new FourierTestPair(
new double[,] { { 1.0, 1.0 }, { 1.0, 1.0 } },
new ComplexImage(new double[,] { { 0.0, 0.0 }, { 0.0, 4.0 } } )) // 2x2
,new FourierTestPair(
new double[,] { { 2.0, 2.0 }, { 1.0, 1.0 } },
new ComplexImage(new double[,] { { 0.0, 2.0 }, { 0.0, 6.0 } } )) //2x2
,new FourierTestPair(
new double[,] { { 1.0, 1.0, 1.0 }, { 1.0, 1.0, 1.0 }, { 1.0, 1.0, 1.0 } },
new ComplexImage(new Complex[,] {
{ new Complex(1.0, 0.0), new Complex(1.0, 1.7321), new Complex(1.0, - 1.7321) },
{ new Complex(1.0, 1.7321), new Complex(-2.0, 3.4641), new Complex(4.0, 0) },
{ new Complex(1.0, -1.7321), new Complex(4.0, 0.0), new Complex(-2.0, - 3.4641) } } )) //3x3
,new FourierTestPair(
new double[,] { { 2.0, 2.0, 2.0 }, { 1.0, 1.0, 1.0 }, { 0.0, 0.0, 0.0 } },
new ComplexImage(new Complex[,] {
{ new Complex(1.0, 0.0), new Complex(1.0, 1.7321), new Complex(1.0, - 1.7321) },
{ new Complex(2.5, 0.8660), new Complex(1.0, 5.1962), new Complex(4.0, - 3.4641) },
{ new Complex(2.5, -0.8660), new Complex(4.0, 3.4641), new Complex(1.0, - 5.1962) } } )) //3x3
,new FourierTestPair(
new double[,] { { 3.0, 3.0, 3.0, 3.0 }, { 2.0, 2.0, 2.0, 2.0 }, { 1.0, 1.0, 1.0, 1.0 }, { 0.0, 0.0, 0.0, 0.0 } },
new ComplexImage(new Complex[,] {
{ new Complex( 0.0, 0.0), new Complex(0.0, 0.0), new Complex(8.0, 0.0) , new Complex(0.0, 0.0)},
{ new Complex( 0.0, 0.0), new Complex(0.0, 0.0), new Complex(8.0, 8.0) , new Complex(0.0, 0.0)},
{ new Complex( 0.0, 0.0), new Complex(0.0, 0.0), new Complex(24.0, 0.0) , new Complex(0.0, 0.0)},
{ new Complex( 0.0, 0.0), new Complex(0.0, 0.0), new Complex(8.0, -8.0) , new Complex(0.0, 0.0)}} )) //4x4
};
/*
* TEMPLATE:
*
new FourierTestPair(
new double[,] { { 1.0, 1.0 }, { 1.0, 1.0 } },
new ComplexImage(new double[,] { { 4.0, 0.0 }, { 0.0, 0.0 } } ))
*/
foreach (FourierTestPair test in testToDo)
{
TransformTestTestData(df, i, test.Item1, test.Item2);
i += 1;
}
}
public void FourierBluesteinMatchesNaiveOnRealSineNonPowerOfTwo(FourierOptions options)
{
var dft = new DiscreteFourierTransform();
var samples = SignalGenerator.EquidistantPeriodic(w => new Complex(Math.Sin(w), 0), Constants.Pi2, 0, 14);
VerifyMatchesNaiveComplex(
samples,
12,
s => dft.NaiveForward(s, options),
s => dft.BluesteinForward(s, options));
VerifyMatchesNaiveComplex(
samples,
12,
s => dft.NaiveInverse(s, options),
s => dft.BluesteinInverse(s, options));
}
public void FourierBluesteinMatchesNaiveOnRandomPowerOfTwo([Values(FourierOptions.Default, FourierOptions.Matlab, FourierOptions.NumericalRecipes)] FourierOptions options)
{
var dft = new DiscreteFourierTransform();
var samples = SignalGenerator.Random((u, v) => new Complex(u, v), _uniform, 0x80);
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveForward(s, options),
s => dft.BluesteinForward(s, options));
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveInverse(s, options),
s => dft.BluesteinInverse(s, options));
}
public void FourierBluesteinMatchesNaiveOnRandomPowerOfTwo([Values(FourierOptions.Default, FourierOptions.Matlab, FourierOptions.NumericalRecipes)] FourierOptions options)
{
var dft = new DiscreteFourierTransform();
var samples = SignalGenerator.Random((u, v) => new Complex(u, v), _uniform, 0x80);
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveForward(s, options),
s => dft.BluesteinForward(s, options));
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveInverse(s, options),
s => dft.BluesteinInverse(s, options));
}
public void FourierRadix2MatchesNaiveOnRealSine(FourierOptions options)
{
var dft = new DiscreteFourierTransform();
var samples = SignalGenerator.EquidistantPeriodic(w => new Complex(Math.Sin(w), 0), Constants.Pi2, 0, 16);
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveForward(s, options),
s => dft.Radix2Forward(s, options));
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveInverse(s, options),
s => dft.Radix2Inverse(s, options));
}
public void FourierRadix2MatchesNaiveOnRealSine(FourierOptions options)
{
var dft = new DiscreteFourierTransform();
var samples = SignalGenerator.EquidistantPeriodic(w => new Complex(Math.Sin(w), 0), Constants.Pi2, 0, 16);
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveForward(s, options),
s => dft.Radix2Forward(s, options));
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveInverse(s, options),
s => dft.Radix2Inverse(s, options));
}
public void FourierRadix2MatchesNaiveOnRandom(FourierOptions options)
{
var dft = new DiscreteFourierTransform();
var samples = SignalGenerator.Random((u, v) => new Complex(u, v), _uniform, 0x80);
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveForward(s, options),
s => dft.Radix2Forward(s, options));
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveInverse(s, options),
s => dft.Radix2Inverse(s, options));
}
public void FourierBluesteinIsReversible(FourierOptions options)
{
var dft = new DiscreteFourierTransform();
VerifyIsReversibleComplex(
0x7FFF,
1e-12,
s =>
{
dft.BluesteinForward(s, options);
return s;
},
s =>
{
dft.BluesteinInverse(s, options);
return s;
});
}
public void FourierRadix2IsReversible(FourierOptions options)
{
var dft = new DiscreteFourierTransform();
VerifyIsReversibleComplex(
0x8000,
1e-12,
s =>
{
dft.Radix2Forward(s, options);
return(s);
},
s =>
{
dft.Radix2Inverse(s, options);
return(s);
});
}
public void FourierBluesteinIsReversible(FourierOptions options)
{
var dft = new DiscreteFourierTransform();
VerifyIsReversibleComplex(
0x7FFF,
1e-12,
s =>
{
dft.BluesteinForward(s, options);
return(s);
},
s =>
{
dft.BluesteinInverse(s, options);
return(s);
});
}
/// <summary>
/// Initiate the Series
/// </summary>
/// <param name="Control">Control Options for the FourierLevSet</param>
protected FourierLevSetBase(FourierLevSetControl Control)
{
this.control = Control;
this.FourierP = Control.FourierP;
this.current_samplP = Control.samplP;
this.numFp = FourierP.Length;
this.DomainSize = Control.DomainSize;
this.h_min = Control.h_min;
this.mode = Control.mode;
this.FourierEvolve = Control.FourierEvolve;
// set up DFT
invDFT_coeff = new Complex[numFp];
for (int sp = 0; sp < numFp; sp++)
{
invDFT_coeff[sp] = (Complex)current_samplP[sp];
}
DFT = new DiscreteFourierTransform();
DFT_coeff = DFT.NaiveForward(invDFT_coeff, FourierOptions.Matlab);
}
private static List <double[]> featureExtraction(List <double[]> sinyal)
{
Console.WriteLine("DFT");
var watch = new Stopwatch();
watch.Start();
var jumlahKanal = sinyal.Count();
var jumlahSample = sinyal.First().Length / 2;
var dft = new List <double[]>();
for (int i = 0; i < jumlahKanal; i++)
{
dft.Add(new double[jumlahSample]);
}
Parallel.For(0, jumlahKanal, i => {
dft[i] = new DiscreteFourierTransform().discreteFourierTransform(sinyal[i]).Take(jumlahSample).ToArray();
});
watch.Stop();
Console.WriteLine("Selesai DFT " + watch.ElapsedMilliseconds + " ms");
return(dft);
}
public void TransformTest()
{
var r = new Random();
var t1 = new DiscreteFourierTransform();
var t2 = new FastFourierTransform();
var points = new List <Point>();
var points2 = new List <Point>();
var x = 0;
points.Add(new Point(x++, 1));
points.Add(new Point(x++, 2));
points.Add(new Point(x++, 3));
points.Add(new Point(x, 1));
x = 0;
points2.Add(new Point(x, 1));
points2.Add(new Point(x, 1));
points2.Add(new Point(x, 1));
points2.Add(new Point(x, 1));
points2.Add(new Point(x, 0));
points2.Add(new Point(x, 0));
points2.Add(new Point(x, 0));
points2.Add(new Point(x, 0));
var first = t1.Transform(points.ToArray()).ToList();
var second = t2.Transform(points.ToArray()).ToList();
var f2 = t1.Transform(points2.ToArray()).ToList();
var s2 = t2.Transform(points2.ToArray()).ToList();
for (int i = 0; i < first.Count; i++)
{
Assert.AreEqual(first[i], second[i]);
}
}
private void btnDFT_Click(object sender, EventArgs e)
{
////signal + noise
//double fs = 256; //sampling rate
//double fw = 5; //signal frequency
//double fn = 50; //noise frequency
//double n = 5; //number of periods to show
//double A = 10; //signal amplitude
//double N = 1; //noise amplitude
//int size = (int)(n * fs / fw); //sample size
//var t = Enumerable.Range(1, size).Select(p => p * 1 / fs).ToArray();
chart1.Series[2].Points.Clear();
var yf3 = new DiscreteFourierTransform().discreteFourierTransform(yf2).Take(y.Length / 2).ToArray();
for (int i = 0; i < yf3.Length; i++)
{
//chart1.Series[0].Points.AddY(y[i]);
chart1.Series[2].Points.AddY(yf3[i]);
}
//double[] yf3 = bandpassnarrow.ProcessSamples(y); //Bandpass Narrow
}
public void Radix2ThrowsWhenNotPowerOfTwo()
{
var samples = SignalGenerator.Random((u, v) => new Complex(u, v), GetUniform(1), 0x7F);
var dft = new DiscreteFourierTransform();
Assert.Throws(
typeof (ArgumentException),
() => dft.Radix2Forward(samples, FourierOptions.Default));
Assert.Throws(
typeof (ArgumentException),
() => dft.Radix2Inverse(samples, FourierOptions.Default));
Assert.Throws(
typeof (ArgumentException),
() => DiscreteFourierTransform.Radix2(samples, -1));
Assert.Throws(
typeof (ArgumentException),
() => DiscreteFourierTransform.Radix2Parallel(samples, -1));
}
public void FourierBluesteinMatchesNaiveOnRealSineNonPowerOfTwo([Values(FourierOptions.Default, FourierOptions.Matlab, FourierOptions.NumericalRecipes)] FourierOptions options)
{
var dft = new DiscreteFourierTransform();
var samples = SignalGenerator.EquidistantPeriodic(w => new Complex(Math.Sin(w), 0), Constants.Pi2, 0, 14);
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveForward(s, options),
s => dft.BluesteinForward(s, options));
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveInverse(s, options),
s => dft.BluesteinInverse(s, options));
}
private void bgProses_DoWork(object sender, DoWorkEventArgs e)
{
this.Invoke((MethodInvoker) delegate
{
tlpMain.Enabled = false;
btnOpenDatasets.Enabled = false;
});
_SamplingRate = (int)numSamplingRate.Value;
_ChannelsChoose = tbChannels.Text.Split(',').Select(Int32.Parse).ToArray();
_TriggerTimeStart = (int)numTriggerStart.Value;
_TriggerTimeStop = (int)numTriggerStop.Value;
_EpocSizeTraining = (_TriggerTimeStop - _TriggerTimeStart) * _SamplingRate;
_Artifact = new List <List <bool> >();
_ClassLabel = new List <List <int> >();
_Trigger = new List <List <int> >();
_Data = new List <List <double[]> >();
for (int i = 0; i < _dataPath.Count; i++)
{
ReadArtifactSelection(_artifactPath[i]);
updateAktivitas("Datasets " + Path.GetFileNameWithoutExtension(_dataPath[i]) + " List Artifact Load Done");
}
for (int i = 0; i < _dataPath.Count(); i++)
{
ReadClassLabels(_classLabelPath[i]);
}
for (int i = 0; i < _dataPath.Count; i++)
{
ReadTrigger(_triggerPath[i]);
updateAktivitas("Datasets " + Path.GetFileNameWithoutExtension(_dataPath[i]) + " List Trigger Load Done");
}
for (int i = 0; i < _dataPath.Count; i++)
{
ReadData(_dataPath[i]);
updateAktivitas("Datasets " + Path.GetFileNameWithoutExtension(_dataPath[i]) + " List Data EEG Load Done");
}
int countFile = _Data.Count();
_Subject = new double[countFile][][][];//subject //perStimulus //times //Channels
Parallel.For(0, countFile, i =>
{
var perData = new double[_Trigger[i].Count][][];
Parallel.For(0, _Trigger[i].Count(), j =>
{
var startFrom = _Trigger[i][j] + (_SamplingRate * _TriggerTimeStart);
var endTo = startFrom + (_SamplingRate * (_TriggerTimeStop - _TriggerTimeStart));
var newArrayTemp = new double[endTo - startFrom][];
int l = 0;
for (int k = startFrom; k < endTo; k++)
{
newArrayTemp[l] = _Data[i][k];
l++;
}
var pivotArray = new double[newArrayTemp[0].Length][];
Parallel.For(0, pivotArray.Length, k =>
{
var pivotTimes = new double[newArrayTemp.Length];
Parallel.For(0, pivotTimes.Length, m =>
{
pivotTimes[m] = newArrayTemp[m][k];
});
pivotArray[k] = pivotTimes;
});
perData[j] = pivotArray;
});
_Subject[i] = perData;
});
var ApplyingButterworth = new double[countFile][][][];
Parallel.For(0, ApplyingButterworth.Length, i =>
{
var subjectRow = new double[_Subject[i].Length][][];
Parallel.For(0, subjectRow.Length, j =>
{
var channels = new double[_Subject[i][j].Length][];
Parallel.For(0, channels.Length, k =>
{
var bandpass = new DesignButterworth(Filter.BandPass, 5, 8, 30, _SamplingRate);
bandpass.Input = _Subject[i][j][k];
bandpass.iirInitialization();
bandpass.compute();
channels[k] = bandpass.Output;
});
subjectRow[j] = channels;
});
updateAktivitas("Subject-" + i + " Butterworth Done");
ApplyingButterworth[i] = subjectRow;
});
var ApplyingDFT = new double[countFile][][][];
Parallel.For(0, ApplyingDFT.Length, i =>
{
var subjectRow = new double[ApplyingButterworth[i].Length][][];
Parallel.For(0, subjectRow.Length, j =>
{
var channels = new double[ApplyingButterworth[i][j].Length][];
Parallel.For(0, channels.Length, k =>
{
var jumlahSample = ApplyingButterworth[i][j][k].Length / 2;
channels[k] = new DiscreteFourierTransform().discreteFourierTransform(ApplyingButterworth[i][j][k]).Take(jumlahSample).ToArray();
});
subjectRow[j] = channels;
});
updateAktivitas("Subject-" + i + " DFT Done");
ApplyingDFT[i] = subjectRow;
});
var ExtractFeature = new double[countFile][][];
Parallel.For(0, ExtractFeature.Length, i =>
{
var subjectRow = new double[ApplyingDFT[i].Length][];
Parallel.For(0, subjectRow.Length, j =>
{
subjectRow[j] = dftToOneRowFeature(ApplyingDFT[i][j]);
});
updateAktivitas("Subject-" + i + " Extraction Feature Done");
ExtractFeature[i] = subjectRow;
});
Parallel.For(0, ExtractFeature.Length, i =>
{
for (int j = 0; j < ExtractFeature[i].Length; j++)
{
if (!_Artifact[i][j])
{
saveFeature(ExtractFeature[i][j], Path.GetFileNameWithoutExtension(_dataPath[i]), _ClassLabel[i][j]);
}
}
updateAktivitas("Subject-" + i + " Save Feature Done");
});
this.Invoke((MethodInvoker) delegate
{
tlpMain.Enabled = true;
btnOpenDatasets.Enabled = true;
});
if (bgProses.CancellationPending)
{
e.Cancel = true;
return;
}
}
static void Main(string[] args)
{
opendatasets();
Console.WriteLine("Masukkan Samplingrate : ");
_SamplingRate = Convert.ToInt32(Console.ReadLine());
Console.WriteLine("Channels (Contoh Input : 0,1,2,..,N (Lihat Dokumentasi Datasets)) : ");
_ChannelsChoose = Array.ConvertAll(Console.ReadLine().Split(','), int.Parse);
Console.WriteLine("Masukkan Times Triggers Start : ");
_TriggerTimeStart = Convert.ToInt32(Console.ReadLine());
Console.WriteLine("Masukkan Times Triggers Stop : ");
_TriggerTimeStop = Convert.ToInt32(Console.ReadLine());
_EpocSizeTraining = (_TriggerTimeStop - _TriggerTimeStart) * _SamplingRate;
_Artifact = new List <List <bool> >();
_ClassLabel = new List <List <int> >();
_Trigger = new List <List <int> >();
_Data = new List <List <double[]> >();
for (int i = 0; i < _dataPath.Count; i++)
{
ReadArtifactSelection(_artifactPath[i]);
Console.WriteLine("Datasets " + Path.GetFileNameWithoutExtension(_dataPath[i]) + " List Artifact Load Done");
}
for (int i = 0; i < _dataPath.Count(); i++)
{
ReadClassLabels(_classLabelPath[i]);
}
for (int i = 0; i < _dataPath.Count; i++)
{
ReadTrigger(_triggerPath[i]);
Console.WriteLine("Datasets " + Path.GetFileNameWithoutExtension(_dataPath[i]) + " List Trigger Load Done");
}
for (int i = 0; i < _dataPath.Count; i++)
{
ReadData(_dataPath[i]);
Console.WriteLine("Datasets " + Path.GetFileNameWithoutExtension(_dataPath[i]) + " List Data EEG Load Done");
}
int countFile = _Data.Count();
_Subject = new double[countFile][][][];//subject //perStimulus //times //Channels
Parallel.For(0, countFile, i =>
{
var perData = new double[_Trigger[i].Count][][];
Parallel.For(0, _Trigger[i].Count(), j =>
{
var startFrom = _Trigger[i][j] + (_SamplingRate * _TriggerTimeStart);
var endTo = startFrom + (_SamplingRate * (_TriggerTimeStop - _TriggerTimeStart));
var newArrayTemp = new double[endTo - startFrom][];
int l = 0;
for (int k = startFrom; k < endTo; k++)
{
newArrayTemp[l] = _Data[i][k];
l++;
}
var pivotArray = new double[newArrayTemp[0].Length][];
Parallel.For(0, pivotArray.Length, k =>
{
var pivotTimes = new double[newArrayTemp.Length];
Parallel.For(0, pivotTimes.Length, m =>
{
pivotTimes[m] = newArrayTemp[m][k];
});
pivotArray[k] = pivotTimes;
});
perData[j] = pivotArray;
});
_Subject[i] = perData;
});
var ApplyingButterworth = new double[countFile][][][];
Parallel.For(0, ApplyingButterworth.Length, i =>
{
var subjectRow = new double[_Subject[i].Length][][];
Parallel.For(0, subjectRow.Length, j =>
{
var channels = new double[_Subject[i][j].Length][];
Parallel.For(0, channels.Length, k =>
{
var bandpass = new DesignButterworth(Filter.BandPass, 5, 8, 30, _SamplingRate);
bandpass.Input = _Subject[i][j][k];
bandpass.iirInitialization();
bandpass.compute();
channels[k] = bandpass.Output;
});
subjectRow[j] = channels;
});
Console.WriteLine("Subject-" + i + " Butterworth Done");
ApplyingButterworth[i] = subjectRow;
});
var ApplyingDFT = new double[countFile][][][];
Parallel.For(0, ApplyingDFT.Length, i =>
{
var subjectRow = new double[ApplyingButterworth[i].Length][][];
Parallel.For(0, subjectRow.Length, j =>
{
var channels = new double[ApplyingButterworth[i][j].Length][];
Parallel.For(0, channels.Length, k =>
{
var jumlahSample = ApplyingButterworth[i][j][k].Length / 2;
channels[k] = new DiscreteFourierTransform().discreteFourierTransform(ApplyingButterworth[i][j][k]).Take(jumlahSample).ToArray();
});
subjectRow[j] = channels;
});
Console.WriteLine("Subject-" + i + " DFT Done");
ApplyingDFT[i] = subjectRow;
});
var ExtractFeature = new double[countFile][][];
Parallel.For(0, ExtractFeature.Length, i =>
{
var subjectRow = new double[ApplyingDFT[i].Length][];
Parallel.For(0, subjectRow.Length, j =>
{
subjectRow[j] = dftToOneRowFeature(ApplyingDFT[i][j]);
});
Console.WriteLine("Subject-" + i + " Extraction Feature Done");
ExtractFeature[i] = subjectRow;
});
Parallel.For(0, ExtractFeature.Length, i =>
{
for (int j = 0; j < ExtractFeature[i].Length; j++)
{
if (!_Artifact[i][j])
{
saveFeature(ExtractFeature[i][j], Path.GetFileNameWithoutExtension(_dataPath[i]), _ClassLabel[i][j]);
}
}
Console.WriteLine("Subject-" + i + " Save Feature Done");
});
}
public void FourierRadix2MatchesNaiveOnRandom(FourierOptions options)
{
var dft = new DiscreteFourierTransform();
var samples = SignalGenerator.Random((u, v) => new Complex(u, v), GetUniform(1), 0x80);
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveForward(s, options),
s => dft.Radix2Forward(s, options));
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveInverse(s, options),
s => dft.Radix2Inverse(s, options));
}
static void Main(string[] args)
{
Stopwatch stopwatch = new Stopwatch();
stopwatch.Start();
_Artifact = new List <List <bool> >();
_ClassLabel = new List <List <int> >();
_Trigger = new List <List <int> >();
foreach (string file in Directory.EnumerateFiles("datasets", "*.HDR_ArtifactSelection"))
{
var contents = File.ReadLines(file);
var newArtifact = new List <bool>();
foreach (var row in contents)
{
newArtifact.Add(Convert.ToBoolean(Convert.ToInt16(row)));
}
_Artifact.Add(newArtifact);
}
foreach (string file in Directory.EnumerateFiles("datasets", "*.HDR_Classlabel"))
{
var contents = File.ReadLines(file);
var newLabel = new List <int>();
foreach (var row in contents)
{
if (row == "NaN")
{
newLabel.Add(5);
}
else
{
newLabel.Add(Convert.ToInt16(row));
}
}
_ClassLabel.Add(newLabel);
}
foreach (string file in Directory.EnumerateFiles("datasets", "*.HDR_TRIG"))
{
var contents = File.ReadLines(file);
var newTrigger = new List <int>();
foreach (var row in contents)
{
newTrigger.Add(Convert.ToInt32(row));
}
_Trigger.Add(newTrigger);
}
_Data = new List <List <double[]> >();
foreach (string file in Directory.EnumerateFiles("datasets", "*.data"))
{
var contents = File.ReadLines(file);
var newData = new List <double[]>();
foreach (var row in contents)
{
var timesAllChannels = Array.ConvertAll(row.Split(" ".ToCharArray(), StringSplitOptions.RemoveEmptyEntries), Double.Parse);
var times = new double[] { timesAllChannels[26], timesAllChannels[5], timesAllChannels[9], timesAllChannels[36] };
newData.Add(times);
}
_Data.Add(newData);
}
int countFile = _Data.Count();
_Subject = new double[countFile][][][];//subject //perStimulus //times //Channels
Parallel.For(0, countFile, i =>
{
var perData = new double[_Trigger[i].Count][][];
Parallel.For(0, _Trigger[i].Count(), j =>
{
var startFrom = _Trigger[i][j] + (_SamplingRate * (_FixationCross + _BlankScreen));
var endTo = startFrom + (_SamplingRate * _StimulusShown);
var newArrayTemp = new double[endTo - startFrom][];
int l = 0;
for (int k = startFrom; k < endTo; k++)
{
newArrayTemp[l] = _Data[i][k];
l++;
}
var pivotArray = new double[newArrayTemp[0].Length][];
Parallel.For(0, pivotArray.Length, k =>
{
var pivotTimes = new double[newArrayTemp.Length];
Parallel.For(0, pivotTimes.Length, m =>
{
pivotTimes[m] = newArrayTemp[m][k];
});
pivotArray[k] = pivotTimes;
});
perData[j] = pivotArray;
});
_Subject[i] = perData;
});
var ApplyingButterworth = new double[countFile][][][];
Parallel.For(0, ApplyingButterworth.Length, i =>
{
var subjectRow = new double[_Subject[i].Length][][];
Parallel.For(0, subjectRow.Length, j =>
{
var channels = new double[_Subject[i][j].Length][];
Parallel.For(0, channels.Length, k =>
{
var bandpass = new DesignButterworth(Filter.BandPass, 5, 8, 30, _SamplingRate);
bandpass.Input = _Subject[i][j][k];
bandpass.iirInitialization();
bandpass.compute();
channels[k] = bandpass.Output;
});
subjectRow[j] = channels;
});
Console.WriteLine("Subject-" + i + " Butterworth Done");
ApplyingButterworth[i] = subjectRow;
});
var ApplyingDFT = new double[countFile][][][];
Parallel.For(0, ApplyingDFT.Length, i =>
{
var subjectRow = new double[ApplyingButterworth[i].Length][][];
Parallel.For(0, subjectRow.Length, j =>
{
var channels = new double[ApplyingButterworth[i][j].Length][];
Parallel.For(0, channels.Length, k =>
{
var jumlahSample = ApplyingButterworth[i][j][k].Length / 2;
channels[k] = new DiscreteFourierTransform().discreteFourierTransform(ApplyingButterworth[i][j][k]).Take(jumlahSample).ToArray();
});
subjectRow[j] = channels;
});
Console.WriteLine("Subject-" + i + " DFT Done");
ApplyingDFT[i] = subjectRow;
});
var ExtractFeature = new double[countFile][][];
Parallel.For(0, ExtractFeature.Length, i =>
{
var subjectRow = new double[ApplyingDFT[i].Length][];
Parallel.For(0, subjectRow.Length, j =>
{
subjectRow[j] = dftToOneRowFeature(ApplyingDFT[i][j]);
});
Console.WriteLine("Subject-" + i + " Extraction Feature Done");
ExtractFeature[i] = subjectRow;
});
Parallel.For(0, ExtractFeature.Length, i =>
{
for (int j = 0; j < ExtractFeature[i].Length; j++)
{
if (!_Artifact[i][j])
{
saveFeature(ExtractFeature[i][j], i, _ClassLabel[i][j] - 1);
}
}
Console.WriteLine("Subject-" + i + " Save Feature Done");
});
Console.WriteLine(stopwatch.ElapsedMilliseconds);
Console.ReadKey();
stopwatch.Stop();
}
public void FourierNaiveIsReversible([Values(FourierOptions.Default, FourierOptions.Matlab)] FourierOptions options)
{
var dft = new DiscreteFourierTransform();
VerifyIsReversibleComplex(
0x80,
1e-12,
s => dft.NaiveForward(s, options),
s => dft.NaiveInverse(s, options));
}
public void FourierBluesteinMatchesNaiveOnRealSineNonPowerOfTwo(FourierOptions options)
{
var dft = new DiscreteFourierTransform();
var samples = Sample.EquidistantPeriodic(w => new Complex(Math.Sin(w), 0), Constants.Pi2, 0, 14);
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveForward(s, options),
s => dft.BluesteinForward(s, options));
VerifyMatchesNaiveComplex(
samples,
1e-12,
s => dft.NaiveInverse(s, options),
s => dft.BluesteinInverse(s, options));
}
public void FourierRadix2IsReversible([Values(FourierOptions.Default, FourierOptions.Matlab)] FourierOptions options)
{
var dft = new DiscreteFourierTransform();
VerifyIsReversibleComplex(
0x8000,
1e-12,
s =>
{
dft.Radix2Forward(s, options);
return s;
},
s =>
{
dft.Radix2Inverse(s, options);
return s;
});
}
