public void TestSvc()
{
var clf = new Svc <int>(kernel: Kernel.Linear, probability: true);
clf.Fit(X, Y);
var spClf = new Svc <int>(kernel: Kernel.Linear, probability: true);
spClf.Fit(SparseMatrix.OfMatrix(X), Y);
Assert.IsTrue(spClf.Predict(T).SequenceEqual(true_result));
Assert.IsTrue(spClf.SupportVectors is SparseMatrix);
Assert.IsTrue(clf.SupportVectors.AlmostEquals(spClf.SupportVectors));
Assert.IsTrue(spClf.DualCoef is SparseMatrix);
Assert.IsTrue(clf.DualCoef.AlmostEquals(spClf.DualCoef));
Assert.IsTrue(spClf.Coef is SparseMatrix);
Assert.IsTrue(clf.Coef.AlmostEquals(spClf.Coef));
Assert.IsTrue(clf.Support.SequenceEqual(spClf.Support));
Assert.IsTrue(clf.Predict(T).SequenceEqual(spClf.Predict(T)));
// refit with a different dataset
clf.Fit(X2, Y2);
spClf.Fit(SparseMatrix.OfMatrix(X2), Y2);
Assert.IsTrue(clf.SupportVectors.AlmostEquals(spClf.SupportVectors));
Assert.IsTrue(clf.DualCoef.AlmostEquals(spClf.DualCoef));
Assert.IsTrue(clf.Coef.AlmostEquals(spClf.Coef));
Assert.IsTrue(clf.Support.SequenceEqual(spClf.Support));
Assert.IsTrue(clf.Predict(T2).SequenceEqual(spClf.Predict(T2)));
Assert.IsTrue(clf.PredictProba(T2).AlmostEquals(spClf.PredictProba(T2), 0.001));
}
public void TestDecisionFunction()
{
// multi class:
var clf = new Svc <int>(kernel: Kernel.Linear, c: 0.1);
clf.Fit(iris.Data, iris.Target);
var dec = (iris.Data * clf.Coef.Transpose()).AddRowVector(clf.Intercept);
Assert.IsTrue(dec.AlmostEquals(clf.DecisionFunction(iris.Data)));
// binary:
clf.Fit(X, Y);
dec = (X * clf.Coef.Transpose()).AddRowVector(clf.Intercept);
int[] prediction = clf.Predict(X);
Assert.IsTrue(dec.AlmostEquals(clf.DecisionFunction(X)));
var b = clf.DecisionFunction(X).Column(0).Select(v => clf.Classes[v > 0 ? 1 : 0]);
Assert.IsTrue(prediction.SequenceEqual(b));
var expected = DenseMatrix.OfArray(new[, ] {
{ -1.0 }, { -0.66 }, { -1.0 }, { 0.66 }, { 1.0 }, { 1.0 }
});
Assert.IsTrue(clf.DecisionFunction(X).AlmostEquals(expected, 1E-2));
}
//
private static void SinglePredict(Svc secureSvc, double[] feature, int i, CKKSEncoder encoder, Encryptor encryptor, Decryptor decryptor,
Stopwatch innerProductStopwatch, Stopwatch degreeStopwatch, Stopwatch negateStopwatch, Stopwatch serverDecisionStopWatch, double scale, Result[] results)
{
double finalResult = 0;
Console.WriteLine($"start {i} \n");
var plaintexts = new Plaintext();
var featuresCiphertexts = new Ciphertext();
encoder.Encode(feature, scale, plaintexts);
encryptor.Encrypt(plaintexts, featuresCiphertexts);
// Server side start
var cyphetResult = secureSvc.Predict(featuresCiphertexts, true, true, innerProductStopwatch, degreeStopwatch, negateStopwatch, serverDecisionStopWatch);
// Server side end
//timePredictSum.Stop();
Plaintext plainResult = new Plaintext();
decryptor.Decrypt(cyphetResult, plainResult);
List <double> result = new List <double>();
encoder.Decode(plainResult, result);
finalResult = result[0];
int estimation = finalResult > 0 ? 0 : 1;
Console.WriteLine($"\n ************************************************");
Console.WriteLine($"SVC estimation{i} is : {estimation} , result : {finalResult}");
//file.WriteLine($"{i} , {estimation} , {finalResult} ");
Console.WriteLine($"************************************************ \n");
results[i] = new Result(finalResult, estimation);
//Console.WriteLine($"SecureSVC estimation{i} is : {estimation} , finalResult = {finalResult} , Time = {timePredictSum.ElapsedMilliseconds}");
}
/// <summary>
/// Make sure some tweaking of parameters works.
///
/// We change clf.dual_coef_ at run time and expect .predict() to change
/// accordingly. Notice that this is not trivial since it involves a lot
/// of C/Python copying in the libsvm bindings.
///
/// The success of this test ensures that the mapping between libsvm and
/// the python classifier is complete.
/// </summary>
public void TestTweakParams()
{
var clf = new Svc <int>(kernel: Kernel.Linear, c: 1.0);
clf.Fit(X, Y);
Assert.IsTrue(clf.DualCoef.AlmostEquals(DenseMatrix.OfArray(new[, ] {
{ 0.25, -0.25 }
})));
Assert.IsTrue(clf.Predict(DenseMatrix.OfArray(new[, ] {
{ -.1, -.1 }
})).SequenceEqual(new[] { 1 }));
clf.DualCoef = DenseMatrix.OfArray(new[, ] {
{ 0.0, 1.0 }
});
Assert.IsTrue(clf.Predict(DenseMatrix.OfArray(new[, ] {
{ -.1, -.1 }
})).SequenceEqual(new[] { 2 }));
}
/// <summary>
/// Make some classification predictions on a toy dataset using a SVC
///
/// If binary is True restrict to a binary classification problem instead of a
/// multiclass classification problem
/// </summary>
/// <param name="x"></param>
/// <param name="y"></param>
/// <param name="binary"></param>
private static Tuple <int[], int[], Matrix <double> > MakePrediction(
Matrix <double> x = null,
int[] y = null,
bool binary = false)
{
if (x == null && y == null)
{
// import some data to play with
var dataset = IrisDataset.Load();
x = dataset.Data;
y = dataset.Target;
}
if (binary)
{
// restrict to a binary classification task
x = x.RowsAt(y.Indices(v => v < 2));
y = y.Where(v => v < 2).ToArray();
}
int nSamples = x.RowCount;
int nFeatures = x.ColumnCount;
var rng = new Random(37);
int[] p = Shuffle(rng, Enumerable.Range(0, nSamples).ToArray());
x = x.RowsAt(p);
y = y.ElementsAt(p);
var half = nSamples / 2;
// add noisy features to make the problem harder and avoid perfect results
rng = new Random(0);
x = x.HStack(DenseMatrix.CreateRandom(nSamples, 200, new Normal {
RandomSource = rng
}));
// run classifier, get class probabilities and label predictions
var clf = new Svc <int>(kernel: Kernel.Linear, probability: true);
clf.Fit(x.SubMatrix(0, half, 0, x.ColumnCount), y.Take(half).ToArray());
Matrix <double> probasPred = clf.PredictProba(x.SubMatrix(half, x.RowCount - half, 0, x.ColumnCount));
if (binary)
{
// only interested in probabilities of the positive case
// XXX: do we really want a special API for the binary case?
probasPred = probasPred.SubMatrix(0, probasPred.RowCount, 1, 1);
}
var yPred = clf.Predict(x.SubMatrix(half, x.RowCount - half, 0, x.ColumnCount));
var yTrue = y.Skip(half).ToArray();
return(Tuple.Create(yTrue, yPred, probasPred));
}
public void TestSvcWithCustomKernel()
{
var clfLin = new Svc <int>(kernel: Kernel.Linear);
clfLin.Fit(SparseMatrix.OfMatrix(X), Y);
var clfMylin =
new Svc <int>(kernel: Kernel.FromFunction((x, y) => x * y.Transpose()));
clfMylin.Fit(SparseMatrix.OfMatrix(X), Y);
Assert.IsTrue(
clfLin.Predict(SparseMatrix.OfMatrix(X)).SequenceEqual(clfMylin.Predict(SparseMatrix.OfMatrix(X))));
}
public void TestLibsvmIris()
{
// shuffle the dataset so that labels are not ordered
foreach (var k in new[] { Kernel.Linear, Kernel.Rbf })
{
var clf = new Svc <int>(kernel: k);
clf.Fit(iris.Data, iris.Target);
var pred = clf.Predict(iris.Data);
var matchingN = pred.Zip(iris.Target, Tuple.Create).Where(t => t.Item1 == t.Item2).Count();
Assert.IsTrue(1.0 * matchingN / pred.Length > 0.9);
Assert.IsTrue(clf.Classes.SequenceEqual(clf.Classes.OrderBy(v => v)));
}
}
public void TestSingleSample_1D()
{
var clf = new Svc <int>();
clf.Fit(X, Y);
var p = clf.Predict(X.Row(0).ToRowMatrix());
Assert.AreEqual(Y[0], p[0]);
//todo:
//clf = svm.LinearSVC(random_state=0).fit(X, Y)
//clf.predict(X[0])
}
/// <summary>
/// Make some classification predictions on a toy dataset using a SVC
///
/// If binary is True restrict to a binary classification problem instead of a
/// multiclass classification problem
/// </summary>
/// <param name="x"></param>
/// <param name="y"></param>
/// <param name="binary"></param>
private static Tuple<int[], int[], Matrix<double>> MakePrediction(
Matrix<double> x = null,
int[] y = null,
bool binary = false)
{
if (x == null && y == null)
{
// import some data to play with
var dataset = IrisDataset.Load();
x = dataset.Data;
y = dataset.Target;
}
if (binary)
{
// restrict to a binary classification task
x = x.RowsAt(y.Indices(v => v < 2));
y = y.Where(v => v < 2).ToArray();
}
int nSamples = x.RowCount;
int nFeatures = x.ColumnCount;
var rng = new Random(37);
int[] p = Shuffle(rng, Enumerable.Range(0, nSamples).ToArray());
x = x.RowsAt(p);
y = y.ElementsAt(p);
var half = nSamples/2;
// add noisy features to make the problem harder and avoid perfect results
rng = new Random(0);
x = x.HStack(DenseMatrix.CreateRandom(nSamples, 200, new Normal{RandomSource = rng}));
// run classifier, get class probabilities and label predictions
var clf = new Svc<int>(kernel: Kernel.Linear, probability: true);
clf.Fit(x.SubMatrix(0, half, 0, x.ColumnCount), y.Take(half).ToArray());
Matrix<double> probasPred = clf.PredictProba(x.SubMatrix(half, x.RowCount - half, 0, x.ColumnCount));
if (binary)
{
// only interested in probabilities of the positive case
// XXX: do we really want a special API for the binary case?
probasPred = probasPred.SubMatrix(0, probasPred.RowCount, 1, 1);
}
var yPred = clf.Predict(x.SubMatrix(half, x.RowCount - half, 0, x.ColumnCount));
var yTrue = y.Skip(half).ToArray();
return Tuple.Create(yTrue, yPred, probasPred);
}
public void TestLibsvmParameters()
{
var clf = new Svc <int>(kernel: Kernel.Linear);
clf.Fit(X, Y);
Assert.IsTrue(clf.DualCoef.AlmostEquals(DenseMatrix.OfArray(new[, ] {
{ 0.25, -.25 }
})));
Assert.IsTrue(clf.Support.SequenceEqual(new[] { 1, 3 }));
Assert.IsTrue(
clf.SupportVectors.AlmostEquals(
DenseMatrix.OfRows(2, X.ColumnCount, new[] { X.Row(1), X.Row(3) })));
Assert.IsTrue(clf.Intercept.SequenceEqual(new[] { 0.0 }));
Assert.IsTrue(clf.Predict(X).SequenceEqual(Y));
}
public void TestSvcWithCallableKernel()
{
// create SVM with callable linear kernel, check that results are the same
// as with built-in linear kernel
var svmCallable = new Svc <int>(kernel: Kernel.FromFunction((x, y) => x * (y.Transpose())),
probability: true);
svmCallable.Fit(X, Y);
var svmBuiltin = new Svc <int>(kernel: Kernel.Linear, probability: true);
svmBuiltin.Fit(X, Y);
Assert.IsTrue(svmCallable.DualCoef.AlmostEquals(svmBuiltin.DualCoef));
Assert.IsTrue(svmCallable.Intercept.AlmostEquals(svmBuiltin.Intercept));
Assert.IsTrue(svmCallable.Predict(X).SequenceEqual(svmBuiltin.Predict(X)));
Assert.IsTrue(svmCallable.PredictProba(X).AlmostEquals(svmBuiltin.PredictProba(X), 1));
Assert.IsTrue(svmCallable.DecisionFunction(X).AlmostEquals(svmBuiltin.DecisionFunction(X), 2));
}
public void TestSvcIris()
{
foreach (var k in new[] { Kernel.Linear, Kernel.Poly, Kernel.Rbf })
{
var spClf = new Svc <int>(kernel: k);
spClf.Fit(SparseMatrix.OfMatrix(iris.Data), iris.Target);
var clf = new Svc <int>(kernel: k);
clf.Fit(DenseMatrix.OfMatrix(iris.Data), iris.Target);
Assert.IsTrue(clf.SupportVectors.AlmostEquals(spClf.SupportVectors));
Assert.IsTrue(clf.DualCoef.AlmostEquals(spClf.DualCoef));
Assert.IsTrue(
clf.Predict(DenseMatrix.OfMatrix(iris.Data)).SequenceEqual(
spClf.Predict(SparseMatrix.OfMatrix(iris.Data))));
if (k == Kernel.Linear)
{
Assert.IsTrue(clf.Coef.AlmostEquals(spClf.Coef));
}
}
}
public void test_weight()
{
var clf = new Svc <int>(classWeightEstimator: ClassWeightEstimator <int> .Explicit(new Dictionary <int, double> {
{ 1, 0.1 }
}));
// we give a small weights to class 1
clf.Fit(X, Y);
// so all predicted values belong to class 2
Assert.IsTrue(clf.Predict(X).SequenceEqual(Enumerable.Repeat(2, 6)));
/*
* X_, y_ = make_classification(n_samples=200, n_features=10,
* weights=[0.833, 0.167], random_state=2)
*
* for clf in (linear_model.LogisticRegression(),
* svm.LinearSVC(random_state=0), svm.SVC()):
* clf.set_params(class_weight={0: .1, 1: 10})
* clf.fit(X_[:100], y_[:100])
* y_pred = clf.predict(X_[100:])
* assert_true(f1_score(y_[100:], y_pred) > .3)
* */
}
public void TestLibsvmParameters()
{
var clf = new Svc<int>(kernel: Kernel.Linear);
clf.Fit(X, Y);
Assert.IsTrue(clf.DualCoef.AlmostEquals(DenseMatrix.OfArray(new[,] {{0.25, -.25}})));
Assert.IsTrue(clf.Support.SequenceEqual(new[] {1, 3}));
Assert.IsTrue(
clf.SupportVectors.AlmostEquals(
DenseMatrix.OfRows(2, X.ColumnCount, new[] {X.Row(1), X.Row(3)})));
Assert.IsTrue(clf.Intercept.SequenceEqual(new[] {0.0}));
Assert.IsTrue(clf.Predict(X).SequenceEqual(Y));
}
public void TestSvcIris()
{
foreach (var k in new[] {Kernel.Linear, Kernel.Poly, Kernel.Rbf})
{
var spClf = new Svc<int>(kernel: k);
spClf.Fit(SparseMatrix.OfMatrix(iris.Data), iris.Target);
var clf = new Svc<int>(kernel: k);
clf.Fit(DenseMatrix.OfMatrix(iris.Data), iris.Target);
Assert.IsTrue(clf.SupportVectors.AlmostEquals(spClf.SupportVectors));
Assert.IsTrue(clf.DualCoef.AlmostEquals(spClf.DualCoef));
Assert.IsTrue(
clf.Predict(DenseMatrix.OfMatrix(iris.Data)).SequenceEqual(
spClf.Predict(SparseMatrix.OfMatrix(iris.Data))));
if (k == Kernel.Linear)
{
Assert.IsTrue(clf.Coef.AlmostEquals(spClf.Coef));
}
}
}
public void TestSvcWithCustomKernel()
{
var clfLin = new Svc<int>(kernel: Kernel.Linear);
clfLin.Fit(SparseMatrix.OfMatrix(X), Y);
var clfMylin =
new Svc<int>(kernel: Kernel.FromFunction((x, y) => x*y.Transpose()));
clfMylin.Fit(SparseMatrix.OfMatrix(X), Y);
Assert.IsTrue(
clfLin.Predict(SparseMatrix.OfMatrix(X)).SequenceEqual(clfMylin.Predict(SparseMatrix.OfMatrix(X))));
}
public void TestSvcWithCallableKernel()
{
// create SVM with callable linear kernel, check that results are the same
// as with built-in linear kernel
var svmCallable = new Svc<int>(kernel: Kernel.FromFunction((x, y) => x*(y.Transpose())),
probability: true);
svmCallable.Fit(X, Y);
var svmBuiltin = new Svc<int>(kernel: Kernel.Linear, probability: true);
svmBuiltin.Fit(X, Y);
Assert.IsTrue(svmCallable.DualCoef.AlmostEquals(svmBuiltin.DualCoef));
Assert.IsTrue(svmCallable.Intercept.AlmostEquals(svmBuiltin.Intercept));
Assert.IsTrue(svmCallable.Predict(X).SequenceEqual(svmBuiltin.Predict(X)));
Assert.IsTrue(svmCallable.PredictProba(X).AlmostEquals(svmBuiltin.PredictProba(X), 1));
Assert.IsTrue(svmCallable.DecisionFunction(X).AlmostEquals(svmBuiltin.DecisionFunction(X), 2));
}
public void test_weight()
{
var clf = new Svc<int>(classWeightEstimator: ClassWeightEstimator<int>.Explicit(new Dictionary<int, double> {{1, 0.1}}));
// we give a small weights to class 1
clf.Fit(X, Y);
// so all predicted values belong to class 2
Assert.IsTrue(clf.Predict(X).SequenceEqual(Enumerable.Repeat(2, 6)));
/*
X_, y_ = make_classification(n_samples=200, n_features=10,
weights=[0.833, 0.167], random_state=2)
for clf in (linear_model.LogisticRegression(),
svm.LinearSVC(random_state=0), svm.SVC()):
clf.set_params(class_weight={0: .1, 1: 10})
clf.fit(X_[:100], y_[:100])
y_pred = clf.predict(X_[100:])
assert_true(f1_score(y_[100:], y_pred) > .3)
* */
}
public void TestDecisionFunction()
{
// multi class:
var clf = new Svc<int>(kernel: Kernel.Linear, c: 0.1);
clf.Fit(iris.Data, iris.Target);
var dec = (iris.Data*clf.Coef.Transpose()).AddRowVector(clf.Intercept);
Assert.IsTrue(dec.AlmostEquals(clf.DecisionFunction(iris.Data)));
// binary:
clf.Fit(X, Y);
dec = (X*clf.Coef.Transpose()).AddRowVector(clf.Intercept);
int[] prediction = clf.Predict(X);
Assert.IsTrue(dec.AlmostEquals(clf.DecisionFunction(X)));
var b = clf.DecisionFunction(X).Column(0).Select(v => clf.Classes[v > 0 ? 1 : 0]);
Assert.IsTrue(prediction.SequenceEqual(b));
var expected = DenseMatrix.OfArray(new[,] {{-1.0}, {-0.66}, {-1.0}, {0.66}, {1.0}, {1.0}});
Assert.IsTrue(clf.DecisionFunction(X).AlmostEquals(expected, 1E-2));
}
/// <summary>
/// Make sure some tweaking of parameters works.
///
/// We change clf.dual_coef_ at run time and expect .predict() to change
/// accordingly. Notice that this is not trivial since it involves a lot
/// of C/Python copying in the libsvm bindings.
///
/// The success of this test ensures that the mapping between libsvm and
/// the python classifier is complete.
/// </summary>
public void TestTweakParams()
{
var clf = new Svc<int>(kernel: Kernel.Linear, c: 1.0);
clf.Fit(X, Y);
Assert.IsTrue(clf.DualCoef.AlmostEquals(DenseMatrix.OfArray(new[,]{{0.25, -0.25}})));
Assert.IsTrue(clf.Predict(DenseMatrix.OfArray(new[,] {{-.1, -.1}})).SequenceEqual(new[] {1}));
clf.DualCoef = DenseMatrix.OfArray(new[,] {{0.0, 1.0}});
Assert.IsTrue(clf.Predict(DenseMatrix.OfArray(new[,] {{-.1, -.1}})).SequenceEqual(new[] {2}));
}
static async Task Main(string[] args)
{
Console.WriteLine("Secure Iris");
// Get Input from resource file or as args
double[][] features;
int numberOfFeatures = 4;
int numOfRows = 0;
// Option to load from args and not the whole dataset that is stored in resources
if (args.Length >= numberOfFeatures)
{
numOfRows = args.Length / numberOfFeatures;
// Features:
features = new double[numOfRows][];
for (int i = 0; i < numOfRows; i++)
{
features[i] = new double[numberOfFeatures];
}
for (int i = 0, l = args.Length; i < l; i++)
{
features[i / numberOfFeatures][i % numberOfFeatures] = Double.Parse(args[i]);
}
}
else // load the whole dataset from resources
{
List <double[]> rows = new List <double[]>();
var bytes = Properties.Resources.iris;
numOfRows = 0;
features = SVCUtilities.SvcUtilities.LoadFeatures(bytes, numberOfFeatures, ref numOfRows);
}
Stopwatch clientStopwatch = new Stopwatch();
clientStopwatch.Start();
//svm algorithm parametrs calculated in python : training result
double[][] vectors = new double[3][];
vectors[0] = new[] { 4.5, 2.3, 1.3, 0.3 };
vectors[1] = new[] { 5.1, 3.3, 1.7, 0.5 };
vectors[2] = new[] { 5.1, 2.5, 3.0, 1.1 };
double[][] coefficients = new double[1][];
coefficients[0] = new double[] { -0.07724840262003278, -0.6705185831514366, 0.7477669857714694 };
double[] intercepts = { 1.453766563649063 };
// SEAL parameters client side
Console.WriteLine("SecureSVC : ");
EncryptionParameters parms = new EncryptionParameters(SchemeType.CKKS);
ulong polyModulusDegree = 16384;
int power = 40;
double scale = Math.Pow(2.0, power);
if (power >= 20 && power < 40)
{
parms.CoeffModulus = CoeffModulus.Create(polyModulusDegree,
new int[] { 60, 20, 21, 22, 23, 24, 25, 26, 27, 60 });
}
else if (power >= 40 && power < 60)
{
parms.CoeffModulus = CoeffModulus.Create(polyModulusDegree,
new int[] { 60, 40, 40, 40, 40, 40, 40, 40, 60 });
}
else if (power == 60)
{
polyModulusDegree = 32768;
parms.CoeffModulus = CoeffModulus.Create(polyModulusDegree,
new int[] { 60, 60, 60, 60, 60, 60, 60, 60, 60 });
}
parms.PolyModulusDegree = polyModulusDegree;
var context = new SEALContext(parms);
// Key generation
KeyGenerator keygen = new KeyGenerator(context);
var publicKey = keygen.PublicKey;
var secretKey = keygen.SecretKey;
var relinKeys = keygen.RelinKeys();
var galoisKeys = keygen.GaloisKeys();
var encryptor = new Encryptor(context, publicKey);
var decryptor = new Decryptor(context, secretKey);
var encoder = new CKKSEncoder(context);
clientStopwatch.Stop();
using (System.IO.StreamWriter file =
new System.IO.StreamWriter(
$@"{OutputDir}IrisLinearSecured_{IsParallel}_{DateTime.Now.Day}_{DateTime.Now.ToShortTimeString().ToString().Replace(":", "_")}.txt")
)
{
// Only CONCEPT demonstation how to parrallel all the computation on all machine cpu's
// Though the parallel here is done on the client side , in "real life" this parallel mechanism
// Should on the server side
if (IsParallel)
{
int processorCount = Environment.ProcessorCount;
Console.WriteLine("Number Of Logical Processors: {0}", processorCount);
Svc[] machines = new Svc[processorCount];
Stopwatch[] innerProductStopwatchArr = new Stopwatch[processorCount];
Stopwatch[] negateStopwatchArr = new Stopwatch[processorCount];
Stopwatch[] degreeStopwatchArr = new Stopwatch[processorCount];
Stopwatch[] serverDecisionStopWatchArr = new Stopwatch[processorCount];
Result[] results = new Result[numOfRows];
Task[] tasks = new Task[processorCount];
for (int i = 0; i < processorCount; i++)
{
machines[i] = new Svc(vectors, coefficients, intercepts, "Linear", 0.25, 0.0, 3, 40, publicKey /*, secretKey*/, relinKeys, galoisKeys, 1, 4);
innerProductStopwatchArr[i] = new Stopwatch();
negateStopwatchArr[i] = new Stopwatch();
degreeStopwatchArr[i] = new Stopwatch();
serverDecisionStopWatchArr[i] = new Stopwatch();
}
Stopwatch totalTime = new Stopwatch();
totalTime.Start();
for (int i = 0; i < numOfRows;)
{
for (int j = 0; j < processorCount && i < numOfRows; j++)
{
var secureSvc = machines[i % processorCount];
var feature = features[i];
//Console.WriteLine($"\n\n $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$");
List <object> l = new List <object>();
l.Add(secureSvc); //0
l.Add(feature); //1
l.Add(i); //2
l.Add(encoder); //3
l.Add(encryptor); //4
l.Add(decryptor); //5
l.Add(innerProductStopwatchArr[i % processorCount]);
l.Add(degreeStopwatchArr[i % processorCount]);
l.Add(negateStopwatchArr[i % processorCount]);
l.Add(serverDecisionStopWatchArr[i % processorCount]);
l.Add(results);
tasks[j] = new TaskFactory().StartNew(new Action <object>((test) =>
{
List <object> l2 = (List <object>)test;
SinglePredict((Svc)l2[0], (double[])l2[1], (int)l2[2], (CKKSEncoder)l2[3], (Encryptor)l2[4], (Decryptor)l2[5], (Stopwatch)l2[6], (Stopwatch)l2[7], (Stopwatch)l2[8], (Stopwatch)l2[9], scale, (Result[])l2[10]);
}), l);
i++;
}
await Task.WhenAll(tasks);
}
totalTime.Stop();
for (int i = 0; i < numOfRows; i++)
{
var result = results[i];
file.WriteLine($"{i} , {result.Estimation} , {result.TotalValue} ");
}
double innerProductTime = 0;
double degreeTime = 0;
double negateTime = 0;
double serverTime = 0;
for (int i = 0; i < processorCount; i++)
{
innerProductTime = innerProductStopwatchArr[i].ElapsedMilliseconds;
degreeTime = degreeStopwatchArr[i].ElapsedMilliseconds;
negateTime = negateStopwatchArr[i].ElapsedMilliseconds;
serverTime = serverDecisionStopWatchArr[i].ElapsedMilliseconds;
}
file.WriteLine($" Client time : {clientStopwatch.ElapsedMilliseconds} ms ");
file.WriteLine($" Total time for {numOfRows} samples : {totalTime.ElapsedMilliseconds} ms ");
file.WriteLine($" Avg time : {totalTime.ElapsedMilliseconds * 1000 / numOfRows} microSec ");
file.WriteLine($" Inner Product time for {numOfRows} samples : {innerProductTime} ms ");
file.WriteLine($" Inner Product Avg time : {innerProductTime * 1000 / numOfRows} microSec ");
file.WriteLine($" Degree time for {numOfRows} samples : {degreeTime} ms ");
file.WriteLine($" Degree Avg time : {degreeTime * 1000 / numOfRows} microSec ");
file.WriteLine($" Negate time for {numOfRows} samples : {negateTime} ms ");
file.WriteLine($" Negate Avg time : {negateTime * 1000 / numOfRows} microSec ");
file.WriteLine($" Decision time for {numOfRows} samples : {serverTime} ms ");
file.WriteLine($" Decision Avg time : {serverTime * 1000 / numOfRows} microSec ");
}
else
{
//Initiate Stopwatch for performance measure
Stopwatch innerProductStopwatch = new Stopwatch();
Stopwatch negateStopwatch = new Stopwatch();
Stopwatch degreeStopwatch = new Stopwatch();
Stopwatch serverDecisionStopWatch = new Stopwatch();
int featureSizeWithSpace = numberOfFeatures;
int batchSize = 200;
if (batchSize > 1)
{
featureSizeWithSpace = numberOfFeatures * 2;
}
Svc clf = new Svc(vectors, coefficients, intercepts, "Linear", 0.25, 0.0, 3, 40, publicKey /*, secretKey*/, relinKeys, galoisKeys, batchSize, featureSizeWithSpace);
Stopwatch totalTime = new Stopwatch();
totalTime.Start();
int start = 0;
//double[] batchFeatures = new double[batchSize * featureSizeWithSpace];
for (int i = 0; i < numOfRows;)
{
start = i;
double finalResult = -10000;
double[][] batchRows = new double[batchSize][];
for (int j = 0; j < batchSize && i < numOfRows; j++)
{
batchRows[j] = features[i];
i++;
}
double[] batchFeatures = GetBatchFeatures(batchRows, batchSize, numberOfFeatures, featureSizeWithSpace);
var plaintexts = new Plaintext();
var featuresCiphertexts = new Ciphertext();
encoder.Encode(batchFeatures, scale, plaintexts);
encryptor.Encrypt(plaintexts, featuresCiphertexts);
//Server side start
var cypherResult = clf.Predict(featuresCiphertexts, true, true, innerProductStopwatch, degreeStopwatch, negateStopwatch, serverDecisionStopWatch);
// Server side end
Plaintext plainResult = new Plaintext();
decryptor.Decrypt(cypherResult, plainResult);
List <double> result = new List <double>();
encoder.Decode(plainResult, result);
for (int j = 0; j < batchSize && start < numOfRows; j++)
{
finalResult = result[j * featureSizeWithSpace];
int estimation = finalResult > 0 ? 0 : 1;
Console.WriteLine($"\n ************************************************");
Console.WriteLine($"SVC estimation{i} is : {estimation} , result : {finalResult}");
file.WriteLine($"{start} , {estimation} , {finalResult} ");
Console.WriteLine($"************************************************ \n");
start++;
}
}
totalTime.Stop();
file.WriteLine($" Client time : {clientStopwatch.ElapsedMilliseconds} ms ");
file.WriteLine($" Total time for {numOfRows} samples : {totalTime.ElapsedMilliseconds} ms ");
file.WriteLine($" Avg time : {totalTime.ElapsedMilliseconds * 1000 / numOfRows} microSec ");
file.WriteLine($" Inner Product time for {numOfRows} samples : {innerProductStopwatch.ElapsedMilliseconds} ms ");
file.WriteLine($" Inner Product Avg time : {innerProductStopwatch.ElapsedMilliseconds * 1000 / numOfRows} microSec ");
file.WriteLine($" Degree time for {numOfRows} samples : {degreeStopwatch.ElapsedMilliseconds} ms ");
file.WriteLine($" Degree Avg time : {degreeStopwatch.ElapsedMilliseconds * 1000 / numOfRows} microSec ");
file.WriteLine($" Negate time for {numOfRows} samples : {negateStopwatch.ElapsedMilliseconds} ms ");
file.WriteLine($" Negate Avg time : {negateStopwatch.ElapsedMilliseconds * 1000 / numOfRows} microSec ");
file.WriteLine($" Decision time for {numOfRows} samples : {serverDecisionStopWatch.ElapsedMilliseconds} ms ");
file.WriteLine($" Decision Avg time : {serverDecisionStopWatch.ElapsedMilliseconds * 1000 / numOfRows} microSec ");
}
}
}
public void TestSvc()
{
var clf = new Svc<int>(kernel: Kernel.Linear, probability: true);
clf.Fit(X, Y);
var spClf = new Svc<int>(kernel: Kernel.Linear, probability: true);
spClf.Fit(SparseMatrix.OfMatrix(X), Y);
Assert.IsTrue(spClf.Predict(T).SequenceEqual(true_result));
Assert.IsTrue(spClf.SupportVectors is SparseMatrix);
Assert.IsTrue(clf.SupportVectors.AlmostEquals(spClf.SupportVectors));
Assert.IsTrue(spClf.DualCoef is SparseMatrix);
Assert.IsTrue(clf.DualCoef.AlmostEquals(spClf.DualCoef));
Assert.IsTrue(spClf.Coef is SparseMatrix);
Assert.IsTrue(clf.Coef.AlmostEquals(spClf.Coef));
Assert.IsTrue(clf.Support.SequenceEqual(spClf.Support));
Assert.IsTrue(clf.Predict(T).SequenceEqual(spClf.Predict(T)));
// refit with a different dataset
clf.Fit(X2, Y2);
spClf.Fit(SparseMatrix.OfMatrix(X2), Y2);
Assert.IsTrue(clf.SupportVectors.AlmostEquals(spClf.SupportVectors));
Assert.IsTrue(clf.DualCoef.AlmostEquals(spClf.DualCoef));
Assert.IsTrue(clf.Coef.AlmostEquals(spClf.Coef));
Assert.IsTrue(clf.Support.SequenceEqual(spClf.Support));
Assert.IsTrue(clf.Predict(T2).SequenceEqual(spClf.Predict(T2)));
Assert.IsTrue(clf.PredictProba(T2).AlmostEquals(spClf.PredictProba(T2), 0.001));
}
static void Main(string[] args)
{
Console.WriteLine("Plain Iris");
double[][] features;
int numOfRows = 0;
// Option to load from args and not the whole dataset that is stored in resources
if (args.Length >= 4)
{
numOfRows = args.Length / 4;
// Features:
features = new double[numOfRows][];
for (int i = 0; i < numOfRows; i++)
{
features[i] = new double[4];
}
for (int i = 0, l = args.Length; i < l; i++)
{
features[i / 4][i % 4] = Double.Parse(args[i]);
}
}
else // load the whole dataset from resources
{
List <double[]> rows = new List <double[]>();
var bytes = Properties.Resources.iris;
numOfRows = 0;
features = SVCUtilities.SvcUtilities.LoadFeatures(bytes, 4, ref numOfRows);
}
// The support vectors and the coefficients of the Iris classification algorithm
double[][] vectors = new double[3][];
vectors[0] = new[] { 4.5, 2.3, 1.3, 0.3 };
vectors[1] = new[] { 5.1, 3.3, 1.7, 0.5 };
vectors[2] = new[] { 5.1, 2.5, 3.0, 1.1 };
double[][] coefficients = new double[1][];
coefficients[0] = new double[] { -0.0005100630977269122, -0.008885899026071108, 0.009395962123798021 };
double[] intercepts = { 1.1358388232934824 };
int[] weights = { 2, 1 };
Console.WriteLine("SVC : ");
// Estimator constructor
Svc clf = new Svc(2, 2, vectors, coefficients, intercepts, weights, "Poly", 0.25, 0.0, 3);
using (System.IO.StreamWriter file =
new System.IO.StreamWriter(
$@"{OutputDir}IrisPolyPlain_{DateTime.Now.Day}_{DateTime.Now.ToShortTimeString().ToString().Replace(":","_")}.txt")
)
{
Stopwatch totalTime = new Stopwatch();
totalTime.Start();
//Run prediction onthe loaded samples
for (int i = 0; i < numOfRows; i++)
{
double finalResult = -10000;
int estimation = clf.Predict(features[i], out finalResult);
Console.WriteLine($"\n ************************************************");
Console.WriteLine($"SVC estimation{i} is : {estimation} , result : {finalResult}");
//write classification result to file
file.WriteLine($"{i} , {estimation} , {finalResult} ");
Console.WriteLine($"************************************************ \n");
}
totalTime.Stop();
//Write performance data to file
file.WriteLine($" Total time for {numOfRows} samples : {totalTime.ElapsedMilliseconds} ms ");
file.WriteLine($" Avg time : {totalTime.ElapsedMilliseconds *1000 / numOfRows} microSec ");
}
}
static void Main(string[] args)
{
Console.WriteLine("Plain Mashroom");
double[,] vectorsMashroomD = { { -0.8403433999584713, 0.9532703900465632, -0.9838993878642219, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.05347685426934428, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, 0.2843298100961381, -0.8771690980243297 }, { -0.8403433999584713, 0.14012794477794924, -1.376724168039617, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, 3.4332552519568216, 0.9480808566164142, 1.4284264144452141, 0.2843298100961381, 1.448588654048012 }, { 1.029712237713905, 0.9532703900465632, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.6184260923501674, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.8771690980243297 }, { -0.21699152073434588, -1.4861569457592787, -1.376724168039617, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, 2.278612115023465, 1.494682719884397, 1.4658499494714023, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, -2.3724904782364917, 0.622441390325499, -0.94101658068771, 0.0, 0.14203663498716684, -0.2561317410190009, -1.2722157358170676, 1.4284264144452141, 0.2843298100961381, 0.28570977801184116 }, { -0.21699152073434588, -1.4861569457592787, -1.376724168039617, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, 2.278612115023465, 1.494682719884397, 1.4658499494714023, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, -0.94101658068771, 0.0, 0.14203663498716684, -0.2561317410190009, -1.2722157358170676, 1.4284264144452141, 0.2843298100961381, 0.28570977801184116 }, { -2.087047158406722, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, -1.970316152682646, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.2289977647710673, -1.1448057498013176, 0.8389893256787506, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, -0.5143892000079515, -0.2957296600062443 }, { 1.029712237713905, -1.4861569457592787, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.8771690980243297 }, { -0.8403433999584713, 0.9532703900465632, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.05347685426934428, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.8771690980243297 }, { 1.029712237713905, 0.14012794477794924, -0.5910746076888268, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, 2.278612115023465, -0.6690383093994308, 0.6184260923501674, 0.8735106372181927, 1.7814601887614623, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, -1.2722157358170676, -0.250470603921817, -0.5143892000079515, -0.2957296600062443 }, { -0.8403433999584713, -1.4861569457592787, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.6184260923501674, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.8771690980243297 }, { -0.21699152073434588, 0.14012794477794924, 1.3730492931881484, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, 2.278612115023465, -0.6690383093994308, 1.4658499494714023, -1.1448057498013176, -1.0459524004866725, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, 3.4332552519568216, 0.9480808566164142, 1.4284264144452141, -1.313108210112041, -0.2957296600062443 }, { 1.029712237713905, -1.4861569457592787, -0.19824982751343181, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, 0.6184260923501674, -1.1448057498013176, 1.7814601887614623, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 1.0830488202002277, 2.0300280920660976 }, { -0.8403433999584713, 0.14012794477794924, 1.7658740733635434, 1.185916567160356, -1.970316152682646, 0.16289645171177966, 2.278612115023465, 1.494682719884397, 0.05347685426934428, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, 0.2843298100961381, -0.8771690980243297 }, { -0.8403433999584713, 0.9532703900465632, 0.1945749526619632, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, 3.4332552519568216, 0.9480808566164142, 1.4284264144452141, 0.2843298100961381, 1.448588654048012 }, { 1.029712237713905, 0.14012794477794924, -0.19824982751343181, -0.8432296400028142, 0.40656202865404567, -6.1388691373667745, -0.4388636369510842, -0.6690383093994308, 1.7483245685118138, -1.1448057498013176, -1.0459524004866725, 0.6837776537937139, 0.5863846591895536, -0.42928778084845187, -0.4166805929300313, 0.0, -3.9790548744261973, -0.2561317410190009, 0.9480808566164142, -1.5096433676970904, 0.2843298100961381, 0.28570977801184116 }, { -0.8403433999584713, 0.9532703900465632, 0.1945749526619632, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, 3.4332552519568216, 0.9480808566164142, 1.4284264144452141, 1.0830488202002277, 1.448588654048012 }, { 1.029712237713905, -1.4861569457592787, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, -1.4810169520224028, -1.4653525684453887, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.8771690980243297 }, { 1.029712237713905, 0.14012794477794924, -0.5910746076888268, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, 2.278612115023465, -0.6690383093994308, 0.6184260923501674, 0.8735106372181927, 1.7814601887614623, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, -1.2722157358170676, -0.250470603921817, -2.9105462303202203, -0.2957296600062443 }, { 1.029712237713905, -1.4861569457592787, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.8771690980243297 }, { -0.8403433999584713, 0.9532703900465632, -1.376724168039617, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, 2.278612115023465, 1.494682719884397, 1.4658499494714023, -1.1448057498013176, -0.10348153740396093, -2.534652044595703, 0.5863846591895536, 0.622441390325499, -0.94101658068771, 0.0, 0.14203663498716684, -0.2561317410190009, -1.2722157358170676, 1.4284264144452141, 0.2843298100961381, 0.28570977801184116 }, { -2.087047158406722, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, -1.970316152682646, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.05347685426934428, -1.1448057498013176, 0.8389893256787506, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, -0.5143892000079515, -0.2957296600062443 }, { 1.029712237713905, -1.4861569457592787, -0.19824982751343181, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, 0.6184260923501674, -1.1448057498013176, 1.7814601887614623, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, 2.0300280920660976 }, { -2.087047158406722, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, -0.5441892438806311, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.05347685426934428, -1.1448057498013176, 0.8389893256787506, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, -0.5143892000079515, 0.8671492160299267 }, { -0.8403433999584713, 0.14012794477794924, -0.19824982751343181, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, -1.1448057498013176, -0.10348153740396093, 2.2929925029884224, 2.065822227902576, -0.9551523664354273, -0.94101658068771, 0.0, 0.14203663498716684, 3.4332552519568216, 0.9480808566164142, 1.4284264144452141, 1.0830488202002277, -0.8771690980243297 }, { -0.8403433999584713, 0.9532703900465632, 1.3730492931881484, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, 1.4658499494714023, -1.1448057498013176, -1.0459524004866725, 0.6837776537937139, -2.3724904782364917, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, -0.7171415877086972, -1.0899191131053325, 0.2843298100961381, -0.8771690980243297 }, { -0.8403433999584713, -1.4861569457592787, -0.9838993878642219, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.05347685426934428, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, 0.2843298100961381, -0.8771690980243297 }, { 1.029712237713905, -1.4861569457592787, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.05347685426934428, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, 0.2843298100961381, -0.8771690980243297 }, { 1.029712237713905, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, -1.970316152682646, 0.16289645171177966, 2.278612115023465, 1.494682719884397, 0.05347685426934428, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, 0.2843298100961381, -0.8771690980243297 }, { -0.8403433999584713, -1.4861569457592787, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.1833753304309906, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.8771690980243297 }, { -0.8403433999584713, 0.9532703900465632, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, -1.4810169520224028, -1.4653525684453887, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.8771690980243297 }, { -0.8403433999584713, 0.14012794477794924, 1.3730492931881484, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, 2.278612115023465, -0.6690383093994308, 0.6184260923501674, 0.8735106372181927, 1.7814601887614623, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, -1.2722157358170676, -0.250470603921817, -0.5143892000079515, -0.2957296600062443 }, { 1.029712237713905, 0.9532703900465632, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, -1.4810169520224028, -1.4653525684453887, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.8771690980243297 }, { 1.029712237713905, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, -1.970316152682646, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, -1.1448057498013176, 0.8389893256787506, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, -0.5143892000079515, -0.2957296600062443 }, { -0.8403433999584713, 0.14012794477794924, -0.5910746076888268, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, 2.278612115023465, -0.6690383093994308, 0.6184260923501674, 0.8735106372181927, 1.7814601887614623, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, -1.2722157358170676, -0.250470603921817, -2.9105462303202203, -0.2957296600062443 }, { 1.029712237713905, 0.9532703900465632, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.1833753304309906, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.8771690980243297 }, { -0.8403433999584713, 0.14012794477794924, -0.5910746076888268, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, 2.278612115023465, -0.6690383093994308, 0.6184260923501674, 0.8735106372181927, 1.7814601887614623, 0.6837776537937139, -2.3724904782364917, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, -1.2722157358170676, -0.250470603921817, -0.5143892000079515, -0.2957296600062443 }, { 1.029712237713905, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, -0.5441892438806311, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, -1.1448057498013176, 0.8389893256787506, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, -0.5143892000079515, -0.2957296600062443 }, { -0.8403433999584713, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, -1.970316152682646, 0.16289645171177966, 2.278612115023465, 1.494682719884397, 0.05347685426934428, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, 0.2843298100961381, -0.8771690980243297 }, { 1.029712237713905, 0.14012794477794924, 1.7658740733635434, 1.185916567160356, -0.5441892438806311, 0.16289645171177966, 2.278612115023465, 1.494682719884397, 0.05347685426934428, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, 1.0087021598534562, 0.2843298100961381, -0.8771690980243297 }, { -0.21699152073434588, 0.9532703900465632, -1.376724168039617, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, 2.278612115023465, 1.494682719884397, 1.4658499494714023, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, -0.94101658068771, 0.0, 0.14203663498716684, -0.2561317410190009, -1.2722157358170676, 1.4284264144452141, 0.2843298100961381, 0.28570977801184116 }, { 1.029712237713905, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, -1.970316152682646, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.2289977647710673, -1.1448057498013176, 0.8389893256787506, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, -0.5143892000079515, -0.2957296600062443 }, { -0.8403433999584713, 0.9532703900465632, 1.3730492931881484, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, -0.5114723838114789, -1.1448057498013176, -1.0459524004866725, 0.6837776537937139, -2.3724904782364917, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, -0.7171415877086972, -1.0899191131053325, 1.0830488202002277, -0.8771690980243297 }, { -0.21699152073434588, 0.14012794477794924, -0.19824982751343181, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, -1.1448057498013176, -0.10348153740396093, 2.2929925029884224, 2.065822227902576, -0.9551523664354273, -0.94101658068771, 0.0, 0.14203663498716684, 3.4332552519568216, 0.9480808566164142, 1.4284264144452141, 1.0830488202002277, -0.8771690980243297 }, { 1.029712237713905, 0.9532703900465632, 1.3730492931881484, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, -0.5114723838114789, -1.1448057498013176, -1.0459524004866725, 0.6837776537937139, -2.3724904782364917, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, -0.7171415877086972, -1.0899191131053325, 1.0830488202002277, -0.8771690980243297 }, { 1.029712237713905, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, -0.5441892438806311, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.05347685426934428, -1.1448057498013176, 0.8389893256787506, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, -0.5143892000079515, 0.8671492160299267 }, { -0.8403433999584713, 0.9532703900465632, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.1833753304309906, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.8771690980243297 }, { -0.8403433999584713, 0.9532703900465632, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.6184260923501674, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.8771690980243297 }, { -2.087047158406722, 0.14012794477794924, 1.3730492931881484, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, 2.278612115023465, -0.6690383093994308, 1.4658499494714023, -1.1448057498013176, -1.0459524004866725, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, 3.4332552519568216, 0.9480808566164142, 1.4284264144452141, -1.313108210112041, -0.2957296600062443 }, { 1.029712237713905, -1.4861569457592787, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.05347685426934428, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, -1.4653525684453887, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, 0.2843298100961381, -0.8771690980243297 }, { 1.029712237713905, 0.14012794477794924, 1.7658740733635434, 1.185916567160356, -0.5441892438806311, 0.16289645171177966, 2.278612115023465, 1.494682719884397, 1.4658499494714023, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, 0.2843298100961381, -0.8771690980243297 }, { 1.029712237713905, 0.9532703900465632, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.6184260923501674, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.8771690980243297 }, { 1.029712237713905, 0.9532703900465632, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.05347685426934428, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.8771690980243297 }, { -0.8403433999584713, -1.4861569457592787, -0.19824982751343181, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, -1.4810169520224028, -1.4653525684453887, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.8771690980243297 }, { 1.029712237713905, 0.14012794477794924, -0.5910746076888268, 1.185916567160356, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -1.0899191131053325, 0.2843298100961381, -0.2957296600062443 }, { 1.029712237713905, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.6184260923501674, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -1.0899191131053325, -0.5143892000079515, -0.2957296600062443 }, { 1.029712237713905, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.5114723838114789, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -1.0899191131053325, 0.2843298100961381, -0.2957296600062443 }, { -0.8403433999584713, 0.9532703900465632, -0.5910746076888268, -0.8432296400028142, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.6184260923501674, -1.1448057498013176, -0.10348153740396093, -0.9254371954009945, -0.893052909523469, 0.09657680473852359, -3.0383605317184252, 0.0, 0.14203663498716684, -0.2561317410190009, -0.16206743960032674, -1.0899191131053325, 0.2843298100961381, 1.448588654048012 }, { 1.029712237713905, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, 0.881937664921384, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, 1.4658499494714023, -1.1448057498013176, 1.7814601887614623, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, 0.2843298100961381, 2.0300280920660976 }, { 1.029712237713905, -1.4861569457592787, 1.3730492931881484, -0.8432296400028142, -1.4949405164153078, 0.16289645171177966, 2.278612115023465, 1.494682719884397, -0.7939470028518905, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, -0.5143892000079515, -0.8771690980243297 }, { 1.029712237713905, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, 0.881937664921384, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, 0.05347685426934428, -1.1448057498013176, 1.7814601887614623, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, -0.2957296600062443 }, { -0.8403433999584713, 0.14012794477794924, -0.5910746076888268, 1.185916567160356, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.5114723838114789, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -1.0899191131053325, 0.2843298100961381, -0.2957296600062443 }, { -0.8403433999584713, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -1.0899191131053325, 0.2843298100961381, -0.2957296600062443 }, { 1.029712237713905, 0.14012794477794924, 1.3730492931881484, -0.8432296400028142, -1.4949405164153078, 0.16289645171177966, 2.278612115023465, 1.494682719884397, -0.7939470028518905, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, -0.5143892000079515, -0.8771690980243297 }, { 1.029712237713905, 0.14012794477794924, -0.5910746076888268, 1.185916567160356, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.5114723838114789, 0.8735106372181927, -0.10348153740396093, -2.534652044595703, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -1.0899191131053325, -0.5143892000079515, -0.2957296600062443 }, { -2.087047158406722, 0.9532703900465632, 1.3730492931881484, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, 2.278612115023465, 1.494682719884397, 1.4658499494714023, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, 1.4284264144452141, -2.1118272202161306, 0.28570977801184116 }, { -0.21699152073434588, 0.9532703900465632, 1.7658740733635434, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, 2.278612115023465, 1.494682719884397, 1.7483245685118138, -1.1448057498013176, 0.8389893256787506, 2.2929925029884224, 2.065822227902576, 1.1483059759124745, 1.156327370343005, 0.0, 4.263128144400531, -0.2561317410190009, -1.2722157358170676, 1.4284264144452141, -2.1118272202161306, 0.28570977801184116 }, { -0.8403433999584713, 0.9532703900465632, -0.19824982751343181, -0.8432296400028142, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, -1.3588962409327137, 0.8735106372181927, -1.0459524004866725, -0.9254371954009945, -0.893052909523469, 0.09657680473852359, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, -1.2722157358170676, 1.4284264144452141, 0.2843298100961381, 1.448588654048012 }, { -0.8403433999584713, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.6184260923501674, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -1.0899191131053325, -0.5143892000079515, -0.2957296600062443 }, { 1.029712237713905, -1.4861569457592787, 0.1945749526619632, -0.8432296400028142, -1.4949405164153078, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, 1.1833753304309906, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, 0.2843298100961381, -0.8771690980243297 }, { -0.8403433999584713, 0.14012794477794924, 0.1945749526619632, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.7939470028518905, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, 3.4332552519568216, 0.9480808566164142, 0.5889779052616985, 0.2843298100961381, 0.8671492160299267 }, { 1.029712237713905, 0.14012794477794924, -0.19824982751343181, 1.185916567160356, 0.881937664921384, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, -0.2289977647710673, -1.1448057498013176, 1.7814601887614623, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.6701948585135749, 0.2843298100961381, 2.0300280920660976 }, { -0.8403433999584713, 0.9532703900465632, 1.7658740733635434, -0.8432296400028142, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.7939470028518905, -1.1448057498013176, -0.10348153740396093, -0.9254371954009945, -0.893052909523469, 0.09657680473852359, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, -0.16206743960032674, -1.0899191131053325, 0.2843298100961381, -0.8771690980243297 }, { -0.8403433999584713, 0.14012794477794924, -0.19824982751343181, -0.8432296400028142, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, -1.3588962409327137, 0.8735106372181927, -1.0459524004866725, -0.9254371954009945, -0.893052909523469, 0.09657680473852359, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, -1.2722157358170676, 1.4284264144452141, 0.2843298100961381, 1.448588654048012 }, { -0.8403433999584713, 0.9532703900465632, -0.9838993878642219, -0.8432296400028142, 1.3573133011887224, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, -1.3588962409327137, 0.8735106372181927, -1.0459524004866725, -0.9254371954009945, -0.893052909523469, 0.09657680473852359, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, -1.2722157358170676, 1.4284264144452141, 0.2843298100961381, -0.8771690980243297 }, { 1.029712237713905, 0.9532703900465632, -0.5910746076888268, -0.8432296400028142, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.7939470028518905, -1.1448057498013176, -0.10348153740396093, -0.9254371954009945, -0.893052909523469, 0.09657680473852359, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, -0.16206743960032674, -1.0899191131053325, 0.2843298100961381, 1.448588654048012 }, { 1.029712237713905, 0.14012794477794924, 0.1945749526619632, -0.8432296400028142, -1.4949405164153078, 0.16289645171177966, 2.278612115023465, 1.494682719884397, 1.1833753304309906, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, 0.2843298100961381, -0.8771690980243297 }, { 1.029712237713905, 0.14012794477794924, -0.5910746076888268, 1.185916567160356, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.6184260923501674, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -1.0899191131053325, 0.2843298100961381, -0.2957296600062443 }, { 1.029712237713905, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.5114723838114789, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, -2.3724904782364917, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -1.0899191131053325, -0.5143892000079515, -0.2957296600062443 }, { 1.029712237713905, 0.14012794477794924, 1.3730492931881484, -0.8432296400028142, -1.4949405164153078, 0.16289645171177966, 2.278612115023465, 1.494682719884397, 1.1833753304309906, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, 0.2843298100961381, -0.8771690980243297 }, { 1.029712237713905, 0.9532703900465632, -0.9838993878642219, -0.8432296400028142, 1.3573133011887224, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, -1.3588962409327137, 0.8735106372181927, -1.0459524004866725, -0.9254371954009945, -0.893052909523469, 0.09657680473852359, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, -1.2722157358170676, 1.4284264144452141, 0.2843298100961381, -0.8771690980243297 }, { -0.8403433999584713, 0.9532703900465632, 0.1945749526619632, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, 3.4332552519568216, 0.9480808566164142, 0.5889779052616985, 0.2843298100961381, 0.8671492160299267 }, { -0.8403433999584713, 0.14012794477794924, 0.1945749526619632, 1.185916567160356, 0.40656202865404567, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, 3.4332552519568216, 0.9480808566164142, 0.5889779052616985, 0.2843298100961381, 0.8671492160299267 }, { -0.8403433999584713, -1.4861569457592787, -0.5910746076888268, -0.8432296400028142, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.7939470028518905, -1.1448057498013176, -0.10348153740396093, -0.9254371954009945, -0.893052909523469, -0.9551523664354273, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, -0.16206743960032674, -1.0899191131053325, 0.2843298100961381, 1.448588654048012 }, { 1.029712237713905, -1.4861569457592787, 1.3730492931881484, -0.8432296400028142, -1.4949405164153078, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, -0.7939470028518905, -1.1448057498013176, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, -0.5143892000079515, -0.8771690980243297 }, { -0.8403433999584713, -1.4861569457592787, -0.5910746076888268, -0.8432296400028142, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 0.6184260923501674, -1.1448057498013176, -0.10348153740396093, -0.9254371954009945, -0.893052909523469, 0.09657680473852359, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, -0.16206743960032674, -1.0899191131053325, 0.2843298100961381, 1.448588654048012 }, { 1.029712237713905, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.5114723838114789, 0.8735106372181927, -0.10348153740396093, -2.534652044595703, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -1.0899191131053325, -0.5143892000079515, -0.2957296600062443 }, { -0.8403433999584713, 0.14012794477794924, -0.5910746076888268, 1.185916567160356, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, 1.4658499494714023, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -1.0899191131053325, 0.2843298100961381, -0.2957296600062443 }, { -2.087047158406722, 0.9532703900465632, 1.7658740733635434, -0.8432296400028142, 0.40656202865404567, 0.16289645171177966, 2.278612115023465, 1.494682719884397, 1.7483245685118138, -1.1448057498013176, 0.8389893256787506, 2.2929925029884224, 2.065822227902576, 1.1483059759124745, 1.156327370343005, 0.0, 4.263128144400531, -0.2561317410190009, -1.2722157358170676, 1.4284264144452141, -2.1118272202161306, 0.28570977801184116 }, { 1.029712237713905, 0.14012794477794924, 1.3730492931881484, 1.185916567160356, 0.881937664921384, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, -0.2289977647710673, -1.1448057498013176, 1.7814601887614623, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, 0.2843298100961381, 2.0300280920660976 }, { -0.8403433999584713, 0.9532703900465632, -0.5910746076888268, -0.8432296400028142, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.7939470028518905, -1.1448057498013176, -0.10348153740396093, -0.9254371954009945, -0.893052909523469, 0.09657680473852359, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, -0.16206743960032674, -1.0899191131053325, 0.2843298100961381, 1.448588654048012 }, { 1.029712237713905, 0.14012794477794924, -0.5910746076888268, 1.185916567160356, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.5114723838114789, 0.8735106372181927, -0.10348153740396093, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -1.0899191131053325, -0.5143892000079515, -0.2957296600062443 }, { 1.029712237713905, -1.4861569457592787, -0.5910746076888268, -0.8432296400028142, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.5114723838114789, -1.1448057498013176, -0.10348153740396093, -0.9254371954009945, -0.893052909523469, 0.09657680473852359, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, -0.16206743960032674, -1.0899191131053325, 0.2843298100961381, 1.448588654048012 }, { -0.8403433999584713, -1.4861569457592787, 1.7658740733635434, -0.8432296400028142, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.7939470028518905, -1.1448057498013176, -0.10348153740396093, -0.9254371954009945, -0.893052909523469, -0.9551523664354273, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, -0.16206743960032674, -1.0899191131053325, 0.2843298100961381, 1.448588654048012 }, { 1.029712237713905, 0.14012794477794924, -0.19824982751343181, 1.185916567160356, 0.881937664921384, 0.16289645171177966, -0.4388636369510842, 1.494682719884397, 1.4658499494714023, -1.1448057498013176, 1.7814601887614623, 0.6837776537937139, 0.5863846591895536, 0.622441390325499, 0.6319913825853262, 0.0, 0.14203663498716684, -0.2561317410190009, 0.9480808566164142, -0.250470603921817, 0.2843298100961381, 2.0300280920660976 }, { -0.8403433999584713, 0.9532703900465632, 1.7658740733635434, -0.8432296400028142, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.7939470028518905, -1.1448057498013176, -0.10348153740396093, -0.9254371954009945, -0.893052909523469, -0.9551523664354273, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, -0.16206743960032674, -1.0899191131053325, 0.2843298100961381, -0.8771690980243297 }, { 1.029712237713905, -1.4861569457592787, 1.7658740733635434, -0.8432296400028142, -1.0195648801479693, 0.16289645171177966, -0.4388636369510842, -0.6690383093994308, -0.7939470028518905, -1.1448057498013176, -0.10348153740396093, -0.9254371954009945, -0.893052909523469, -3.058610708783329, 0.10765539482764744, 0.0, 0.14203663498716684, -0.2561317410190009, -0.16206743960032674, -1.0899191131053325, 0.2843298100961381, 1.448588654048012 } };
double[,] coefficientsMashroomD = { { -1.3798597708867266, -4.883607955379984, -5.548165790648151, -0.061757313767708176, -1.0865719010661703, -0.01426424141400819, -0.6163936088063474, -4.260153928818506, -1.2569041780574652, -2.1466027594616666, -2.2687652579772193, -6.118469300074238, -0.4924357275715736, -19.86696648294654, -0.34593690639143176, -1.0568472890871559, -0.14068891807316228, -0.9774166529477862, -7.130533106146039, -1.4840676028542605, -1.7639384438998986, -1.2819441836162062, -5.406881303996927, -0.9428957399138065, -2.322454934072298, -1.256062020922827, -2.860208657375192, -8.666078918366267, -0.4259696231740045, -0.473397647886441, -0.573640257921649, -0.16074059078640462, -6.261905222713596, -0.2049332572451397, -1.424903904529542, -0.12479587490241022, -8.958852855238515, -0.12189667565667259, -0.24280512531928305, -8.553587621769204, -3.292456463251167, -0.1290079928645057, -1.5624125899701915, -1.4162751660162036, -9.559136292753655, -5.651622437034512, -9.154899613658696, -0.6599745540188126, -0.07990402107507785, -5.260578627046839, -2.724836345832232, -13.289853100018485, -0.9342464605060691, 9.909437436603506, 11.101371474384855, 14.60593660919845, 0.5929858932655261, 2.397355266695858, 1.4186969140985455, 2.2551230051915527, 13.044814491403326, 2.537949671413006, 0.09808894794076044, 0.451166340737226, 2.9190463941871516, 0.2879197684262461, 2.5310436365072775, 6.335447288910088, 0.46137414846171526, 2.5343253229422906, 1.192585411419088, 1.6910118124713702, 7.046825275416602, 1.88315353921033, 0.19950359763170292, 4.279542727719369, 0.15618881292275336, 4.969094934430523, 6.899153231002094, 0.2435028745144476, 22.331397870782144, 8.619190232607641, 1.9735084693990874, 2.0925254204676658, 1.7430664187590996, 3.363379375281906, 0.7355121005045911, 0.19870584895371457, 3.306725180785799, 3.834438723728394, 1.0693253739909567, 3.596272507907274, 4.818436678576192, 2.1717114699735163, 4.046615221790625, 0.936049495114594 } };
double[] interceptsMashroom = { -0.3209481162941494 };
int[] weightsMashroom = { 53, 43 };
double[][] vectorsMashroom = SVCUtilities.SvcUtilities.ToJaggedArray(vectorsMashroomD);
double[][] coefficientsMashroom = SVCUtilities.SvcUtilities.ToJaggedArray(coefficientsMashroomD);
double[][] features;
int numOfRows = 0;
if (args.Length >= 4)
{
numOfRows = args.Length / 4;
// Features:
features = new double[numOfRows][];
for (int i = 0; i < numOfRows; i++)
{
features[i] = new double[4];
}
for (int i = 0, l = args.Length; i < l; i++)
{
features[i / 4][i % 4] = Double.Parse(args[i]);
}
}
else
{
List <double[]> rows = new List <double[]>();
var bytesMashrooms = Properties.Resources.mashrooms;
numOfRows = 0;
features = SVCUtilities.SvcUtilities.LoadFeatures(bytesMashrooms, vectorsMashroom[0].Length, ref numOfRows);
}
Svc clf = new Svc(2, 2, vectorsMashroom, coefficientsMashroom, interceptsMashroom, weightsMashroom, "Poly", 0.045454545454545456, 0.0, 2);
using (System.IO.StreamWriter file =
new System.IO.StreamWriter(
$@"{OutputDir}MashroomPlain_{DateTime.Now.Day}_{DateTime.Now.ToShortTimeString().ToString().Replace(":", "_")}.txt")
)
{
Stopwatch totalTime = new Stopwatch();
totalTime.Start();
for (int i = 0; i < numOfRows; i++)
{
double finalResult = -10000;
int estimation = clf.Predict(features[i], out finalResult);
Console.WriteLine($"\n ************************************************");
Console.WriteLine($"SVC estimation{i} is : {estimation} , result : {finalResult}");
file.WriteLine($"{i} , {estimation} , {finalResult} ");
Console.WriteLine($"************************************************ \n");
}
totalTime.Stop();
file.WriteLine($" Total time for {numOfRows} samples : {totalTime.ElapsedMilliseconds} ms ");
file.WriteLine($" Avg time : {totalTime.ElapsedMilliseconds * 1000 / numOfRows} microSec ");
}
}
public void TestSingleSample_1D()
{
var clf = new Svc<int>();
clf.Fit(X, Y);
var p = clf.Predict(X.Row(0).ToRowMatrix());
Assert.AreEqual(Y[0], p[0]);
//todo:
//clf = svm.LinearSVC(random_state=0).fit(X, Y)
//clf.predict(X[0])
}
public void TestPrecomputed()
{
var clf = new Svc<int>(kernel: Kernel.Precomputed);
// Gram matrix for train data (square matrix)
// (we use just a linear kernel)
var k = X*(X.Transpose());
clf.Fit(k, Y);
// Gram matrix for test data (rectangular matrix)
var kt = T*X.Transpose();
var pred = clf.Predict(kt);
try
{
clf.Predict(kt.Transpose());
Assert.Fail();
}
catch (ArgumentException)
{
}
Assert.IsTrue(clf.DualCoef.AlmostEquals(DenseMatrix.OfArray(new[,] {{0.25, -.25}})));
Assert.IsTrue(clf.Support.SequenceEqual(new[] {1, 3}));
Assert.IsTrue(clf.Intercept.SequenceEqual(new[] {0.0}));
Assert.IsTrue(pred.SequenceEqual(true_result));
// Gram matrix for test data but compute KT[i,j]
// for support vectors j only.
kt = kt.CreateMatrix(kt.RowCount, kt.ColumnCount);
for (int i = 0; i < T.RowCount; i++)
{
foreach (var j in clf.Support)
{
kt[i, j] = T.Row(i)*X.Row(j);
}
}
pred = clf.Predict(kt);
Assert.IsTrue(pred.SequenceEqual(true_result));
// same as before, but using a callable function instead of the kernel
// matrix. kernel is just a linear kernel
clf = new Svc<int>(kernel: Kernel.FromFunction((x, y) => x*y.Transpose()));
clf.Fit(X, Y);
pred = clf.Predict(T);
Assert.IsTrue(clf.DualCoef.AlmostEquals(DenseMatrix.OfArray(new[,] {{0.25, -.25}})));
Assert.IsTrue(clf.Support.SequenceEqual(new[] {1, 3}));
Assert.IsTrue(clf.Intercept.SequenceEqual(new[] {0.0}));
Assert.IsTrue(pred.SequenceEqual(true_result));
// test a precomputed kernel with the iris dataset
// and check parameters against a linear SVC
clf = new Svc<int>(kernel: Kernel.Precomputed);
var clf2 = new Svc<int>(kernel: Kernel.Linear);
k = iris.Data*iris.Data.Transpose();
clf.Fit(k, iris.Target);
clf2.Fit(iris.Data, iris.Target);
pred = clf.Predict(k);
Assert.IsTrue(clf.Support.SequenceEqual(clf2.Support));
Assert.IsTrue(clf.DualCoef.AlmostEquals(clf2.DualCoef));
Assert.IsTrue(clf.Intercept.AlmostEquals(clf2.Intercept));
var matchingN = pred.Zip(iris.Target, Tuple.Create).Where(t => t.Item1 == t.Item2).Count();
Assert.IsTrue(1.0*matchingN/pred.Length > 0.99);
// Gram matrix for test data but compute KT[i,j]
// for support vectors j only.
k = k.CreateMatrix(k.RowCount, k.ColumnCount);
for (int i = 0; i < iris.Data.RowCount; i++)
{
foreach (var j in clf.Support)
k[i, j] = iris.Data.Row(i)*iris.Data.Row(j);
}
pred = clf.Predict(k);
matchingN = pred.Zip(iris.Target, Tuple.Create).Where(t => t.Item1 == t.Item2).Count();
Assert.IsTrue(1.0*matchingN/pred.Length > 0.99);
clf = new Svc<int>(kernel: Kernel.FromFunction((x, y) => x*y.Transpose()));
clf.Fit(iris.Data, iris.Target);
matchingN = pred.Zip(iris.Target, Tuple.Create).Where(t => t.Item1 == t.Item2).Count();
Assert.IsTrue(1.0*matchingN/pred.Length > 0.99);
}
public void TestLibsvmIris()
{
// shuffle the dataset so that labels are not ordered
foreach (var k in new[] {Kernel.Linear, Kernel.Rbf})
{
var clf = new Svc<int>(kernel: k);
clf.Fit(iris.Data, iris.Target);
var pred = clf.Predict(iris.Data);
var matchingN = pred.Zip(iris.Target, Tuple.Create).Where(t => t.Item1 == t.Item2).Count();
Assert.IsTrue(1.0*matchingN/pred.Length > 0.9);
Assert.IsTrue(clf.Classes.SequenceEqual(clf.Classes.OrderBy(v => v)));
}
}
public void TestPrecomputed()
{
var clf = new Svc <int>(kernel: Kernel.Precomputed);
// Gram matrix for train data (square matrix)
// (we use just a linear kernel)
var k = X * (X.Transpose());
clf.Fit(k, Y);
// Gram matrix for test data (rectangular matrix)
var kt = T * X.Transpose();
var pred = clf.Predict(kt);
try
{
clf.Predict(kt.Transpose());
Assert.Fail();
}
catch (ArgumentException)
{
}
Assert.IsTrue(clf.DualCoef.AlmostEquals(DenseMatrix.OfArray(new[, ] {
{ 0.25, -.25 }
})));
Assert.IsTrue(clf.Support.SequenceEqual(new[] { 1, 3 }));
Assert.IsTrue(clf.Intercept.SequenceEqual(new[] { 0.0 }));
Assert.IsTrue(pred.SequenceEqual(true_result));
// Gram matrix for test data but compute KT[i,j]
// for support vectors j only.
kt = kt.CreateMatrix(kt.RowCount, kt.ColumnCount);
for (int i = 0; i < T.RowCount; i++)
{
foreach (var j in clf.Support)
{
kt[i, j] = T.Row(i) * X.Row(j);
}
}
pred = clf.Predict(kt);
Assert.IsTrue(pred.SequenceEqual(true_result));
// same as before, but using a callable function instead of the kernel
// matrix. kernel is just a linear kernel
clf = new Svc <int>(kernel: Kernel.FromFunction((x, y) => x * y.Transpose()));
clf.Fit(X, Y);
pred = clf.Predict(T);
Assert.IsTrue(clf.DualCoef.AlmostEquals(DenseMatrix.OfArray(new[, ] {
{ 0.25, -.25 }
})));
Assert.IsTrue(clf.Support.SequenceEqual(new[] { 1, 3 }));
Assert.IsTrue(clf.Intercept.SequenceEqual(new[] { 0.0 }));
Assert.IsTrue(pred.SequenceEqual(true_result));
// test a precomputed kernel with the iris dataset
// and check parameters against a linear SVC
clf = new Svc <int>(kernel: Kernel.Precomputed);
var clf2 = new Svc <int>(kernel: Kernel.Linear);
k = iris.Data * iris.Data.Transpose();
clf.Fit(k, iris.Target);
clf2.Fit(iris.Data, iris.Target);
pred = clf.Predict(k);
Assert.IsTrue(clf.Support.SequenceEqual(clf2.Support));
Assert.IsTrue(clf.DualCoef.AlmostEquals(clf2.DualCoef));
Assert.IsTrue(clf.Intercept.AlmostEquals(clf2.Intercept));
var matchingN = pred.Zip(iris.Target, Tuple.Create).Where(t => t.Item1 == t.Item2).Count();
Assert.IsTrue(1.0 * matchingN / pred.Length > 0.99);
// Gram matrix for test data but compute KT[i,j]
// for support vectors j only.
k = k.CreateMatrix(k.RowCount, k.ColumnCount);
for (int i = 0; i < iris.Data.RowCount; i++)
{
foreach (var j in clf.Support)
{
k[i, j] = iris.Data.Row(i) * iris.Data.Row(j);
}
}
pred = clf.Predict(k);
matchingN = pred.Zip(iris.Target, Tuple.Create).Where(t => t.Item1 == t.Item2).Count();
Assert.IsTrue(1.0 * matchingN / pred.Length > 0.99);
clf = new Svc <int>(kernel: Kernel.FromFunction((x, y) => x * y.Transpose()));
clf.Fit(iris.Data, iris.Target);
matchingN = pred.Zip(iris.Target, Tuple.Create).Where(t => t.Item1 == t.Item2).Count();
Assert.IsTrue(1.0 * matchingN / pred.Length > 0.99);
}