C# IContinuousDistribution类代码示例

        public static List<PatternClass> Create2DimTeachingVectors(double p1, double p2, IContinuousDistribution generatorFirstFeature1, IContinuousDistribution generatorFirstFeature2, IContinuousDistribution generatorSecondFeature1, IContinuousDistribution generatorSecondFeature2, int nrOfTeachingVectors)
        {
            List<PatternClass> createdTeachingVectors = new List<PatternClass>();
             if (probabilityGenerator == null)
             {
                 probabilityGenerator = new ContinuousUniform(0, 1);
                 probabilityGenerator.RandomSource = new Random(DateTime.Now.Millisecond + 10);
             }

             for (int i = 0; i < nrOfTeachingVectors; i++)
             {
                 double value = 0;
                 double value1 = 0;
                 int classNumber = CreateClass(p1, p2);

                 if (classNumber == 1)
                 {
                     value = generatorFirstFeature1.Sample();
                     value1 = generatorSecondFeature1.Sample();
                 }
                 else if (classNumber == 2)
                 {
                     value = generatorFirstFeature2.Sample();
                     value1 = generatorSecondFeature2.Sample();
                 }

                 createdTeachingVectors.Add(new PatternClass(new FeatureVector(value,value1), classNumber));
             }

             return createdTeachingVectors;
        }

        public static List<PatternClass> Create2dimSampleObject(IContinuousDistribution generator, IContinuousDistribution generator1, int count, int classNumber)
        {
            List<PatternClass> sampleObjects = new List<PatternClass>();

                         probabilityGenerator = new ContinuousUniform(0, 1);
             probabilityGenerator.RandomSource = new Random(DateTime.Now.Millisecond + 10);

            for (int i = 0; i < count; i++)
            {
                sampleObjects.Add(new PatternClass(new FeatureVector(generator.Sample(), generator1.Sample()), classNumber));
            }
            return sampleObjects;
        }

        public static IEnumerable<Tuple<PointF, PointF>> SyntheticFeatureAlignmentData(
            int numGoodAlignments,
            int numBadAlignments,
            IContinuousDistribution featureNoise,
            IContinuousDistribution pointDistribution)
        {
            var rigidTrans = DenseMatrix.CreateRandom(3, 2, pointDistribution);

            var sourcePoints = randSourceMatrix(numGoodAlignments, pointDistribution);
            var targetPoints = NoisyTransform(sourcePoints, rigidTrans, featureNoise);

            return alignedPointsFromRows(sourcePoints.RowEnumerator(), targetPoints.RowEnumerator())
                     .Concat(alignedPointsFromRows(randSourceMatrix(numBadAlignments, pointDistribution).RowEnumerator(), DenseMatrix.CreateRandom(numBadAlignments, 2, pointDistribution).RowEnumerator()));
        }

        public static double CalculateBayesRiskUniformDistribution(IContinuousDistribution generator1, IContinuousDistribution generator2, double P1, double P2)
        {
            double dens1 = generator1.Density(generator1.Mean);
            double dens2 = generator2.Density(generator2.Mean);

            if (dens1 > dens2)
            {
                double range = generator1.Maximum - generator2.Minimum;
                if (range > 0.0)
                {
                    return range * dens2  * 100;
                }

            }
            else if (dens2 > dens1)
            {
                double range = generator2.Maximum - generator1.Minimum;
                if (range > 0.0)
                {
                    return range * dens1 * 100;
                }
            }
            else
            {
                double range1 = generator2.Maximum - generator1.Minimum;
                double range2 = generator1.Maximum - generator2.Minimum;

                if (range1 > 0.0)
                {
                    return range1 * dens1*100;
                }

                if (range2 > 0.0)
                {
                    return range2 * dens1 * 100;
                }
            }

            return 0.0;
        }

 /// <summary>
 /// Create a new dense matrix with values sampled from the provided random distribution.
 /// </summary>
 public static DenseMatrix CreateRandom(int rows, int columns, IContinuousDistribution distribution)
 {
     var storage = new DenseColumnMajorMatrixStorage<Complex>(rows, columns);
     for (var i = 0; i < storage.Data.Length; i++)
     {
         storage.Data[i] = new Complex(distribution.Sample(), distribution.Sample());
     }
     return new DenseMatrix(storage);
 }

        /// <summary>
        /// Generates a vector with random elements
        /// </summary>
        /// <param name="length">Number of elements in the vector.</param>
        /// <param name="randomDistribution">Continuous Random Distribution or Source</param>
        /// <returns>
        /// A vector with n-random elements distributed according
        /// to the specified random distribution.
        /// </returns>
        /// <exception cref="ArgumentNullException">If the length vector is non poisitive<see langword="null" />.</exception> 
        public override Vector Random(int length, IContinuousDistribution randomDistribution)
        {
            if (length < 0)
            {
                throw new ArgumentException(Resources.ArgumentMustBePositive, "length");
            }

            var v = (SparseVector)this.CreateVector(length);
            for (var index = 0; index < v.Count; index++)
            {
                v[index] = randomDistribution.Sample();
            }

            return v;
        }

 /// <summary>
 /// Create a new diagonal matrix with diagonal values sampled from the provided random distribution.
 /// </summary>
 public static DiagonalMatrix CreateRandom(int rows, int columns, IContinuousDistribution distribution)
 {
     return new DiagonalMatrix(new DiagonalMatrixStorage<double>(rows, columns, Generate.Random(Math.Min(rows, columns), distribution)));
 }

 /// <summary>
 /// Create a sample source from a continuous generator.
 /// </summary>
 public GeneratorSource(IContinuousDistribution distribution)
 {
     _distribution = distribution;
 }

        private Series CreateDistributionSerie(IContinuousDistribution distributionn)
        {
            Series createdSerie = new Series();
            createdSerie.ChartType = SeriesChartType.SplineArea;

            for (double i = distributionn.Mean - (8 * distributionn.StdDev); i < distributionn.Mean + (8 * distributionn.StdDev); i += 0.5)
            {
                if (distributionn is ContinuousUniform)
                {
                    createdSerie.Points.Add(new DataPoint((int)i, distributionn.Density(i)));
                }
                else
                {
                    createdSerie.Points.Add(new DataPoint(i, distributionn.Density(i)));

                }
            }

            return createdSerie;
        }

        /// <summary>
        /// Vapnik Chervonenkis test.
        /// </summary>
        /// <param name="epsilon">The error we are willing to tolerate.</param>
        /// <param name="delta">The error probability we are willing to tolerate.</param>
        /// <param name="s">The samples to use for testing.</param>
        /// <param name="dist">The distribution we are testing.</param>
        public static void ContinuousVapnikChervonenkisTest(double epsilon, double delta, double[] s, IContinuousDistribution dist)
        {
            // Using VC-dimension, we can bound the probability of making an error when estimating empirical probability
            // distributions. We are using Theorem 2.41 in "All Of Nonparametric Statistics".
            // http://books.google.com/books?id=MRFlzQfRg7UC&lpg=PP1&dq=all%20of%20nonparametric%20statistics&pg=PA22#v=onepage&q=%22shatter%20coe%EF%AC%83cients%20do%20not%22&f=false .</para>
            // For intervals on the real line the VC-dimension is 2.
            Assert.Greater(s.Length, Math.Ceiling(32.0 * Math.Log(16.0 / delta) / epsilon / epsilon));

            var histogram = new Histogram(s, NumberOfHistogramBuckets);
            for (var i = 0; i < NumberOfHistogramBuckets; i++)
            {
                var p = dist.CumulativeDistribution(histogram[i].UpperBound) - dist.CumulativeDistribution(histogram[i].LowerBound);
                var pe = histogram[i].Count/(double)s.Length;
                Assert.Less(Math.Abs(p - pe), epsilon, dist.ToString());
            }
        }

 /// <summary>
 /// Create a new diagonal matrix with diagonal values sampled from the provided random distribution.
 /// </summary>
 public static DiagonalMatrix CreateRandom(int rows, int columns, IContinuousDistribution distribution)
 {
     return new DiagonalMatrix(DiagonalMatrixStorage<Complex>.OfInit(rows, columns,
         i => new Complex(distribution.Sample(), distribution.Sample())));
 }

 /// <summary>
 /// Create a new dense vector with values sampled from the provided random distribution.
 /// </summary>
 public static DenseVector CreateRandom(int length, IContinuousDistribution distribution)
 {
     return new DenseVector(DenseVectorStorage<double>.OfInit(length,
         i => distribution.Sample()));
 }

 private static double CalculateIntegral1(IContinuousDistribution dist1, IContinuousDistribution dist2, double commonPoint, double p1, double p2)
 {
     double sum = 0.0;
     double sum1 = 0.0;
     for (double i = dist1.Mean - (8 * dist1.StdDev); i < dist1.Mean + (8 * dist1.StdDev); i += 0.0001)
     {
         double d1 = p1*dist1.Density(i);
         double d2 = p2*dist2.Density(i);
             if (d1 <= d2)
             {
                 sum += 0.0001 * d1;
             }
             else
             {
                 sum += 0.0001 * d2;
             }
     }
     return sum;
 }

        private static double CalculateIntegralUniform(IContinuousDistribution dist1, IContinuousDistribution dist2,double p1, double p2)
        {
            bool found = false;
            double zakresDist1 = dist1.Maximum - dist1.Minimum;
            double zakresDist2 = dist2.Maximum - dist2.Minimum;

            if (dist2.Maximum <= dist1.Maximum && dist1.Maximum < dist2.Maximum )
            {

            }

            double sum = 0.0;
            if (p1 > p2)
                return p2 * sum;
            else
                return p1 * sum;
        }

        /// <summary>
        /// Find the common point beetwen 2 distributions
        /// </summary>
        /// <param name="dist1">First Distribution</param>
        /// <param name="dist2">Second Distribution</param>
        /// <param name="commonPoint"></param>
        /// <returns></returns>
        public static bool FindCommonPointNormalDistribution(IContinuousDistribution dist1,IContinuousDistribution dist2,ref double commonPoint)
        {
            bool found = false;
            double step = 0.0;
            if (dist2.Mean > dist1.Mean)
                step = 0.001;
            else
                step = -0.001;

                for (double i = dist1.Mean; i < dist2.Mean; i += 0.001)
                {
                    if (Math.Abs(dist1.Density(i) - dist2.Density(i)) < 0.00001)
                    {
                        commonPoint = i;
                        found = true;
                        break;
                    }
                }

            return found;
        }