All Classes

Class Description
Accuracy
Return the number of correct predictions in the given data set.
AccuracyBinary
Return the number of correct predictions in the given data set.
ActivationFunction
This enum represents the available activation functions f: Rn→ Rn for use with neural networks.
AdvancedLinearAlgebra
This computation directory contains variuous linear algebra functions.
AffineMap
Apply an affine map to a vector
BackPropagationOutput
This class represents the output of back propagation on a single layer.
BackSubstitution
Use backward substitution to compute a vector x such that ax = b, where a is upper triangular square matrix.
ChiSquareTest
Compute the Χ2-test for goodness of fit of the given observatinos.
Convolution
Compute the discrete convolution of two vectors
CoxGradientContinuous  
CoxGradientDiscrete  
CoxRegressionContinuous
Estimate the coefficients of a Cox model on the given data using gradient descent.
CoxRegressionDiscrete
Estimate the coefficients of a Cox model on the given data using gradient descent.
DefaultFilteredStatistics  
DefaultLinearAlgebra  
DefaultMachineLearning  
DefaultSampler  
DefaultStatistics  
DivideBySInt  
FilteredStatistics
Various analysis methods for filtered data, eg a data set and a bit vector of the same length with entry i indicating whether the i'th data point should be considered.
FindTiedGroups
Find sets of mututally equal in a list of secret shared integers: If the i'th and j'th elements are the same in the result, it indicates that the i'th and j'th elements are equal in the input data.
ForwardPropagationOutput
This class represents the output of forward propagation on a single layer.
ForwardSubstitution
Use forward substitution to compute a vector x such that ax = b, where a is lower triangular square matrix.
FTest
Compute the F-test for equal mean (one-way-anova) for the given data sets.
GramSchmidt
Perform the Gram-Schmidt process on a list of linearly independent vectors.
Histogram
Compute a 1-dimensional histogram for a data set.
HistogramFiltered
Compute a 1-dimensional histogram for a data set.
InvertLowerTriangularMatrix
Invert lower triangular matrix.
InvertUpperTriangularMatrix
Invert upper triangular matrix.
KruskallWallisTest
Compute the Kruskall-Wallis test statistic on k groups, also known as one-way ANOVA on ranks.
Layer
Instances of this class represents fully connected layers in a neural network.
LeakyBreakTies
Assuming that the input data is sorted, this computation outputs the ranks of the elements using the given strategy.
LeakyFrequencyTable
Compute the frequencies of entries in the given data.
LeakyKAnonymity
Compute a k-anonymous version of a dataset.
LinearInverseProblem
Solve a linear inverse problem, eg.
LinearRegression
Fit a linear model to the given dataset and output estimates for the coefficients and some model diagnostics (see LinearRegression.LinearRegressionResult).
LinearRegression.LinearRegressionResult  
LogisticRegression
A naive implementation of logistic regression, not optimized for secure computation.
LogisticRegressionGD
A gradient descent algorithm to fit a logistic model to a dataset.
LogisticRegressionPrediction  
MachineLearning
This computation library contains various functions for machine learning.
MahalanobisDistance
Compute the Mahalanobis Distance of all samples in a data set.
MatrixUtils  
MatrixUtils.MatrixPopulator<E>  
MaxList
Given a list X = [x0, ..., xn-1] with n a power of 2, this method returns a pair [max(X), i : xi = max(X)].
MaxPair
Returns [max(x, y), 1 if y = max(x, y) and 0 otherwise]
MLP
This class represents layered neural networks.
MoorePenrosePseudoInverse
Compute the Moore-Penrose pseudo inverse of a matrix A.
MultiDimensionalArray<T>
A multi-dimensional array is a data collection where entries are indexed by a fixed length vector (the length equals the dimension of the array).
MultiDimensionalHistogram
Compute a multi-dimensiona histogram on a dataset
NoisyHistogram
Compute a differentially private histogram for a dataset.
NoisyStats
Compute differentially private estimates for the coefficients of a linear model fitted on a dataset.
NormalizeVector
Normalize a vector.
OneSampleTTest
Compute a t-test statistics on a sample for the null hypothesis the mean of the sample is equal to mu.
OneSampleTTestFiltered
Compute a t-test statistics on a sample for the null hypothesis the mean of the sample is equal to mu.
OneSampleTTestFiltered.FilteredTTestResult  
PearsonCorrelation
Compute the correlation between two samples.
Predict
Assuming that the given neural network has n dimensional output, this function applies the network to the given input and finds the index of the output i with 0 ≤ i < n containing the largest number.
Projection
Compute the projection of a vector onto another.
QRAlgorithm
Compute eigenvalues of a matrix using the iterative QR algorithm.
QRDecomposition
Compute the QR-decomposition of an mxn-matrix A with m ≥ n and full column rank.
Ranks
Output ranks with averaged ties and correction term for Kruskall-Wallis.
RealUtils  
Relu
Compute the rectified linear function f(x) = x if x > 0 and f(x) = 0 otherwise.
ReluDerivative
Compute the derivative of the Relu function, eg.
SampleBernoulliDistribution
Sample a number from a Bernoulli distribution which is 0 with probability p and 1 with probability 1-p.
SampleCategoricalDistribution
Sample an element from a categorical distribution.
SampleCovariance
Compute the unbiased covariance matrix for the given observations
SampleExponentialDistribution
This computation samples from an exponential distribution with rate 1/lambda and location 0.
SampleIrwinHallDistribution
Sample a number from an Irwin-Hall distribution which is the sum of n iid U(0,1) distributions.
SampleLaplaceDistribution
This computation samples from a Laplace distribution with scale b and location 0.
SampleMean
Compute the mean of a list of observations.
SampleMeanFiltered
Compute the sample mean of a set of samples.
SampleMedian
Compute the sample median of a sample
SampleNormalDistribution
Sample a number from an approximately standard normal distribution.
SampleQuantiles
Compute some quantiles for a sample
Sampler
This computation library contains functions which samples random values from various distributions.
SampleRademacherDistribution
This computation samples from a Rademacher distribution which can be -1 or +1 each with probability 1/2.
SampleStandardDeviation
Compute the standard deviation of a list of observations.
SampleUniformDistribution
Sample a number uniformly in the interval [0,1).
SampleVariance
Compute the sample variance for a list of observations.
SampleVarianceFiltered
Compute the sample variance of a set of samples.
Sigmoid
Compute the sigmoid (logistic) function f(x) = 1 / (1 + e-x).
SigmoidDerivative
Compute the derivative of the Sigmoid function f'(x)given the function value y = f(x)in the point.
SimpleLinearRegression
This computation returns coefficients a and b based on a simple linear regression of the observed x and y values.
SimpleLinearRegression.SimpleLinearRegressionResult  
SimpleLinearRegressionTTest
Test for the null hypothesis H0: β = β0 where β is a coefficient in a linear model.
SP
Calculate the sum of products (aka the dot product) of two samples
SPD
Compute the sum of products of deviations of two samples.
SSD
Compute the sum of squared deviations
SSDFiltered  
SSE
Compute the sum of squared estimate of errors (aka the residual sum of squares)
Statistics
This computation library contains various statistical functions.
SumFiltered  
SurvivalInfoContinuous
Represents a data point in data for survival analysis with continuous covariates.
SurvivalInfoDiscrete
Represents a data point in data for survival analysis with only discrete covariates on finite sets.
SurvivalInfoSorter<T>
Convert an instance of type T into a Pair of a DRes<SInt> and a List<DRes<SInt>> and back for use with Collections.sort(List) to sort on the time of event parameter.
SurvivalInfoSorterContinuous  
SurvivalInfoSorterDiscrete  
TransposedMatrixAction
This computation multiplies the transpose of the given matrix to a vector without explicitly representing the matrix in its transposed form.
Triple<A,​B,​C>
Instances of this class holds three values of arbitrary type.
TwoDimensionalHistogram
Compute a two-dimensional histogram for a given two dimensional data set.
TwoSampleTTest
This implements the calculation of a t-test statistics for two samples where it can be assumed that the variances are equal.
USS
Compute the uncorrected sum of squares
USSFiltered  
VectorUtils  
VectorUtils.EntrywiseBinaryOp<A,​B,​C>  
VectorUtils.EntrywiseUnaryOp<A,​C>