Package dk.alexandra.fresco.stat.regression.logistic

Class LogisticRegression

Object

LogisticRegression

All Implemented Interfaces:

Computation<ArrayList<DRes<SFixed>>,ProtocolBuilderNumeric>

public class LogisticRegression
extends Object
implements Computation<ArrayList<DRes<SFixed>>,ProtocolBuilderNumeric>

A naive implementation of logistic regression, not optimized for secure computation. Given a data set consisting of column vectors and a list of expected classifications (0 or 1), this computation performs a linear regression on the data and the expected outcome as log-odds. See also https://en.wikipedia.org/wiki/Logistic_regression.

Constructor Summary

Constructors

Constructor	Description
LogisticRegression(Matrix<DRes<SFixed>> data, ArrayList<DRes<SFixed>> expected, double[] beta, IntToDoubleFunction rate, int epochs)

Method Summary

Modifier and Type	Method	Description
DRes<ArrayList<DRes<SFixed>>>	buildComputation(ProtocolBuilderNumeric builder)

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

LogisticRegression

public LogisticRegression(Matrix<DRes<SFixed>> data, ArrayList<DRes<SFixed>> expected, double[] beta, IntToDoubleFunction rate, int epochs)

Method Details

buildComputation

public DRes<ArrayList<DRes<SFixed>>> buildComputation(ProtocolBuilderNumeric builder)

Specified by:

buildComputation in interface Computation<ArrayList<DRes<SFixed>>,ProtocolBuilderNumeric>