Merge pull request #1 from radumg/BiLT

BiLT - LinearRegression & NaiveBayes Classifier

src/DynAI.sln

@@ -0,0 +1,22 @@
+
+Microsoft Visual Studio Solution File, Format Version 12.00
+# Visual Studio 15
+VisualStudioVersion = 15.0.26430.16
+MinimumVisualStudioVersion = 10.0.40219.1
+Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "DynAI", "DynAI\MachineLearning.csproj", "{5990DC45-28B6-4E1E-B3EA-C6549EE270F7}"
+EndProject
+Global
+	GlobalSection(SolutionConfigurationPlatforms) = preSolution
+		Debug|Any CPU = Debug|Any CPU
+		Release|Any CPU = Release|Any CPU
+	EndGlobalSection
+	GlobalSection(ProjectConfigurationPlatforms) = postSolution
+		{5990DC45-28B6-4E1E-B3EA-C6549EE270F7}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
+		{5990DC45-28B6-4E1E-B3EA-C6549EE270F7}.Debug|Any CPU.Build.0 = Debug|Any CPU
+		{5990DC45-28B6-4E1E-B3EA-C6549EE270F7}.Release|Any CPU.ActiveCfg = Release|Any CPU
+		{5990DC45-28B6-4E1E-B3EA-C6549EE270F7}.Release|Any CPU.Build.0 = Release|Any CPU
+	EndGlobalSection
+	GlobalSection(SolutionProperties) = preSolution
+		HideSolutionNode = FALSE
+	EndGlobalSection
+EndGlobal

src/DynAI/MachineLearning.cs

@@ -0,0 +1,201 @@
+using Accord.MachineLearning.Bayes;
+using Accord.Math;
+using Accord.Math.Random;
+using Accord.Statistics.Filters;
+using Accord.Statistics.Models.Regression.Linear;
+using System;
+using System.Collections.Generic;
+using System.Linq;
+using System.Text;
+using System.Threading.Tasks;
+
+namespace AI.MachineLearning
+{
+    #region Linear Regression
+
+    /// <summary>
+    /// Class providing support for linear regression ML algorithms
+    /// </summary>
+    public class LinearRegression
+    {
+        #region public properties
+        // dataset
+        public double[] inputs { get; private set; }
+        public double[] outputs { get; private set; }
+        public double testValue { get; private set; }
+
+        // regression
+        public SimpleLinearRegression regression { get; private set; }
+        public OrdinaryLeastSquares ols;
+
+        // state & result
+        public bool learned { get; private set; }
+        public double result { get; private set; }
+        #endregion
+
+        /// <summary>
+        /// Constructs a new LinearRegression machine.
+        /// </summary>
+        public LinearRegression(List<double> inputList, List<double> outputList)
+        {
+            // validation
+            if (inputList == null || outputList == null) throw new ArgumentNullException("Neither the input list nor the output list can be NULL");
+
+            // initialise seed value
+            Generator.Seed = new Random().Next();
+
+            // process input and output lists into arrays
+            inputs = inputList.ToArray();
+            outputs = outputList.ToArray();
+
+            // set up linear regression using OLS
+            regression = new SimpleLinearRegression();
+            ols = new OrdinaryLeastSquares();
+
+            // nulls
+            testValue = new double();
+            result = new double();
+            this.learned = false;
+        }
+
+        /// <summary>
+        /// Use the object's inputs and outputs to learn the model of the linear regression, using OrdinaryLeastSquares
+        /// </summary>
+        public LinearRegression Learn()
+        {
+            regression = this.ols.Learn(inputs, outputs);
+            learned = true;
+
+            return this;
+        }
+
+        /// <summary>
+        /// Using the learned model, predict an output for the specified input
+        /// </summary>
+        /// <param name="test">The value to use as input for the prediction</param>
+        /// <returns>The predicted value</returns>
+        public double Predict(double test)
+        {
+            // don't predict if we haven't learned the model yet
+            if (this.learned != true) throw new Exception("Cannot predict before the machine has learned.");
+
+            // check we haven't already predicted for this input
+            if (test == this.testValue && this.learned == true) return this.result;
+
+            // predict
+            this.testValue = test;
+            this.result = this.regression.Transform(this.testValue);
+
+            return this.result;
+        }
+    }
+    #endregion
+
+    #region Classifiers
+    /// <summary>
+    /// Class providing support for Naive Bayes classification machines.
+    /// </summary>
+    public class NaiveBayes
+    {
+        #region public properties
+        // dataset
+        public string[][] dataset { get; private set; }
+        public string[] columns { get; private set; }
+        public string outputColumn { get; private set; }
+        public int[][] inputs;
+        public int[] outputs;
+
+        // classifier
+        public Accord.MachineLearning.Bayes.NaiveBayes classifier;
+        public NaiveBayesLearning learner;
+        public Codification codebook { get; private set; }
+
+        // state & result
+        public bool learned { get; private set; }
+        public string[] testValue { get; private set; }
+        public string result { get; private set; }
+        public double[] probs { get; private set; }
+        #endregion
+
+        /// <summary>
+        /// Constructs a new NaiveBayes classification machine.
+        /// </summary>
+        public NaiveBayes(string[][] data, List<string> columnList, string outputColumn)
+        {
+            // validation
+            if (data == null || columnList == null || outputColumn==null) throw new ArgumentNullException("Neither the input list nor the column list can be NULL");
+
+            // initialise seed value
+            Generator.Seed = new Random().Next();
+
+            // process input and output lists into arrays
+            this.dataset = data;
+            this.columns = columnList.ToArray();
+            this.outputColumn = outputColumn;
+
+            // Create a new codification codebook to
+            // convert strings into discrete symbols
+            this.codebook = new Codification(columns, this.dataset);
+
+            // Extract input and output pairs to train
+            int[][] symbols = this.codebook.Transform(this.dataset);
+            this.inputs = symbols.Get(null, 0, -1); // Gets all rows, from 0 to the last (but not the last)
+            this.outputs = symbols.GetColumn(-1);     // Gets only the last column
+
+            // Create a new Naive Bayes learning
+            this.learner = new NaiveBayesLearning();
+
+            // nulls
+            testValue = null;
+            result = null;
+            probs = null;
+            this.learned = false;
+        }
+
+        /// <summary>
+        /// Use the object's inputs and outputs to learn the model of the linear regression, using OrdinaryLeastSquares
+        /// </summary>
+        public NaiveBayes Learn()
+        {
+            this.classifier = this.learner.Learn(inputs, outputs);
+            this.learned = true;
+
+            return this;
+        }
+
+        /// <summary>
+        /// Using the learned model, predict an output for the specified input
+        /// </summary>
+        /// <param name="test">The value to use as input for the prediction</param>
+        /// <returns>The predicted value</returns>
+        public string Predict(string[] test)
+        {
+            // don't predict if we haven't learned the model yet
+            if (this.learned != true) throw new Exception("Cannot predict before the machine has learned.");
+
+            // check we haven't already predicted for this input
+            if (test == this.testValue && this.learned == true) return this.result;
+
+            // predict
+            // First encode the test instance
+            int[] instance = this.codebook.Transform(test);
+
+            // Let us obtain the numeric output that represents the answer
+            int codeword = this.classifier.Decide(instance);
+
+            // Now let us convert the numeric output to an actual answer
+            this.result = this.codebook.Revert(this.outputColumn, codeword);
+
+            // We can also extract the probabilities for each possible answer
+            this.probs = this.classifier.Probabilities(instance);
+
+            return this.result;
+        }
+    }
+    #endregion
+
+    #region Helpers
+
+    #endregion
+
+}

src/DynAI/MachineLearning.csproj

@@ -0,0 +1,75 @@
+<?xml version="1.0" encoding="utf-8"?>
+<Project ToolsVersion="15.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
+  <Import Project="$(MSBuildExtensionsPath)\$(MSBuildToolsVersion)\Microsoft.Common.props" Condition="Exists('$(MSBuildExtensionsPath)\$(MSBuildToolsVersion)\Microsoft.Common.props')" />
+  <PropertyGroup>
+    <Configuration Condition=" '$(Configuration)' == '' ">Debug</Configuration>
+    <Platform Condition=" '$(Platform)' == '' ">AnyCPU</Platform>
+    <ProjectGuid>{5990DC45-28B6-4E1E-B3EA-C6549EE270F7}</ProjectGuid>
+    <OutputType>Library</OutputType>
+    <AppDesignerFolder>Properties</AppDesignerFolder>
+    <RootNamespace>AI</RootNamespace>
+    <AssemblyName>AI</AssemblyName>
+    <TargetFrameworkVersion>v4.5.2</TargetFrameworkVersion>
+    <FileAlignment>512</FileAlignment>
+    <NuGetPackageImportStamp>
+    </NuGetPackageImportStamp>
+  </PropertyGroup>
+  <PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Debug|AnyCPU' ">
+    <DebugSymbols>true</DebugSymbols>
+    <DebugType>full</DebugType>
+    <Optimize>false</Optimize>
+    <OutputPath>bin\Debug\</OutputPath>
+    <DefineConstants>DEBUG;TRACE</DefineConstants>
+    <ErrorReport>prompt</ErrorReport>
+    <WarningLevel>4</WarningLevel>
+    <DocumentationFile>bin\Debug\AI.xml</DocumentationFile>
+  </PropertyGroup>
+  <PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Release|AnyCPU' ">
+    <DebugType>pdbonly</DebugType>
+    <Optimize>true</Optimize>
+    <OutputPath>bin\Release\</OutputPath>
+    <DefineConstants>TRACE</DefineConstants>
+    <ErrorReport>prompt</ErrorReport>
+    <WarningLevel>4</WarningLevel>
+  </PropertyGroup>
+  <ItemGroup>
+    <Reference Include="Accord, Version=3.7.0.0, Culture=neutral, PublicKeyToken=fa1a88e29555ccf7, processorArchitecture=MSIL">
+      <HintPath>..\packages\Accord.3.7.0\lib\net45\Accord.dll</HintPath>
+    </Reference>
+    <Reference Include="Accord.MachineLearning, Version=3.7.0.0, Culture=neutral, PublicKeyToken=fa1a88e29555ccf7, processorArchitecture=MSIL">
+      <HintPath>..\packages\Accord.MachineLearning.3.7.0\lib\net45\Accord.MachineLearning.dll</HintPath>
+    </Reference>
+    <Reference Include="Accord.Math, Version=3.7.0.0, Culture=neutral, PublicKeyToken=fa1a88e29555ccf7, processorArchitecture=MSIL">
+      <HintPath>..\packages\Accord.Math.3.7.0\lib\net45\Accord.Math.dll</HintPath>
+    </Reference>
+    <Reference Include="Accord.Math.Core, Version=3.7.0.0, Culture=neutral, PublicKeyToken=fa1a88e29555ccf7, processorArchitecture=MSIL">
+      <HintPath>..\packages\Accord.Math.3.7.0\lib\net45\Accord.Math.Core.dll</HintPath>
+    </Reference>
+    <Reference Include="Accord.Statistics, Version=3.7.0.0, Culture=neutral, PublicKeyToken=fa1a88e29555ccf7, processorArchitecture=MSIL">
+      <HintPath>..\packages\Accord.Statistics.3.7.0\lib\net45\Accord.Statistics.dll</HintPath>
+    </Reference>
+    <Reference Include="System" />
+    <Reference Include="System.Core" />
+    <Reference Include="System.Xml.Linq" />
+    <Reference Include="System.Data.DataSetExtensions" />
+    <Reference Include="Microsoft.CSharp" />
+    <Reference Include="System.Data" />
+    <Reference Include="System.Net.Http" />
+    <Reference Include="System.Xml" />
+  </ItemGroup>
+  <ItemGroup>
+    <Compile Include="MachineLearning.cs" />
+    <Compile Include="Properties\AssemblyInfo.cs" />
+  </ItemGroup>
+  <ItemGroup>
+    <None Include="packages.config" />
+  </ItemGroup>
+  <Import Project="$(MSBuildToolsPath)\Microsoft.CSharp.targets" />
+  <Import Project="..\packages\Accord.3.7.0\build\Accord.targets" Condition="Exists('..\packages\Accord.3.7.0\build\Accord.targets')" />
+  <Target Name="EnsureNuGetPackageBuildImports" BeforeTargets="PrepareForBuild">
+    <PropertyGroup>
+      <ErrorText>This project references NuGet package(s) that are missing on this computer. Use NuGet Package Restore to download them.  For more information, see http://go.microsoft.com/fwlink/?LinkID=322105. The missing file is {0}.</ErrorText>
+    </PropertyGroup>
+    <Error Condition="!Exists('..\packages\Accord.3.7.0\build\Accord.targets')" Text="$([System.String]::Format('$(ErrorText)', '..\packages\Accord.3.7.0\build\Accord.targets'))" />
+  </Target>
+</Project>

src/DynAI/Properties/AssemblyInfo.cs

@@ -0,0 +1,36 @@
+using System.Reflection;
+using System.Runtime.CompilerServices;
+using System.Runtime.InteropServices;
+
+// General Information about an assembly is controlled through the following
+// set of attributes. Change these attribute values to modify the information
+// associated with an assembly.
+[assembly: AssemblyTitle("DynAI Machine Learning")]
+[assembly: AssemblyDescription("")]
+[assembly: AssemblyConfiguration("")]
+[assembly: AssemblyCompany("")]
+[assembly: AssemblyProduct("DynAI Machine Learning")]
+[assembly: AssemblyCopyright("Copyright © Radu Gidei 2017")]
+[assembly: AssemblyTrademark("")]
+[assembly: AssemblyCulture("")]
+
+// Setting ComVisible to false makes the types in this assembly not visible
+// to COM components.  If you need to access a type in this assembly from
+// COM, set the ComVisible attribute to true on that type.
+[assembly: ComVisible(false)]
+
+// The following GUID is for the ID of the typelib if this project is exposed to COM
+[assembly: Guid("5990dc45-28b6-4e1e-b3ea-c6549ee270f7")]
+
+// Version information for an assembly consists of the following four values:
+//
+//      Major Version
+//      Minor Version
+//      Build Number
+//      Revision
+//
+// You can specify all the values or you can default the Build and Revision Numbers
+// by using the '*' as shown below:
+// [assembly: AssemblyVersion("1.0.*")]
+[assembly: AssemblyVersion("0.1.0.0")]
+[assembly: AssemblyFileVersion("0.1.0.0")]

src/DynAI/packages.config

@@ -0,0 +1,7 @@
+<?xml version="1.0" encoding="utf-8"?>
+<packages>
+  <package id="Accord" version="3.7.0" targetFramework="net452" />
+  <package id="Accord.MachineLearning" version="3.7.0" targetFramework="net452" />
+  <package id="Accord.Math" version="3.7.0" targetFramework="net452" />
+  <package id="Accord.Statistics" version="3.7.0" targetFramework="net452" />
+</packages>