TimHanewich.NeuralNetwork 0.1.0

dotnet add package TimHanewich.NeuralNetwork --version 0.1.0                
NuGet\Install-Package TimHanewich.NeuralNetwork -Version 0.1.0                
This command is intended to be used within the Package Manager Console in Visual Studio, as it uses the NuGet module's version of Install-Package.
<PackageReference Include="TimHanewich.NeuralNetwork" Version="0.1.0" />                
For projects that support PackageReference, copy this XML node into the project file to reference the package.
paket add TimHanewich.NeuralNetwork --version 0.1.0                
#r "nuget: TimHanewich.NeuralNetwork, 0.1.0"                
#r directive can be used in F# Interactive and Polyglot Notebooks. Copy this into the interactive tool or source code of the script to reference the package.
// Install TimHanewich.NeuralNetwork as a Cake Addin
#addin nuget:?package=TimHanewich.NeuralNetwork&version=0.1.0

// Install TimHanewich.NeuralNetwork as a Cake Tool
#tool nuget:?package=TimHanewich.NeuralNetwork&version=0.1.0                

TimHanewich.NeuralNetwork

Neural network system built on .NET for embedding machine learning (AI) in any project

To download the latest version of this NuGet package in a dotnet package, run the following command with the .NET CLI:

dotnet add package TimHanewich.NeuralNetwork

In this example, we will train a new neural network to model a NAND gate (https://en.wikipedia.org/wiki/NAND_gate).
We use the following class to create a NAND gate example. Copy and paste this into your coding environment to use it in this NAND training example.

public class Nand
{
    public float Input1 {get; set;}
    public float Input2 {get; set;}
    public float Output {get; set;}

    public static Nand NewRandom()
    {
        Nand ToReturn = new Nand();
        Random r = new Random();
        int ir = r.Next(0, 4);

        if (ir == 0)
        {
            ToReturn.Input1 = 0;
            ToReturn.Input2 = 0;
            ToReturn.Output = 1;
        }
        else if (ir == 1)
        {
            ToReturn.Input1 = 0;
            ToReturn.Input2 = 1;
            ToReturn.Output = 1;
        }
        else if (ir == 2)
        {
            ToReturn.Input1 = 1;
            ToReturn.Input2 = 0;
            ToReturn.Output = 1;
        }
        else if (ir == 3)
        {
            ToReturn.Input1 = 1;
            ToReturn.Input2 = 1;
            ToReturn.Output = 0;
        }

        return ToReturn;
    }
}

Place this statement at the top of your code to import the necessary resources:

using TimHanewich.NeuralNetwork;

To create a new Neural Network:

NeuralNetwork nn = NeuralNetwork.Create(new int[] {2, 3, 1}, true);

In the above code snippet, the first parameter, an int array, defines the structure of the network; the first value in this array is the number of input values, the last is the number of output values, and the middle values define how many hidden layers there should be and how many neurons should exist in each. The more complex the scenario, the more hidden layers are needed. A NAND gate is a simple model, so we only need one hidden layer with three neurons in it.

Training the Neural Network

for (int t = 0; t < 1000; t++)
{
    Nand n = Nand.NewRandom();
    nn.ForwardPropagate(new float[] {n.Input1, n.Input2});
    nn.BackwardPropagate(new float[] {n.Output}, 0.3f);
}

The ForwardPropagate method passes inputs through the model and returns the neural network's best guess at the correct output. Since we are only training the network, we can ignore the returned guess for now.

The BackwardPropagate method is the process of "training" our network. This method accepts two parameters: the ideal output(s) (an array), and the learning rate, how strongly the model should adjust during this particular training example. The value of 0.3 is slightly high for typical scenarios, but appropriate as this is a training set of only 1,000 examples.

Testing the Neural Network

for (int t = 0; t < 1000; t++)
{
    Nand n = Nand.NewRandom();
    float guess = nn.ForwardPropagate(new float[] {n.Input1, n.Input2})[0];
    Console.Write("Ideal: " + n.Output.ToString() + "    Guess: " + Math.Abs(guess).ToString("#,##0"));
    Console.ReadLine();
}

In the above snippet, we are using the same ForwardPropagate method as before, but this time we are capturing the network's prediction in the guess variable. We then write the ideal (correct) output and the guess value to the console.

The results of the above code:

Ideal: 0    Guess: 0
Ideal: 1    Guess: 1
Ideal: 1    Guess: 1
Ideal: 0    Guess: 0
Ideal: 0    Guess: 0
Ideal: 1    Guess: 1
Ideal: 1    Guess: 1
Ideal: 0    Guess: 0
Ideal: 1    Guess: 1
Ideal: 0    Guess: 0
Ideal: 0    Guess: 0
Ideal: 1    Guess: 1
Product Compatible and additional computed target framework versions.
.NET net5.0 was computed.  net5.0-windows was computed.  net6.0 was computed.  net6.0-android was computed.  net6.0-ios was computed.  net6.0-maccatalyst was computed.  net6.0-macos was computed.  net6.0-tvos was computed.  net6.0-windows was computed.  net7.0 was computed.  net7.0-android was computed.  net7.0-ios was computed.  net7.0-maccatalyst was computed.  net7.0-macos was computed.  net7.0-tvos was computed.  net7.0-windows was computed.  net8.0 was computed.  net8.0-android was computed.  net8.0-browser was computed.  net8.0-ios was computed.  net8.0-maccatalyst was computed.  net8.0-macos was computed.  net8.0-tvos was computed.  net8.0-windows was computed. 
.NET Core netcoreapp2.0 was computed.  netcoreapp2.1 was computed.  netcoreapp2.2 was computed.  netcoreapp3.0 was computed.  netcoreapp3.1 was computed. 
.NET Standard netstandard2.0 is compatible.  netstandard2.1 was computed. 
.NET Framework net461 was computed.  net462 was computed.  net463 was computed.  net47 was computed.  net471 was computed.  net472 was computed.  net48 was computed.  net481 was computed. 
MonoAndroid monoandroid was computed. 
MonoMac monomac was computed. 
MonoTouch monotouch was computed. 
Tizen tizen40 was computed.  tizen60 was computed. 
Xamarin.iOS xamarinios was computed. 
Xamarin.Mac xamarinmac was computed. 
Xamarin.TVOS xamarintvos was computed. 
Xamarin.WatchOS xamarinwatchos was computed. 
Compatible target framework(s)
Included target framework(s) (in package)
Learn more about Target Frameworks and .NET Standard.
  • .NETStandard 2.0

    • No dependencies.

NuGet packages

This package is not used by any NuGet packages.

GitHub repositories

This package is not used by any popular GitHub repositories.

Version Downloads Last updated
0.1.0 544 7/13/2020