n1arash / Nomad

Nomad Lightweight matrix library

NOMAD

Welcome To Nomad, A lightweight Matrix manipulation library currently under development.

Nomad's Matrix class supports a variaty of different functions such as Norm Calculation, Dot Product, Inverse Calculation, Transpose Operation, Function Mapping, Random Initialization, Reshape Operation, Vector Broadcasting, Uniform and Gaussian Random Distribution, etc...

You can find the full list of all the capabilities in the matrix class documentation page.

Documentation

The full API Documentation explaination on all aspects of the library, you can check it out here.

Quickstart

(The library will be available in NuGet as soon as the first release)

The quickstart guide will get you up and running with the nomad library in no time.

using System;
using Nomad.Matrix;
using Nomad.Utility;

namespace Sample
{
    public class Program
    {
        public static void Main()
        {
            // Create a matrix of size 5 (rows) and 10 (columns)
            Matrix mat = new Matrix(5, 10);

            // Randomly Initialize the values of your matrix with Gaussian Distribution 
            // Gaussian Dispersion will be ((high - low) / 2)
            mat.InRandomize(0.0, 1.0, EDistribution.Gaussian);

            // Transpose the Matrix mat and put it in matT
            Matrix matT = mat.T();

            // Calculate the dot product between the two matrices mat and matT and store the result in product
            Matrix product = mat * matT;

            // Flatten the product matrix inplace
            product.InFlatten();

            // Make sure the matrix has been transformed into a vector as expected
            Console.WriteLine(product.Type().ToString());

            // Transpose the vector inplace so it will print in 1 line
            product.InTranspose();

            // Convert the product vector into a string
            string str = product.ToString();

            // Print the product Vector
            Console.WriteLine(str);

            // End the program
            return;
        }
    }
}