uzo2005 / matinv

Got Bored of repeatedly calculating inverse and adjoints of matrices, so I wrote code to automate it.

We are currently doing linear algebra as a course and so I do lot of boring-ass calculations involving matrices. In this repo, I have set out to automate all the boring calculations, so that I can spend less time on my assignments.

Usage

The intended use-case for this project is quick verification of hand-computed results of linear algebra problem sets and theorems. To be able to do this and maintain accuracy, I entirely avoided using floating point types and instead implemented Fraction as a numeric type. Using fractions to compute arithmetic operations ensured that every result maintained 100% accuracy and that no floating-point error could hinder exploration.

For operations which involve row operations, calling showSteps on the result of these functions would print the necessary row operation steps needed to arrive at the final values.

I would strongly advise readers not to use code from this repository in their own applications because I have sacrificed speed of execution for the fun of exploration.

The achilles heel of all my computations is complex numbers. I havent had any need to worry about them yet, so I would postpone their implementations until further notice.

So far I have implemented:

1. Determinants Of Matrices
2. Adjoint(Adjucate?) Of Matrices
3. Inverse Of Matrices (will throw exception for singular matrices)
4. Normal and Reduced Echelon Forms of Matrix, and hence verification of consistency for a linear system of equations
5. LU Factorisation of matrices

There are many things that could go on the list above(e.g transposition of matrices), but I feel like they are not worthy of mention, since their implementations and computations are trivial. If you are interested in all the functionalities, dig into the code and you will find them.

I hope to also implement:

1. Eigenvalues and Eigenvectors
2. Diagonilization of Matrices

Also if I have time, I will make this a website so other people can use it.