PanagiotisPtr / Linear-Algebra-Library-CUDA

Simple implementation of a Linear Algebra Library in CUDA

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Linear Algebra Library CUDA

Simple implementation of a Linear Algebra Library in CUDA

How to run the code

The code is tested on linux ubuntu 16.04 with CUDA 8.0 and C++11 To run the code you will only need to get CUDA 8.0 installed and setup. And run the compile.sh script from the terminal.

Note: I do not have the nvcc compiler directory at the environment variables on my machine so if your directory is different make sure that you change it. Or if you already have nvcc on your environment variables you can remove it and just compile the main.cu file.

About the Project

This is a very simple implementation of vector/matrix operations on the GPU. I am not an expert with CUDA so the code may not be (and probably isn't) very efficient but it works.It mostly was an experimentation to see what it's like writing programs in CUDA. The matrix operations include operations such as Determinant, Inverse and Cofactor of a matrix.

I hope that you will find the code helpful and educational.

Panagiotis Petridis, High School Student

Greece

About

Simple implementation of a Linear Algebra Library in CUDA


Languages

Language:Cuda 99.5%Language:Shell 0.5%