zhaozhixu / TensorLight

A light-weight tensor operation library for C and CUDA

TensorLight

TensorLight is a light-weight tensor operation library for C and CUDA.

Prerequisites

The following steps have been tested for Ubuntu 16.04 but should work with other distros as well. Required packages can be installed using the following command:

sudo apt-get install build-essential perl git pkg-config check

If you want to build with CUDA support, you also have to install CUDA 8.0 or later according to their website CUDA Toolkit 8.0. Remember to put nvcc (usually in /usr/local/cuda/bin) in environment variable PATH.

Building and Installation

1. Clone this repository to your local directory.

   cd <my_working_directory>
   git clone https://github.com/zhaozhixu/TensorLight.git
   cd TensorLight
   

2. Build and install

   First, configure your installation using:

   chmod +x configure
   ./configure
   

   There are options to custom your building and installation process. You can append them after ./configure. For example, use

   ./configure --with-cuda=yes
   

   if you want to build with CUDA support. Detailed ./configure options can be printed using ./configure -h.

   After that, use make to compile the library and run the tests. Then make install to copy the library files and headers into the installation directory, or sudo make install if you don't have the permissions with that directory.

   make
   sudo make install
   

3. Other make options

   Use make info to see other make options. Especially, you can use make clean to clean up the build directory and all object files, and make uninstall to remove library files and headers from the installation directory.

Usage

Include tl_tensor.h in your project to use TensorLight functions.

Documentation is coming soon. But it should be familiar if you have experience with numpy in Python.

You can use the following command to get the compilation and linking flags when building your project.

pkg-config --cflags --libs tensorlight