Today we are pleased to announce the release of ArrayFire v3.0. This new version features major changes to ArrayFire’s visualization library, a new CPU backend, and dense linear algebra for OpenCL devices. It also includes improvements across the board for ArrayFire’s OpenCL backend. A complete list ArrayFire v3.0 updates and new features can be found in the product Release Notes.
With over 8 years of continuous development, the open source ArrayFire library is the top CUDA and OpenCL software library. ArrayFire supports CUDA-capable GPUs, OpenCL devices, and other accelerators. With its easy-to-use API, this hardware-neutral software library is designed for maximum speed without the hassle of writing time-consuming CUDA and OpenCL device code. With ArrayFire’s library functions, developers can maximize productivity and performance. Each of ArrayFire’s functions has been hand-tuned by CUDA and OpenCL experts.
Major updates and new features
- Major changes to the visualization library
- Introducing handle based C API
- New backend: CPU fallback available for systems without GPUs
- Dense linear algebra functions available for all backends
- Support for 64 bit integers
New functions added in the following categories
- Data generation
- Computer Vision
- Image Processing
- Linear Algebra
- Visualization
Benefits of the binary installers
We’re now using MKL in our installers to speed up CPU and OpenCL backends. The CUDA backend is compiled to be optimized for each CUDA architecture, making it more portable. By default, building the open source version compiles only for the GPUs present in the system.
What people are saying
Kent Knox, Senior Member of Technical Staff from AMD, says:
Arrayfire is a model example of how open sourcing scientific libraries should work. They have made their own code open to the public for review by the community at large, and they build upon existing open source math libraries to improve their own. With their investment in robust automation, they enhance the correctness and performance of the overall scientific ecosystem.
Jason Ramapuram, a machine learning engineer, says:
ArrayFire has provided an elegant and simple solution for deploying GPU based machine learning applications. Being able to implement neural networks and auto encoders without delving into the any CUDA/OpenCL/BLAS details has been immensely helpful for research purposes. All of this is bundled in a brilliant open source package with an amazingly helpful team that is very open to implementing and resolving an issues that arise.
Availability
Visit ArrayFire’s website to download ArrayFire v3.0 Installers or GitHub account page to download and build the source code. The ArrayFire software library operates under the BSD 3-Clause License which enables unencumbered deployment and portability of ArrayFire for all uses, including commercially.
Dedicated support and coding services
ArrayFire offers dedicated support packages for ArrayFire users.
ArrayFire serves many clients through consulting and coding services, algorithm development, porting code, and training courses for developers.
Comments 5
Is there an R language interface to arrayfire?
@MySchizoBuddy:disqus We have this: https://github.com/arrayfire/arrayfire_r
This is still in development stages. It should work on OSX and Linux. Due to compiler restrictions, we are not sure it will work on Windows. We are actively looking to provide better support for the R language.
Please write a blog post with examples in R.
You can look at a sample example over here: https://github.com/arrayfire/arrayfire_r/blob/master/examples/Simple/MonteCarloPi.R
We plan to write a blog post in the next month or so.
how to run the benchmarks in the share folder, like normal c++ application? Also where is the opencl kernel of the Naive bayes classifier?