Today ArrayFire announced the release of Version 3.0 of their high-speed software library for GPU computing. The 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.
ArrayFire has provided an elegant and simple solution for deploying GPU based machine learning applications,” said Jason Ramapuram, a machine learning engineer at Qualcomm. “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.”
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
- Benefits of the binary installers
A complete list ArrayFire v3.0 updates and new features can be found in the product Release Notes.