Difference between revisions of "Jetson/Libraries"

From eLinux.org
Jump to: navigation, search
(3rd-party libraries)
(NVIDIA libraries: Added a link to cuDNN and made the other descriptions not version dependent)
Line 5: Line 5:
  
 
== NVIDIA libraries ==
 
== NVIDIA libraries ==
* '''CUDA''': CUDA 6.0 (available from the [https://developer.nvidia.com/jetson-tk1-support Jetson TK1 Support page], following these [[Jetson/Installing_CUDA|instructions]]) works fine. Earlier versions have not been tested.
+
* '''CUDA''': CUDA GPGPU Compute Toolkit is included when you install [https://developer.nvidia.com/jetson-tk1-development-pack JetPack], or can be installed manually by following the [[Jetson/Installing_CUDA|instructions]]). Versions before CUDA 6.0 have not been tested.
* '''OpenCV4Tegra''': OpenCV4Tegra 2.4.9 works fine on [[Jetson TK1]] including the CUDA-accelerated 'gpu' module following these [[instructions]]. Other versions should also work, including
+
* '''cuDNN''': CUDA Deep Neural Network library, with more info on the [[Jetson/cuDNN|cuDNN]] page.
 +
* '''OpenCV4Tegra''': OpenCV4Tegra works well on [[Jetson TK1]] including the CUDA-accelerated 'gpu' module (but not including the 'nonfree' module of patented code). More info about OpenCV4Tegra is on the [[Jetson/Computer_Vision_Performance|Computer Vision Performance]] page. OpenCV4Tegra is included when you install [https://developer.nvidia.com/jetson-tk1-development-pack JetPack], or can be installed manually by following the [[Jetson/Installing_OpenCV|instructions]]. Other versions of open-source OpenCV or closed-source OpenCV4Tegra should also work.
  
 
== 3rd-party libraries ==
 
== 3rd-party libraries ==
 
* '''[https://github.com/arrayfire/arrayfire ArrayFire]''': ArrayFire is a high performance software library for parallel computing with an easy-to-use API. Its array based function set makes parallel programming simple. ArrayFire's multiple backends (CUDA, OpenCL and native CPU) make it platform independent and highly portable. ArrayFire is now Open Source and is available under a BSD 3-Clause license. The source code can be found on the [https://github.com/arrayfire/arrayfire ArrayFire GitHub Page].
 
* '''[https://github.com/arrayfire/arrayfire ArrayFire]''': ArrayFire is a high performance software library for parallel computing with an easy-to-use API. Its array based function set makes parallel programming simple. ArrayFire's multiple backends (CUDA, OpenCL and native CPU) make it platform independent and highly portable. ArrayFire is now Open Source and is available under a BSD 3-Clause license. The source code can be found on the [https://github.com/arrayfire/arrayfire ArrayFire GitHub Page].
 
* '''OpenCV''': OpenCV 2.4.9 works fine when built from source, including the CUDA-accelerated 'gpu' module, as explained in the wiki [[Jetson/Installing_OpenCV|instructions]]. Other versions probably also work fine. The Tegra CPU-optimized OpenCV4Tegra by NVIDIA also works, as mentioned above.
 
* '''OpenCV''': OpenCV 2.4.9 works fine when built from source, including the CUDA-accelerated 'gpu' module, as explained in the wiki [[Jetson/Installing_OpenCV|instructions]]. Other versions probably also work fine. The Tegra CPU-optimized OpenCV4Tegra by NVIDIA also works, as mentioned above.

Revision as of 22:03, 16 December 2014

Introduction

This page shows libraries & packages that have been tested on Jetson TK1, mentioning whether they work or not and also discussing any known issues, etc.

For a list of packages already available to be installed from the command line using apt-get, try searching or browsing the Ubuntu 14.04 LTS armhf repository.

NVIDIA libraries

  • CUDA: CUDA GPGPU Compute Toolkit is included when you install JetPack, or can be installed manually by following the instructions). Versions before CUDA 6.0 have not been tested.
  • cuDNN: CUDA Deep Neural Network library, with more info on the cuDNN page.
  • OpenCV4Tegra: OpenCV4Tegra works well on Jetson TK1 including the CUDA-accelerated 'gpu' module (but not including the 'nonfree' module of patented code). More info about OpenCV4Tegra is on the Computer Vision Performance page. OpenCV4Tegra is included when you install JetPack, or can be installed manually by following the instructions. Other versions of open-source OpenCV or closed-source OpenCV4Tegra should also work.

3rd-party libraries

  • ArrayFire: ArrayFire is a high performance software library for parallel computing with an easy-to-use API. Its array based function set makes parallel programming simple. ArrayFire's multiple backends (CUDA, OpenCL and native CPU) make it platform independent and highly portable. ArrayFire is now Open Source and is available under a BSD 3-Clause license. The source code can be found on the ArrayFire GitHub Page.
  • OpenCV: OpenCV 2.4.9 works fine when built from source, including the CUDA-accelerated 'gpu' module, as explained in the wiki instructions. Other versions probably also work fine. The Tegra CPU-optimized OpenCV4Tegra by NVIDIA also works, as mentioned above.