Repository files navigation

Documentation

NVIDIA has created this project to support newer hardware and improved librariesto NVIDIA GPU users who are using TensorFlow 1.x. With release of TensorFlow 2.0,Google announced that new major releases will not be provided on the TF 1.x branchafter the release of TF 1.15 on October 14 2019. NVIDIA is working with Google andthe community to improve TensorFlow 2.x by adding support for new hardware andlibraries. However, a significant number of NVIDIA GPU users are still usingTensorFlow 1.x in their software ecosystem. This release will maintain APIcompatibility with upstream TensorFlow 1.15 release. This project will be henceforthreferred to as nvidia-tensorflow.

Link to TensorflowREADME

• Ubuntu 18.04 or later (64-bit)
• GPU support requires a CUDA®-enabled card
• For NVIDIA GPUs, the r450 driver must be installed

For wheel installation:

• Python 3.6
• pip 19.0 or later

See thenvidia-tensorflow install guide to use thepip package, topull and run Docker container, andcustomize and extend TensorFlow.

NVIDIA wheels are not hosted on PyPI.org. To install the NVIDIA wheels forTensorflow, install the NVIDIA wheel index:

$ pip install --user nvidia-pyindex

To install the current NVIDIA Tensorflow release:

$ pip install --user nvidia-tensorflow[horovod]

Thenvidia-tensorflow package includes CPU and GPU support for Linux.

By using the software you agree to fully comply with the terms andconditions of the SLA (Software License Agreement):

• CUDA –https://docs.nvidia.com/cuda/eula/index.html#abstract

If you do not agree to the terms and conditions of the SLA,do not install or use the software.

Please review theContribution Guidelines.

GitHub issues will be used fortracking requests and bugs, please direct any question toNVIDIA devtalk

Apache License 2.0