https://github.com/ROCm/tensorflow-upstream

TensorFlow ROCm port
https://github.com/ROCm/tensorflow-upstream

Last synced: 25 days ago
JSON representation

TensorFlow ROCm port

• Host: GitHub
• URL: https://github.com/ROCm/tensorflow-upstream
• Owner: ROCm
• License: apache-2.0
• Fork: true (tensorflow/tensorflow)
• Created: 2018-04-09T21:24:50.000Z (about 6 years ago)
• Default Branch: develop-upstream
• Last Pushed: 2024-04-09T23:02:46.000Z (about 1 month ago)
• Last Synced: 2024-04-10T18:27:19.984Z (about 1 month ago)
• Language: C++
• Homepage: https://tensorflow.org
• Size: 1000 MB
• Stars: 675
• Watchers: 51
• Forks: 89
• Open Issues: 120
• Metadata Files:
  • Readme: README.ROCm.md
  • Changelog: CHANGELOG.md
  • Contributing: CONTRIBUTING.md
  • License: LICENSE
  • Code of conduct: CODE_OF_CONDUCT.md
  • Citation: CITATION.cff
  • Codeowners: CODEOWNERS
  • Security: SECURITY.md

Lists

• awesome-python-data-science - tensorflow-upstream - TensorFlow ROCm port. <img height="20" src="img/tf_big2.png" alt="sklearn"> <img height="20" src="img/amd_big.png" alt="Possible to run on AMD GPU"> (Deep Learning / TensorFlow)
• awesome-datascience - tensorflow-upstream
• awesome-python-machine-learning-resources - GitHub - 16% open · ⏱️ 23.08.2022): (机器学习框架)
• fintech-awesome-libraries - tensorflow-upstream - TensorFlow ROCm port. (Tensor Flow / Automated Machine Learning)
• awesome-starred - ROCmSoftwarePlatform/tensorflow-upstream - TensorFlow ROCm port (others)
• awesome-datascience - tensorflow-upstream
• awesome-python-data-science - tensorflow-upstream - TensorFlow ROCm port. <img height="20" src="img/tf_big2.png" alt="sklearn"> <img height="20" src="img/amd_big.png" alt="Possible to run on AMD GPU"> (Deep Learning / TensorFlow)

README

        
# Tensorflow ROCm port #

## Introduction ##

This repository hosts the port of [Tensorflow](https://github.com/tensorflow/tensorflow) on ROCm platform. It uses various technologies on ROCm platform such as HIP and MIOpen. For details on HIP, please refer [here](https://github.com/GPUOpen-ProfessionalCompute-Tools/HIP). Optimized DNN library calls (via [MIOpen](https://github.com/ROCmSoftwarePlatform/MIOpen)) are also supported within this codebase.

## Installation ##

For further background information on ROCm, refer [here](https://github.com/RadeonOpenCompute/ROCm/blob/master/README.md).

For details on installation and usage, see these links:

* [Basic installation](rocm_docs/tensorflow-install-basic.md)

* [Building from source](rocm_docs/tensorflow-build-from-source.md)

* [Quickstart guide](rocm_docs/tensorflow-quickstart.md)

## Technical details ##

* [Overview of ROCm port](rocm_docs/rocm-port-overview.md)

* [List of supported operators on ROCm](rocm_docs/core_kernels.md)