Supercomputing/horovod：Uber开发的分布式深度学习训练框架。

Enrico Minack

CI: fix GPU tests (#4311)

Move to v6.17.0 queue with CUDA 12.4
Upgrade buildkite plugins
Replace image-repository option with push
Use CUDA 11.8 container images, 12.0 for head
TF nightly has minimum Python version 3.9


Signed-off-by: Enrico Minack github@enrico.minack.dev

6个月前1348次提交

.buildkiteCI: fix GPU tests (#4311)6个月前
.githubCI: fix GPU tests (#4311)6个月前
cmakeBuild: Link CUDA runtime library statically by default (#3867)3年前
dockerPin cudf-cu11 and dask-cudf-cu11 for horovod-nvtabular docker build (#3981)2年前
docsDocs: Fix invalid link to local section (#3937)2年前
examplesFix mnist download url (#4033)2年前
horovodAdd support for disabling autotune parameter for batch prefetch (#4296)9个月前
testCI: fix GPU tests (#4311)6个月前
third_partyUpdate gloo submodule to commit d96897b (#3925)2年前
.clang-formatInitial commit8年前
.gitignorefix store, and add infinite loader (#2954)4年前
.gitmodulesUpdated torch c++ to use new aten api (#3175)4年前
.readthedocs.yamlCI: Pin requests<2.29.2 for docker-compose, update readthedocs build image (#3913)2年前
CHANGELOG.mdBump version to v0.28.1 (#3942)2年前
CMakeLists.txtCompile using C++17 when using PyTorch 2.1 (#3998)1年前
CODE_OF_CONDUCT.mdProposing "Code of Conduct"6年前
CONTRIBUTING.mdUpdated project governance (#2721)5年前
Dockerfile.test.cpuFix CI build and tests (#4310)6个月前
Dockerfile.test.gpuFix CI build and tests (#4310)6个月前
GOVERNANCE.mdUpdate GOVERNANCE.md (#3836)3年前
Jenkinsfile.ppc64leMove ppc64le CI to TF 2.6.0 and PyTorch 1.9.1 (#3208)4年前
LICENSEMove attributions to NOTICE file (#960)7年前
MANIFEST.inMove native libraries build to cmake (#2009)5年前
NOTICEUpdate to CMake 3.13 for better CUDA support and to enable build concurrency (#3261)4年前
README.rstAdd Helm Chart (#3546)3年前
SECURITY.mdCreate SECURITY.md (#3363)4年前
assert-package-versions.shReplace package tf-nightly-gpu by tf-nightly (#3826)3年前
docker-compose.test.ymlCI: fix GPU tests (#4311)6个月前
horovod.expFixed issue with dynamically linking PyTorch module on Mac OSX (#494)7年前
horovod.ldsFixed issue with dynamically linking PyTorch module on Mac OSX (#494)7年前
setup.cfgRefactored tests by run commands and added macOS CI (#2264)5年前
setup.pyUse pytest<8 so skipif continues to work for test classes (#4060)1年前

关于

Uber开发的分布式深度学习训练框架。

人工智能 python c++cmake

README.md

16.9 MB

邀请码

社区
网站首页
关于我们
教学实践
合作伙伴

支持与服务
帮助中心
API文档
Git常用命令
服务协议

加入我们
官网邮箱：gitlink@ccf.org.cn

QQ群

公众号