ubuntu 18.04安装horovod

已于 2022-11-22 19:51:26 修改
阅读量2.2k 收藏 6
点赞数 1
分类专栏： horovod tensorflow 文章标签： ubuntu tensorflow python
于 2020-07-01 14:29:52 首次发布
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/w5688414/article/details/107063150
版权

今天在跑spellGCN的时候发现需要安装horovod，但是安装过程并不是很顺利，我这里分享一下我安装成功的方法：

我的报错日志为：

 File "/home/../anaconda3/envs/ctc/lib/python3.6/site-packages/horovod/tensorflow/__init__.py", line 25, in <module>
    check_extension('horovod.tensorflow', 'HOROVOD_WITH_TENSORFLOW', __file__, 'mpi_lib')
  File "/home/../anaconda3/envs/ctc/lib/python3.6/site-packages/horovod/common/util.py", line 56, in check_extension
    'Horovod with %s=1 to debug the build error.' % (ext_name, ext_env_var))
ImportError: Extension horovod.tensorflow has not been built.  If this is not expected, reinstall Horovod with HOROVOD_WITH_TENSORFLOW=1 to debug the build error.
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:516: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint8 = np.dtype([("qint8", np.int8, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint8 = np.dtype([("quint8", np.uint8, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint16 = np.dtype([("qint16", np.int16, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint16 = np.dtype([("quint16", np.uint16, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint32 = np.dtype([("qint32", np.int32, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  np_resource = np.dtype([("resource", np.ubyte, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:541: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint8 = np.dtype([("qint8", np.int8, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:542: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint8 = np.dtype([("quint8", np.uint8, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:543: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint16 = np.dtype([("qint16", np.int16, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:544: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint16 = np.dtype([("quint16", np.uint16, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:545: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint32 = np.dtype([("qint32", np.int32, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:550: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  np_resource = np.dtype([("resource", np.ubyte, 1)])
Traceback (most recent call last):
  File "../run_spellgcn.py", line 23, in <module>
    import optimization
  File "/home/../SpellGCN/optimization.py", line 23, in <module>
    import horovod.tensorflow as hvd
  File "/home/wugaosheng/anaconda3/envs/ctc/lib/python3.6/site-packages/horovod/tensorflow/__init__.py", line 25, in <module>
    check_extension('horovod.tensorflow', 'HOROVOD_WITH_TENSORFLOW', __file__, 'mpi_lib')
  File "/home/../anaconda3/envs/ctc/lib/python3.6/site-packages/horovod/common/util.py", line 56, in check_extension
    'Horovod with %s=1 to debug the build error.' % (ext_name, ext_env_var))
ImportError: Extension horovod.tensorflow has not been built.  If this is not expected, reinstall Horovod with HOROVOD_WITH_TENSORFLOW=1 to debug the build error.

安装

安装MPI，这一步不知道有没有用，反正我安装成功了：

wget https://download.open-mpi.org/release/open-mpi/v4.0/openmpi-4.0.0.tar.gz
tar -xzvf openmpi-4.0.0.tar.gz
cd openmpi-4.0.0
./configure --prefix=/usr/local --with-cuda-10.0
sudo make all install

安装NCCL 2:

git clone https://github.com/NVIDIA/nccl && cd nccl
make -j8
sudo make install

安装gcc 4.9

sudo vim /etc/apt/sources.list

#添加下面两行，并保存退出
deb http://dk.archive.ubuntu.com/ubuntu/ xenial main
 
deb http://dk.archive.ubuntu.com/ubuntu/ xenial universe


sudo apt update

sudo apt-get install gcc-4.9
sudo apt-get install g++-4.9

安装horvod:

HOROVOD_GPU_ALLREDUCE=NCCL HOROVOD_WITH_TENSORFLOW=1  pip install --no-cache-dir horovod

测试：

Python 3.6.10 |Anaconda, Inc.| (default, Mar 25 2020, 23:51:54)
[GCC 7.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import horovod.tensorflow as hvd
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:516: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint8 = np.dtype([("qint8", np.int8, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint8 = np.dtype([("quint8", np.uint8, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint16 = np.dtype([("qint16", np.int16, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint16 = np.dtype([("quint16", np.uint16, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint32 = np.dtype([("qint32", np.int32, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  np_resource = np.dtype([("resource", np.ubyte, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:541: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint8 = np.dtype([("qint8", np.int8, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:542: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint8 = np.dtype([("quint8", np.uint8, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:543: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint16 = np.dtype([("qint16", np.int16, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:544: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint16 = np.dtype([("quint16", np.uint16, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:545: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint32 = np.dtype([("qint32", np.int32, 1)])
/home/wugaosheng/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:550: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  np_resource = np.dtype([("resource", np.ubyte, 1)])
WARNING:tensorflow:From /home/wugaosheng/anaconda3/envs/ctc/lib/python3.6/site-packages/horovod/tensorflow/__init__.py:163: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.

WARNING:tensorflow:From /home/wugaosheng/anaconda3/envs/ctc/lib/python3.6/site-packages/horovod/tensorflow/__init__.py:189: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.

发现不报错了，应该安装成功了哈。

参考文献

[1].Ubuntu18.04安装gcc4.9 g++4.9失败解决方案（E: 软件包gcc还没有可供安装的候选者）. Ubuntu18.04安装gcc4.9 g++4.9失败解决方案（E: 软件包gcc还没有可供安装的候选者）_让计算机看懂世界的博客-CSDN博客

[2].How To Install Horovod with CUDA and NCCL (on Ubuntu 18.04). How To Install Horovod with CUDA and NCCL on Ubuntu 18.04 – 2KangHo

[3].Horovod.GitHub - horovod/horovod: Distributed training framework for TensorFlow, Keras, PyTorch, and Apache MXNet.

[4].Can't install horovod on power8 machine . https://github.com/horovod/horovod/issues/541

农民小飞侠
关注
  • 1
    点赞
  • 踩
  • 6
    收藏
    觉得还不错? 一键收藏
  • 打赏
    打赏
  • 2
    评论
专栏目录
在Linux服务器上安装horovod安装
m0_44999081的博客
11-04 1530
Horovod的安装步骤
Horovod分布式搭建——UbuntuServer 22.04
yx1405585468的博客
08-08 737
基于Ubuntu22.04的Horovod分布式环境搭建
Ubuntu安装horovod工作记录
最新发布
eternal963的博客
05-18 295
本文记录了本人在环境配置过程中具体使用的一些代码，可供参考，具体使用要根据情况而定。
如何安装Horovod？
sinat_22510827的博客
12-18 1459
如何安装分布式训练的Uber的神器horovod的GPU版本？ 这个是英文的介绍。 一共分3步。 安装NCCL2 安装OpenMpi 安装horovod 注意！！！注意！！！注意！！！最新版本的horovod要求tensorflow>=1.15.0版本。 一、安装NCCL2 首先先去下载NCCL2。地址 下载需要先注册Nvidia的账号，做一个非常简单的问卷。 下载对应的Local installers，安装repository。注意系统的版本和CUDA的版本。比如我是Ubuntu1
Horovod之分布式训练的使用(tensorflow)，注意事项以及加速优化
thkinglee的博客
04-13 2813
文章目录Horovod原理Horovod 安装Tensorflow例子1. Session(不使用hooks)2. MonitoredTrainingSession版本(使用hooks)运行注意事项注意事项总结分布式训练总结参考资料 最近由于工作需要，重新研究了下horovod。 Horovod原理 详细可看原论文 总结: Horovod 使用ring-all-reduce分布式计算方式 运行过...
Ubuntu18.04安装opencv 3.2.0的解决方法
09-15
主要介绍了Ubuntu18.04安装opencv 3.2.0的解决方法，非常不错，具有一定的参考借鉴价值,需要的朋友可以参考下
Ubuntu 18.04安装compose
01-08
Ubuntu 18.04安装compose 安装Docker-ce 更换国内软件源 sudo cp /etc/apt/sources.list /etc/apt/sources.list.bak sudo sed -i 's/archive.ubuntu.com/mirrors.ustc.edu.cn/g' /etc/apt/sources.list sudo apt ...
Ubuntu18.04安装Docker
01-07
１．sudo apt-get remove docker docker-engine docker.io containerd runc 首先移除已经装好的Docker ...３．sudo curl -fsSL ...这就安装完成了 最好再添加阿里云镜像源，这样拉取镜像更快一些，类似
Ubuntu18.04安装和使用Mysql数据库
12-14
在Ubuntu 18.04系统中安装和使用MySQL数据库是一个常见的任务，尤其对于开发者和系统管理员来说。这里提供了一个简化的8步流程来完成这一过程。 1. **安装MySQL服务端**： 使用`sudo apt-get install mysql-server...
Ubuntu18.04安装intel编译器的教程_202204261
08-03
在本文中，我们将详细介绍如何在Ubuntu 18.04系统上安装Intel的编译器套件，包括Intel oneAPI Base Toolkit和Intel oneAPI HPC Toolkit。这些工具集提供了高性能计算所需的编译器、数学库和其他开发工具。遵循以下...
人脸识别 (3) FaceNet安装与测试
light169的专栏
03-23 6129
参见：算法---FaceNet在Tf下的实战篇 - WP的烂笔头 - 博客园 2.2　下载文件：FaceNet源码＋模型 （１）FaceNet在github源码地址为: 　https://github.com/davidsandberg/facenet/tree/master 　FaceNet库源码下载链接：https://codeload.github.com/davidsandberg/facenet/zip/master　 （２）LFW数据集: http://vis-www.cs.uma.
horovod tensorflow 分布式多gpu
心之所向
07-16 2736
# 引入 try: import horovod.tensorflow as hvd hvd.init() except Exception as e: hvd = None print('no horovod') # 打印信息 if hvd: tf.logging.info('Total workers: {}, local workers: {}'.f...
Tensorflow安装成功，但在导入时报错的解决办法
热门推荐
x_lock的博客
08-16 2万+
在Mac上按照官网教程安装成功tensor flow后，但在程序中导入时，仍然报错，包括但不限于以下两个错误。对于这种错误，原因主要在于Mac内默认的python库比较老了，即便通过pip命令安装了新的包，python也会默认导入默认位置的包。这时候需要做的就是删除，有冲突的包，对于以下两个错误，就是分别时numpy和six两个包冲突了。可以在python命令行环境下，通过numpy.version
ImportError: C extension: No module named ‘pandas._libs.tslib‘ not built.错误信息的解决方法
theskylife的博客
12-12 2万+
最近在使用pandas时，在import pandas模块时，出现了以下的错误信息 ： ModuleNotFoundError: No module named 'pandas._libs.tslib' During handling of the above exception, another exception occurred: ImportError ...
Ubuntu20.04下Horovod GPU安装
weixin_45385568的博客
11-08 715
前言：之前在看分布式训练的时候看到了这个框架，尝试安装并使用，但是在安装和使用过程中踩了很多的坑，这里记录一下。（关于分布式训练有个系列文章很不错，链接我放在这里大家和可以看看。） 单机多卡-1.理论基础、单机多卡-2.TensorFlow、单机多卡-3.Pytorch、单机多卡-4.Horovod 1.安装流程 我们先来到horovod的官网：horovod安装引导 这里我们看到了一些基本的要求，包括python3，cmake，g++等等，安装过程比较简单这里不在赘述。同时为了使用显卡，cuda+cud
python tqdm pytorch_pytorch使用horovod多gpu训练的实现
weixin_31832681的博客
03-01 942
pytorch在Horovod上训练步骤分为以下几步：import torchimport horovod.torch as hvd# Initialize Horovod 初始化horovodhvd.init()# Pin GPU to be used to process local rank (one GPU per process) 分配到每个gpu上torch.cuda.set_devi...
[深度学习] 分布式Horovod介绍（四）
摩登都市天空---专栏
04-26 1万+
[深度学习] 分布式模式介绍（一） [深度学习] 分布式Tensorflow介绍（二） [深度学习] 分布式Pytorch 1.0介绍（三） [深度学习] 分布式Horovod介绍（四） 实际应用中，单机多卡的同步式数据并行是最常用的，在论文中最常见的训练方式是单机八卡. 数据再多一般就需要多机多卡. 无论是单机多卡，还是多机多卡，均是分布式训练，在horovod出现之前，使用te...
ubuntu18.04安装后
05-01
安装Ubuntu 18.04并不难，但是在使用之前需要进行一些必要的配置和安装软件包等工作。 安装Ubuntu 18.04 首先，您需要下载Ubuntu 18.04的ISO镜像并将其写入到USB驱动器中。在安装过程中，您需要选择一些首选项，...

“相关推荐”对你有帮助么？

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论 2
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包

打赏作者

农民小飞侠

你的鼓励将是我创作的最大动力

¥1 ¥2 ¥4 ¥6 ¥10 ¥20
扫码支付：¥1
获取中
扫码支付

您的余额不足，请更换扫码支付或充值

打赏作者

实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值