centos 7 pip install tensorflow

最新推荐文章于 2025-06-21 15:59:29 发布

mydear_11000

最新推荐文章于 2025-06-21 15:59:29 发布

阅读量974

点赞数

CC 4.0 BY-SA版权

本文链接：https://blog.youkuaiyun.com/mydear_11000/article/details/52913090

本文介绍如何在Linux环境下安装TensorFlow 0.11版本所需的依赖包，包括配置开发环境所需的各种组件，如Python开发工具、numpy、scipy等，并通过pip安装特定版本的TensorFlow。

echo -e "\e[36m***Installing unzip git-all pkg-config zip g++ zlib1g-dev zlib-devel. Press ENTER when prompted*** \e[0m"
yum -y install unzip git-all pkg-config zip g++ zlib1g-dev zlib-devel

echo -e "\e[36m***Installing swig. Press ENTER when prompted*** \e[0m"
yum -y install swig

echo -e "\e[36m***Installing python-devel python-nose python-setuptools. Press ENTER when prompted*** \e[0m"
yum install python-devel python-nose python-setuptools

echo -e "\e[36m***Installing numpy python-devel scipy. Press ENTER when prompted*** \e[0m"
yum -y install numpy python-devel scipy

echo -e "\e[36m***Installing epel-release. Press ENTER when prompted*** \e[0m"
yum -y install epel-release

echo -e "\e[36m***Installing python-pip python-wheel. Press ENTER when prompted*** \e[0m"
yum install --enablerepo=* python-pip python-wheel
pip install --upgrade pip

echo -e "\e[36m***Installing tensorflow-0.11. Press ENTER when prompted*** \e[0m"
pip install --upgrade https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.11.0rc1-cp27-none-linux_x86_64.whl

===============================================================================================================================
if error:
1 epel error rpm:
cd /opt/
mkdir yum.old
mv /etc/yum.repos.d/epel.repo yum.old/
mv /etc/yum.repos.d/epel-testing.repo yum.old/
yum list
free -g

if nokey:
rpm --import /etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-7

网速太慢：
echo MTU="1400" >> /etc/sysconfig/network-scripts/ifcfg-eth0 添加到文件开机或重启网络服务生效

ifconfig eth0 mtu 1400 临时生效

#git clone https://github.com/tensorflow/tensorflow
export TF_NEED_CUDA=0
export TF_ENABLE_XLA=1
export TF_NEED_JEMALLOC=1
export TF_NEED_OPENCL=0
export TF_NEED_GCP=0
export TF_NEED_HDFS=0
#bazel --output_base=/data1/zhangyuanpeng/workspace/builder/tensorflow/bazel/output --output_user_root=/data1/zhangyuanpeng/workspace
/builder/tensorflow/bazel build -c opt //tensorflow/cc:tutorials_example_trainer
#bazel --output_base=/data1/zhangyuanpeng/workspace/builder/tensorflow/bazel/output --output_user_root=/data1/zhangyuanpeng/workspace
/builder/tensorflow/bazel build -c opt --copt=-mavx --copt=-mavx2 --copt=-mfma --copt=-mfpmath=both --copt=-msse4.2 --copt=-msse4.1 -
-copt=-msse3 //tensorflow/tools/pip_package:build_pip_package
bazel-bin/tensorflow/tools/pip_package/build_pip_package /data/ypzhang/ypzhang/workspace/build/tensorflow.pkg