# TensorFlow 安裝配置
---
## 現(xiàn)有的機器環(huán)境
> `CentOS 7 (Linux 3.10.0-327.el7.x86_64)`
> `gcc 4.8.5`
> `glibc 2.17`
> `cuda 7.5`
> 沒有 `g++`、`libstdc++`、`cudnn`、`Java 1.8` 和 `bzip2`、`unzip`、`swig`、`zlib-devel`
> 因為使用 `Anaconda` 提供 `python` 環(huán)境，所以并不依賴系統(tǒng)的 `python 2.7.5`
## 先使用管理員權(quán)限做一些必要的準備工作
### 1. 設置新用戶 danliu
```
[root ~]# useradd danliu
[root ~]# passwd danliu
```
### 2. 安裝必備軟件
#### g++ 4.8.5
```
[root ~]# yum install gcc-c++
```
libstdc++ 會作為依賴項一并安裝
### bzip2, unzip, swig 和 Java 1.8
```
[root ~]# yum install bzip2 unzip swig zlib-devel java-1.8.0-openjdk-devel
```
### package.zip
[root ~]# cd /home/
[root ~]# unzip packages.zip
[root ~]# cd /packages/
[root ~]# mv * /home/danliu
[root ~]# cd ..
[root ~]# rm -rf packages
[root ~]# cd /home/danliu/
### bazel
bazel 作為構(gòu)建工具，安裝在系統(tǒng)里更方便。
```
[root ~]# chmod +x bazel-0.3.1-installer-linux-x86_64.sh
[root ~]# ./bazel-0.3.1-installer-linux-x86_64.sh
```
## 使用普通賬戶 danliu 登陸并繼續(xù)接下來的安裝
### 安裝 Anaconda
```
[root ~]# chown -R danliu:danliu /home/danliu
[root ~]# su danliu
[danliu ~]$ chmod +x Anaconda2-4.1.1-Linux-x86_64.sh
[danliu ~]$ ./Anaconda2-4.1.1-Linux-x86_64.sh
Do you wish the installer to prepend the Anaconda2 install location
to PATH in your /home/danliu/.bashrc ? [yes|no]
[no] >>> yes
```
默認的安裝目錄是 `~/anaconda2/` (`/root/anaconda2`)
安裝完成后
```
source ~/.bashrc
```
### 添加已有的 cuda 7.5 安裝
和下面的安裝 cudnn v5 一起做
### 安裝 cudnn v5
```
[danliu ~]$ tar -xzvf cudnn-7.5-linux-x64-v5.0-ga.tar.gz
[danliu ~]$ mv cuda/ cudnnv5
```
把下面的內(nèi)容添加到 `~/.bashrc` 里
```
# CUDA
vim ~/.bashrc
export PATH=/usr/local/cuda-7.5/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda-7.5/lib64:$LD_LIBRARY_PATH
export LD_LIBRARY_PATH=$HOME/cudnnv5/lib64:$LD_LIBRARY_PATH
```
### 修改 TensorFlow 的一處 bug
TensorFlow 下的 `third_party/gpus/crosstool/CROSSTOOL` 存在一個可能導致構(gòu)建出錯的問題，如果使用 Ubuntu 14.04 可以無視，使用 CentOS 7 則必須要修改
```
[danliu ~]$ unzip tensorflow-src.zip
[danliu ~]$ mv tensorflow-master tensorflow-cudnnv5
[danliu ~]$ cd tensorflow-cudnnv5
[danliu ~]$ vi third_party/gpus/crosstool/CROSSTOOL
```
在第 66 行和第 67 行之間添加一行`cxx_builtin_include_directory: "/usr/local/cuda-7.5/include"`
```
@@ -64,6 +64,7 @@
cxx_builtin_include_directory: "/usr/lib/gcc/"
cxx_builtin_include_directory: "/usr/local/include"
cxx_builtin_include_directory: "/usr/include"
+? cxx_builtin_include_directory: "/usr/local/cuda-7.5/include"
tool_path { name: "gcov" path: "/usr/bin/gcov" }
# C(++) compiles invoke the compiler (as that is the one knowing where
```
### 配置 TensorFlow
```
[danliu *]$ ./configure
```
注意 cudnn 的路徑是之前設置的 `/home/danliu/cudnnv5`
> ERROR: Problem getting numpy include path.? Is numpy installed?
### 編譯和測試
```
# 下面的 // 不是注釋是一行
[danliu *]$ bazel build -c opt --config=cuda //tensorflow/cc:tutorials_example_trainer
[danliu *]$ bazel-bin/tensorflow/cc/tutorials_example_trainer --use_gpu
```
### 構(gòu)建可以遷移的 tensorflow wheel 包
```
[danliu *]$ bazel build -c opt --config=cuda //tensorflow/tools/pip_package:build_pip_package
[danliu *]$ bazel-bin/tensorflow/tools/pip_package/build_pip_package ./tensorflow_pkg
```
> bazel-bin/tensorflow/tools/pip_package/build_pip_package:行65: rsync: 未找到命令
