深度学习开发环境配置

Deep-Learning

1 查看版本号

查看 CUDA 版本：

cat /usr/local/cuda/version.txt

查看 CUDNN 版本：

cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2

2 查找对应的匹配版本

这个在tensorflow-ros-cpp仓库的readme里面可以查看

我们需要安装的软件如下

• NVIDIA driver
• cuda
• cudnn
  以上三个软件的安装遵循之前的笔记即可
• tensorfow
• tensorflow-ros-cpp

3 TensorFlow 源码编译安装

3.1 Install Bazel

https://docs.bazel.build/versions/master/install-ubuntu.html
注意Bazel的版本，一样可能出现问题，我选择了bazel0.16的release文件安装的

• 卸载 bazel

rm -rf ~/.bazel ~/.cache/bazel/ ~/bin

3.2 Install TensorFlow

3.2.1 安装python tensorflow

• install tensorflow doc
• 安装依赖

sudo apt install python-dev python-pip  # or python3-dev python3-pip
pip install -U --user pip six numpy wheel setuptools mock 'future>=0.17.1'
pip install -U --user keras_applications==1.0.6 --no-deps
pip install -U --user keras_preprocessing==1.0.5 --no-deps

• Download the TensorFlow source code
  直接去release版本里面下载对应的版本，因为直接从git clone下载时间会特别长，下载完成解压
• 进行编译

cd tensorflow
./configure # 除了选择python版本，cuda点y之外，其他的都默认
# 下面两个选择其中之一进行编译
# 1 GPU suport
bazel build --config=opt --config=cuda //tensorflow/tools/pip_package:build_pip_package
bazel build --config=opt --config=cuda //tensorflow:libtensorflow_cc.so
# 2 cpu only
bazel build --config=opt //tensorflow/tools/pip_package:build_pip_package
bazel build --config=opt //tensorflow:libtensorflow_cc.so

• 编译出错解决办法

bazel clean --expunge # 清除编译文件，重新来过

• 编译完成后生成安装文件

bazel-bin/tensorflow/tools/pip_package/build_pip_package ~/tensorflow_package
# 安装
pip3 install ~/tensorflow_package/file_name.whl

• 测试安装是否成功

#!/usr/bin/python3
import tensorflow as tf
hello = tf.constant('Hello, TensorFlow!')
sess = tf.Session()
print(sess.run(hello))

如果弹出以下文字

usr/local/lib/python3.5/dist-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'.

说明是numpy版本和tensorflow版本不和，需要将numpy版本降级，降到1.6应该就ok了

# 1. remove
sudo apt-get remove python3-numpy
# 2. install
sudo python3 -m pip install numpy==1.16

3.2.2 安装tensorflow cpp lib

Install Reference

• protobuf eigen recompile

./tensorflow/contrib/makefile/download_dependencies.sh
source tensorflow/contrib/makefile/build_all_linux.sh
# protobuf
mkdir /tmp/proto
./tensorflow/contrib/makefile/download_dependencies.sh
cd tensorflow/contrib/makefile/downloads/protobuf/
./autogen.sh
./configure --prefix=/tmp/proto/
make
make install
# eigen
mkdir /tmp/eigen
cd ../eigen
mkdir build_dir
cd build_dir
cmake -DCMAKE_INSTALL_PREFIX=/tmp/eigen/ ../
make install
cd ../../../../../..

• 整理库文件和头文件

# lib
mkdir -p ../tf_test/lib
cp bazel-bin/tensorflow/libtensorflow_cc.so ../tf_test/lib/
cp bazel-bin/tensorflow/libtensorflow_framework.so ../tf_test/lib/  # 之前编译r0.12和r1.3版本的库，只需要libtensorflow_cc.so，1.4版本的似乎分成了两个so文件，即还需要libtensorflow_framework.so
cp /tmp/proto/lib/libprotobuf.a ../tf_test/lib/
# include
mkdir -p ../tf_test/include/tensorflow
cp -r bazel-genfiles/* ../tf_test/include/
cp -r tensorflow/cc ../tf_test/include/tensorflow
cp -r tensorflow/core ../tf_test/include/tensorflow
cp -r third_party ../tf_test/include
cp -r /tmp/proto/include/* ../tf_test/include
cp -r /tmp/eigen/include/eigen3/* ../tf_test/include
# nsync
门槛低人-跑
cp -r tensorflow/contrib/makefile/downloads/nsync/public ../tf_test/include/external/nsync/public
# 删除多余cc文件
cd ../tf_test/
find . -name "*.cc" -type f -delete

• cmake中使用

cmake_minimum_required(VERSION 3.0)
project(cpptensorflow)
set(CMAKE_CXX_STANDARD 11)
set(tensorflow_path /home/nrsl/Downloads/tf_test)
link_directories(${tensorflow_path}/lib)
include_directories(
        ${tensorflow_path}/include
)
add_executable(cpptensorflow main.cpp ann_model_loader.h model_loader_base.h ann_model_loader.cpp)
target_link_libraries(cpptensorflow tensorflow_cc tensorflow_framework)

4 Using tensorflow ros cpp

一个傻逼的包，劳资不用你了，我自己链接我的程序，多好用！！

pip install catkin_pkg empy pyyaml

测试tensorflow-ros-cpp安装是否成功
https://github.com/tradr-project/tensorflow_ros_test/tree/kinetic-devel

当然如果还是想要使用。。
编译方法如下，wiki链接

catkin build tensorflow_ros_cpp --cmake-args -DFORCE_TF_PIP_SEARCH="ON" \
    -DTF_PYTHON_VERSION="python3" \
    -DTF_PIP_PATH="/usr/local/lib/python3.5/dist-packages/tensorflow" \
    -DTF_PIP_EXECUTABLE="/home/nrsl/.local/bin/pip3" \
    -DTF_PYTHON_LIBRARY="/usr/lib/x86_64-linux-gnu/libpython3.5m.so"

5 其他错误记录

• 找不到libcu*.so.9.0

ImportError: libcublas.so.9.0: cannot open shared object file: No such file or directory

这是因为链接库的时候找不到cuda9的库，

sudo gedit /etc/ld.so.conf
# 添加
/usr/local/cuda-9.0/lib64
# 生效
sudo ldconfig

• pip3 install permission问题

# 找到pip3的安装位置， 比如我的是 /home/nrsl/.local/bin/pip3.5
locate pip3
# 建立软连接，这样就可以使用sudo pip-py3 install xx_pylib
sudo ln -s /home/nrsl/.local/bin/pip3.5 /usr/local/bin/pip-py3