Linux14.04+cuda7.5+anaconda2+opencv3.0+caffe
来源:互联网 发布:8051单片机引脚映射 编辑:程序博客网 时间:2024/04/29 21:12
现在安装caffe的博客着实很多,但是很多都是复制转载别人的,根本不管对错,有些没写清楚自己的环境,博主自己通过安装,一步一步记下自己的步骤,写下自己踩过的坑,希望大家借鉴,也是给自己做个笔记。
首先介绍博主电脑硬件和所安装的软件版本,gpu是GTX980Ti,Linux版本是14.04(LTS)64位,python环境用的是anaconda2,cuda版本是7.5,OpenCV用的是3.0。
博主在安装之前的机器是纯干净的,啥玩意儿都没有,在安装环境之前,我把硬盘格式化了,所以是特别干净的盘,电脑只有bios,这是我安装的一个前提吧。
一、安装ubuntu
下载ubuntu
制作ubuntu的系统u盘(博主是在win7的环境中制作的系统盘),博主用的是UltraISO(这个好早就下了,所以忘了是哪下的了,网上有很多的,自己百度谷歌雅虎一下吧)
1)博主这个包,解压出来是个exe,可以直接运行,运行界面如下,博主很穷所以选择试用
2)进入如下界面,选择 文件-打开
3)然后找到你刚刚下载的ubuntu的iso文件(ubuntu-14.04-desktop-amd64.iso),为什么要打码,嘿嘿,你猜
4)可以看见软件打开了iso镜像文件
5)选择 启动-写入硬盘映像
6)插入你的u盘,然后更改如下参数(选择USB-HDD,选择便捷启动中的syslinux,最后点击是)操作如下
7)最后选择写入,等待u盘做好
将制作好的系统盘插入即将安装系统的电脑,开机,一直按f2(装系统的电脑也就是要安装ubuntu的电脑主板是华硕的),进入bios后,选择u盘启动,然后可以进入u盘系统盘,选择install ubuntu,之后按照提示安装即可。
安装好后拔出u盘,重启电脑
二、安装cuda
简单来说,按照官方文档安装即可,请认真阅读,不要侥幸的跳过任何一个步骤任何一个错误,我开始用的是.run文件装的,不是很顺手,不方便,还要另装显卡驱动,容易出错,后来同学之前踩过这个坑,建议用deb安装,很方便,换成.deb方式安装的,三条语句搞定。
官方安装文档网址
[下载cuda网址](https://developer.nvidia.com/cuda-toolkit-archive)
下载完后运行md5sum和下面网页中的checksums对比看看下载是否正确完成
接下来三条语句搞定,不过建议还是认真看看文档,就当学习一下英文了
在设置环境变量的时候要注意,官方文件写在终端输入
export PATH=/usr/local/cuda-7.5/bin:$PATHexport LD_LIBRARY_PATH=/usr/local/cuda-7.5/lib64:$LD_LIBRARY_PATH
这样做,当你关闭终端的时候变量就没了。
本人在/etc/profile里末尾添加上面两句,添加完了回到终端source /etc/profile
,这样就不用重启就能改变变量配置了。
接下来是检查是否安装cuda成功,cuda-install-samples-7.5.sh ./ ,表示在当前文件夹下安装samples,然后进入NVIDIA_CUDA-7.5_Samples下make
make成功后cd到NVIDIA_CUDA-7.5_Samples/bin/x86_64/linux/release下,运行./deviceQuery看是否安装成功
运行./bandwidthTest看显卡和cuda有没有连接成功
三、安装anaconda2
安装anaconda,从官网下载anaconda2,我下的是Anaconda2-4.2.0-Linux-x86_64.sh,然后cd到包含该文件的路径下,运行bash Anaconda2-4.2.0-Linux-x86_64.sh,一路enter,第一个问题Do you approve the license terms?选yes,然后问你安装路径,我和[4]一样安装在/usr/local/anaconda2下,之后问你是否添加anaconda路径,选yes,,它自动将anaconda2的bin路径添加到~/.bashrc,然后就安装好了。
在终端输入python
命令,如果没报错,并且显示出如下界面,说明anaconda2安装完毕。
编译caffe时会提示找不到hdf5相关的动态库
默认在usr/lib/x86_64-linux-gnu下找,可以创建anaconda2中的库的软连接,在终端运行如下命令:
sudo ln -s /usr/local/anaconda2/lib/libhdf5_hl.so.10 /usr/lib/x86_64-linux-gnu/libhdf5_hl.so.10sudo ln -s /usr/local/anaconda2/lib/libhdf5.so.10 /usr/lib/x86_64-linux-gnu/libhdf5.so.10
四、安装OpenCV3
OpenCV3,从官网下载opencv3网址,方法看opencv3的installation in linux,在downloads界面的右上角
博主最近发现一个很严重的问题,如果python用的是anaconda的话,按照上述操作编译会有问题,会有没有cv2.so库的情况出现,所以cmake的命令要有所改变,我的命令如下:
cmake -D CMAKE_BUILD_TYPE=RELEASE \ -D INSTALL_C_EXAMPLES=OFF \ -D INSTALL_PYTHON_EXAMPLES=ON \ -D CMAKE_INSTALL_PREFIX=/usr/local \ -D PYTHON_INCLUDE_DIR=/usr/local/anaconda2/include/python2.7/ \ -D PYTHON_INCLUDE_DIR2=/usr/local/anaconda2/include/python2.7/ \ -D PYTHON_LIBRARY=/usr/local/anaconda2/lib/libpython2.7.so \ -D PYTHON_LIBRARY2=/usr/local/anaconda2/lib/libpython2.7.so \ -D PYTHON_PACKAGES_PATH=/usr/local/anaconda2/lib/python2.7/site-packages/ \ -D BUILD_EXAMPLES=ON \ -D BUILD_NEW_PYTHON_SUPPORT=ON \ -D BUILD_opencv_python3=OFF \ -D BUILD_opencv_python2=ON ..
如果你不知道你anaconda装的路径,可以在python环境中执行:
import syssys.prefix
编译出来的cv2.so在文件夹/usr/local/lib/python2.7/site-packages下面,通过使用命令:
sudo ln -s /usr/local/lib/python2.7/site-packages/cv2.so /usr/local/anaconda2/lib/python2.7/site-packages/
在anaconda安装文件夹下建立cv2.so的软连接,
然后在python下执行import cv2,就不会报ImportError: No module named cv2这样的错误了。
五、接下来是编译caffe的过程
官网安装方法,因为官网不是用的anaconda2和opencv3,所以安装方法有点不一样。
下载其他依赖包
sudo apt-get install libatlas-base-dev sudo apt-get install libprotobuf-dev sudo apt-get install libleveldb-dev sudo apt-get install libsnappy-dev sudo apt-get install libboost-all-dev sudo apt-get install libhdf5-serial-dev sudo apt-get install libgflags-dev sudo apt-get install libgoogle-glog-dev sudo apt-get install liblmdb-dev sudo apt-get install protobuf-compiler
下载caffe
github中下载caffe的网址
官网编译方法如下
编译caffe官网网址,下面是根据自己的环境更改caffe文件夹下的makefile.config文件。
根据我现有的环境(linux14.04 + anaconda2 + cuda7.5 + opencv3.0),我的Makefile如下:
## Refer to http://caffe.berkeleyvision.org/installation.html# Contributions simplifying and improving our build system are welcome!# cuDNN acceleration switch (uncomment to build with cuDNN).# USE_CUDNN := 1# CPU-only switch (uncomment to build without GPU support).# CPU_ONLY := 1# uncomment to disable IO dependencies and corresponding data layers# USE_OPENCV := 0# USE_LEVELDB := 0# USE_LMDB := 0# uncomment to allow MDB_NOLOCK when reading LMDB files (only if necessary)# You should not set this flag if you will be reading LMDBs with any# possibility of simultaneous read and write# ALLOW_LMDB_NOLOCK := 1# Uncomment if you're using OpenCV 3OPENCV_VERSION := 3# To customize your choice of compiler, uncomment and set the following.# N.B. the default for Linux is g++ and the default for OSX is clang++# CUSTOM_CXX := g++# CUDA directory contains bin/ and lib/ directories that we need.CUDA_DIR := /usr/local/cuda# On Ubuntu 14.04, if cuda tools are installed via# "sudo apt-get install nvidia-cuda-toolkit" then use this instead:# CUDA_DIR := /usr# CUDA architecture setting: going with all of them.# For CUDA < 6.0, comment the *_50 lines for compatibility.CUDA_ARCH := -gencode arch=compute_20,code=sm_20 \ -gencode arch=compute_20,code=sm_21 \ -gencode arch=compute_30,code=sm_30 \ -gencode arch=compute_35,code=sm_35 \ -gencode arch=compute_50,code=sm_50 \ -gencode arch=compute_50,code=compute_50# BLAS choice:# atlas for ATLAS (default)# mkl for MKL# open for OpenBlasBLAS := atlas# Custom (MKL/ATLAS/OpenBLAS) include and lib directories.# Leave commented to accept the defaults for your choice of BLAS# (which should work)!# BLAS_INCLUDE := /path/to/your/blas# BLAS_LIB := /path/to/your/blas# Homebrew puts openblas in a directory that is not on the standard search path# BLAS_INCLUDE := $(shell brew --prefix openblas)/include# BLAS_LIB := $(shell brew --prefix openblas)/lib# This is required only if you will compile the matlab interface.# MATLAB directory should contain the mex binary in /bin.# MATLAB_DIR := /usr/local# MATLAB_DIR := /Applications/MATLAB_R2012b.app# NOTE: this is required only if you will compile the python interface.# We need to be able to find Python.h and numpy/arrayobject.h.#PYTHON_INCLUDE := /usr/include/python2.7 \ /usr/lib/python2.7/dist-packages/numpy/core/include# Anaconda Python distribution is quite popular. Include path:# Verify anaconda location, sometimes it's in root.ANACONDA_HOME := /usr/local/anaconda2PYTHON_INCLUDE := $(ANACONDA_HOME)/include \ $(ANACONDA_HOME)/include/python2.7 \ $(ANACONDA_HOME)/lib/python2.7/site-packages/numpy/core/include \# Uncomment to use Python 3 (default is Python 2)# PYTHON_LIBRARIES := boost_python3 python3.5m# PYTHON_INCLUDE := /usr/include/python3.5m \# /usr/lib/python3.5/dist-packages/numpy/core/include# We need to be able to find libpythonX.X.so or .dylib.# PYTHON_LIB := /usr/libPYTHON_LIB := $(ANACONDA_HOME)/lib# Homebrew installs numpy in a non standard path (keg only)# PYTHON_INCLUDE += $(dir $(shell python -c 'import numpy.core; print(numpy.core.__file__)'))/include# PYTHON_LIB += $(shell brew --prefix numpy)/lib# Uncomment to support layers written in Python (will link against Python libs)WITH_PYTHON_LAYER := 1# Whatever else you find you need goes here.INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/includeLIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib# If Homebrew is installed at a non standard location (for example your home directory) and you use it for general dependencies# INCLUDE_DIRS += $(shell brew --prefix)/include# LIBRARY_DIRS += $(shell brew --prefix)/lib# Uncomment to use `pkg-config` to specify OpenCV library paths.# (Usually not necessary -- OpenCV libraries are normally installed in one of the above $LIBRARY_DIRS.)# USE_PKG_CONFIG := 1# N.B. both build and distribute dirs are cleared on `make clean`BUILD_DIR := buildDISTRIBUTE_DIR := distribute# Uncomment for debugging. Does not work on OSX due to https://github.com/BVLC/caffe/issues/171# DEBUG := 1# The ID of the GPU that 'make runtest' will use to run unit tests.TEST_GPUID := 0# enable pretty build (comment to see full commands)Q ?= @
接下来,运行下面的语句,编译并检测caffe是否正确编译:
make all #make -j8可以加快编译速度make testmake runtest
出现如下错误,可能是caffe makefile找不到python路径,该配置文件中的anaconda路径不对,检查makefile.config文件中的ANACONDA_HOME := /usr/local/anaconda2语句
如果没有报错,caffe就安装好了,但是此时python还不能import caffe,要使用caffe的python的接口,还需要执行下面语句:
make pycaffe
在终端输入python
进入python运行环境,输入import caffe
,会出现下列错误:
在python环境中,输入下列命令:
import syssys.path.append('/home/hpc/caffe/python') #添加caffe根目录下的python路径,按照自己的路径进行更改import caffe
出现下面界面,说明能成功使用caffe的python接口了。
但是在你添加完caffe的python路径后,导入caffe还是会报错,错误是ImportError: No module named google.protobuf.internal,说明没有按照protobuf,但是之前按照caffe之前是安装过的,caffe也编译完了,还有一个原因就是机器里有两个python环境,一个是系统自带的,另一个就是你anaconda的环境:
pip install protobuf/home/username/anaconda2/bin/pip install protobuf #这是我的anaconda安装路径,路径改成你自己的
最后在python环境中,输入下列命令:
import syssys.path.append('/home/hpc/caffe/python') #添加caffe根目录下的python路径,按照自己的路径进行更改import caffe
应该没有问题了,界面如下
如果是写.py文件,可以在.py文件中添加下列语句:
CAFFE_ROOT = '/home/hpc/caffe'sys.path.append(CAFFE_ROOT+'/python') # Add caffe python path so we can import it
一些指令:
pkg-config --modversion opencv #查看opencv的安装版本protoc --version #查看protobuf安装版本
- Linux14.04+cuda7.5+anaconda2+opencv3.0+caffe
- Linux14.04+cuda7.5+anaconda2+opencv3.0+caffe+Matlab 艰辛之路
- ubuntu14.04 caffe cuda7.5 cudnn anaconda2 opencv3 安装
- linux14.04+cuda7.5
- 完整配置LINUX14.04+CUDA7.5+cudnn+opencv+caffe+Faster-RCNN,配置CUDA7.5
- Ubuntu14.04+CUDA7.5+opencv3.0+MKL+Caffe配置
- ubuntu14.04+cuda7.5+opencv3.0+cudnn+caffe
- Ubuntu14.04+CUDA7.0+Anaconda2+Cudnn+caffe安装配置
- Ubuntu14.04+CUDA7.5+Anaconda2-4.0.0+Matlab2014b+Caffe from Scratch
- GTX1080+Ubuntu16.04+Cuda8.0+Anaconda2+Opencv3+Matlab2014a+caffe
- Ubuntu14.04+CUDA8.0+Opencv3.1+Anaconda2+Caffe安装
- Caffe搭建:Ubuntu14.04 + CUDA7.5 + opencv3.1+python3.5
- Ubuntu14.04+CUDA7.0+Opencv3.0+Caffe 配置
- ubuntu15.04+caffe+cuda7.5+cudnnv4+mkl+opencv3
- caffe+cuda7.0+opencv3.0.0+mkl ubuntu14.04配置
- caffe+cuda7.0+opencv3.0.0+mkl ubuntu14.04配置
- VS2013+CUDA7.5(+cudnn7.0)+Opencv3.0.0+caffe
- ubuntu14.04安装caffe+cuda7.5+cudnn5.0+opencv3.0(亲测有效)
- 【JZOJ3861】【JSOI2014】支线剧情2
- Binary search (3) -- 4Sum II, Kth Smallest Element in a BST
- 修改Tomcat 允许启动(timeout)时间(45秒超时)
- 学习日志-JavaScript的基本概念
- Aop面向切面编程
- Linux14.04+cuda7.5+anaconda2+opencv3.0+caffe
- 数组之链表操作(用数组超时)
- APNs(Apple Push Notification server)服务器代码C#编写
- Ubuntu 14.04 LTS + Qt 4.7.3 交叉编译
- 【JSOI2014】支线剧情2
- BZOJ 1006 神奇的国度 弦图 最大势算法
- Software developer interview questions: The soft part
- [51nod 1222]最小公倍数计数
- multimap容器案例_按照部门_增删改查员工信息