Windows7 Ubuntu16.04 双硬盘双系统搭建Caffe

最新推荐文章于 2022-10-22 21:45:56 发布

Digital2Slave

最新推荐文章于 2022-10-22 21:45:56 发布

阅读量2.2k

点赞数 2

CC 4.0 BY-SA版权

分类专栏： MachineLearning 文章标签： windows ubuntu caffe

本文链接：https://blog.youkuaiyun.com/tianzhaixing2013/article/details/54021552

MachineLearning 专栏收录该内容

21 篇文章

订阅专栏

本文详细记录了Windows7与Ubuntu16.04双硬盘双系统的安装过程，并分享了解决Windows7无法引导的问题。此外，还介绍了如何在Ubuntu环境下安装配置CUDA、cuDNN以及Caffe，包括安装显卡驱动、编译OpenCV等关键步骤。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

更好的阅读体验，参考我公布的Windows && Ubuntu 双硬盘双系统。

Windows && Ubuntu 双硬盘双系统

主要参考SDD+HDD+Win7+Ubuntu12.04双硬盘双系统安装流程进行操作。
我的一些硬件环境：

120GB Intel SSD 和 250GB Samsung SSD
显卡是英伟达的GeForce GTX 750 Ti
内存8G
处理器Intel(R) Core(TM) i5-6500 CPU @ 3.20GHz

安装Windows系统

Windows7 旗舰版64位系统，来自系统之家下载;
利用UltraISO制作U盘启动盘，进行安装(具体教程请自行搜索完成)。

安装Ubuntu系统

分区设置
/boot 4GB (Windows7 系统盘压缩出来的4GB空间) –> Intel 120GB SSD
/ 104GB (Ubuntu16.04 系统的根分区) –> Samsung 250GB SSD 中压缩出120GB空间的104GB
swap 16GB (Ubuntu16.04 系统交换分区) –> Samsung 250GB SSD 中压缩出120GB空间的16GB

8G内存，设置swap大小为16GB。
无法引导进入Windows7
解决方法：
修改/boot/grub/grub.cfg

将

### BEGIN /etc/grub.d/30_os-prober ###
menuentry "Windows 7 (loader) (on /dev/sda1)" --class windows --class os {
insmod part_msdos
insmod ntfs
set root='(hd0,msdos1)'
search --no-floppy --fs-uuid --set=root AAAA9A0DAA99D663
chainloader +1
}
### END /etc/grub.d/30_os-prober ###

修改为：

### BEGIN /etc/grub.d/30_os-prober ###
menuentry "Windows 7 (loader) (on /dev/sda1)" --class windows --class os {
insmod part_msdos
insmod ntfs
set root='(hd0,msdos1)'
search --no-floppy --fs-uuid --set=root AAAA9A0DAA99D663
ntldr ($root)/bootmgr
}
### END /etc/grub.d/30_os-prober ###

Caffe

安装显卡驱动

参考显卡信息

tzx@jt:~$ lspci | grep -i vga
01:00.0 VGA compatible controller: NVIDIA Corporation GM107 [GeForce GTX 750 Ti] (rev a2)
tzx@jt:~$

参考这里安装英伟达显卡驱动
利用nvidia-smi检验英伟达显卡驱动是否安装成功

tzx@jt:~$ nvidia-smi 
Wed Jan  4 15:52:48 2017       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 367.57                 Driver Version: 367.57                    |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce GTX 750 Ti  Off  | 0000:01:00.0      On |                  N/A |
| 33%   35C    P0     1W /  38W |    244MiB /  1998MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID  Type  Process name                               Usage      |
|=============================================================================|
|    0      1022    G   /usr/lib/xorg/Xorg                             122MiB |
|    0      1788    G   compiz                                         120MiB |
+-----------------------------------------------------------------------------+

安装CUDA

官网下载CUDA
下载时，我的选择如下所示：
Operating System Linux
Architecture x86_64 (这个可能不同)
Distribution Ubuntu
Version 16.04
Installer Type deb(local)

$ sudo dpkg -i cuda-repo-ubuntu1604-8-0-local_8.0.44-1_amd64.deb
$ sudo apt-get update
$ sudo apt-get install cuda

测试是否安装成功

$ cd /usr/local/cuda-8.0/samples/1_Utilities/deviceQuery
$ make
$ sudo ./deviceQuery
[sudo] password for tzx: 
./deviceQuery Starting...

 CUDA Device Query (Runtime API) version (CUDART static linking)

Detected 1 CUDA Capable device(s)

Device 0: "GeForce GTX 750 Ti"
  CUDA Driver Version / Runtime Version          8.0 / 8.0
  CUDA Capability Major/Minor version number:    5.0
  Total amount of global memory:                 1999 MBytes (2095841280 bytes)
  ( 5) Multiprocessors, (128) CUDA Cores/MP:     640 CUDA Cores
  GPU Max Clock rate:                            1084 MHz (1.08 GHz)
  Memory Clock rate:                             2700 Mhz
  Memory Bus Width:                              128-bit
  L2 Cache Size:                                 2097152 bytes
  Maximum Texture Dimension Size (x,y,z)         1D=(65536), 2D=(65536, 65536), 3D=(4096, 4096, 4096)
  Maximum Layered 1D Texture Size, (num) layers  1D=(16384), 2048 layers
  Maximum Layered 2D Texture Size, (num) layers  2D=(16384, 16384), 2048 layers
  Total amount of constant memory:               65536 bytes
  Total amount of shared memory per block:       49152 bytes
  Total number of registers available per block: 65536
  Warp size:                                     32
  Maximum number of threads per multiprocessor:  2048
  Maximum number of threads per block:           1024
  Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
  Max dimension size of a grid size    (x,y,z): (2147483647, 65535, 65535)
  Maximum memory pitch:                          2147483647 bytes
  Texture alignment:                             512 bytes
  Concurrent copy and kernel execution:          Yes with 1 copy engine(s)
  Run time limit on kernels:                     Yes
  Integrated GPU sharing Host Memory:            No
  Support host page-locked memory mapping:       Yes
  Alignment requirement for Surfaces:            Yes
  Device has ECC support:                        Disabled
  Device supports Unified Addressing (UVA):      Yes
  Device PCI Domain ID / Bus ID / location ID:   0 / 1 / 0
  Compute Mode:
     < Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 8.0, CUDA Runtime Version = 8.0, NumDevs = 1, Device0 = GeForce GTX 750 Ti
Result = PASS

安装cuDNN

官方下载cuDNN，需要注册账号才能下载。
我下载的是cudnn-8.0-linux-x64-v5.0-ga.tgz！
然后，执行以下命令：

$ cd ~/Downloads
$ sudo tar -zxvf ./cudnn-8.0-linux-x64-v5.0-ga.tgz
$ cd cuda/include
$ sudo cp cudnn.h /usr/local/cuda/include     # 复制头文件
$ cd ../lib64
$ sudo cp lib* /usr/local/cuda/lib64/         # 复制动态链接库
$ cd /usr/local/cuda/lib64/
$ sudo rm -rf libcudnn.so libcudnn.so.5       # 删除原有动态文件
$ sudo ln -s libcudnn.so.5.0.5 libcudnn.so.5  # 生成软衔接
$ sudo ln -s libcudnn.so.5 libcudnn.so        # 生成软链接

安装OpenCV3.2.0

参考我之前写的博客Ubuntu16.04 编译OpenCV 和 Tesseract-OCR。

opencv-3.2.0/3rdparty/ippicv/downloads/linux-808b791a6eac9ed78d32a7666804320e目录下的ippicv_linux_20151201.tgz 和
opencv_contrib-3.2.0/modules/dnn/.download/bd5e3eed635a8d32e2b99658633815ef/v3.1.0目录下的protobuf-cpp-3.1.0.tar.gz两个文件，可能因为网络下载问题，无法下载成功。自己下载后替换到相应的位置。

下载ippicv_linux_20151201.tgz

下载protobuf-cpp-3.1.0.tar.gz

安装Caffe

安装相关依赖库

tzx@jt:~$ sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libhdf5-serial-dev protobuf-compiler
tzx@jt:~$ sudo apt-get install --no-install-recommends libboost-all-dev
tzx@jt:~$ sudo apt-get install libopenblas-dev liblapack-dev libatlas-base-dev
tzx@jt:~$ sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev

获取Caffe代码并修改配置

获取Caffe代码

$ sudo apt-get update
$ sudo apt-get install git
$ mkdir Github
$ cd Github
$ git clone https://github.com/BVLC/caffe.git

修改Makefile.config

$ cd ~/Github/caffe
$ sudo cp Makefile.config.example Makefile.config
$ sudo gedit Makefile.config #打开Makefile.config文件

若使用cuDNN，则将#USE_CUDNN := 1修改成：USE_CUDNN := 1
若使用的OpenCV版本是3的，则将#OPENCV_VERSION := 3修改为：OPENCV_VERSION := 3
若要使用python来编写layer，则将#WITH_PYTHON_LAYER := 1修改为 WITH_PYTHON_LAYER := 1

重要的一项: 将 # Whatever else you find you need goes here. 下面的

INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include
LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib

修改为：

INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include /usr/include/hdf5/serial
LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib /usr/lib/x86_64-linux-gnu /usr/lib/x86_64-linux-gnu/hdf5/serial

修改Makefile

$ sudo gedit Makefile

将

NVCCFLAGS +=-ccbin=$(CXX) -Xcompiler-fPIC $(COMMON_FLAGS)

替换为：

NVCCFLAGS += -D_FORCE_INLINES -ccbin=$(CXX) -Xcompiler -fPIC $(COMMON_FLAGS)

将

LIBRARIES += glog gflags protobuf boost_system boost_filesystem m hdf5_hl hdf5

替换为：

LIBRARIES += glog gflags protobuf boost_system boost_filesystem m hdf5_serial_hl hdf5_serial

注意自己CUDA的版本号！

$ sudo cp /usr/local/cuda-8.0/lib64/libcudart.so.8.0 /usr/local/lib/libcudart.so.8.0 && sudo ldconfig
$ sudo cp /usr/local/cuda-8.0/lib64/libcublas.so.8.0 /usr/local/lib/libcublas.so.8.0 && sudo ldconfig
$ sudo cp /usr/local/cuda-8.0/lib64/libcurand.so.8.0 /usr/local/lib/libcurand.so.8.0 && sudo ldconfig
$ sudo cp /usr/local/cuda-8.0/lib64/libcudnn.so.5 /usr/local/lib/libcudnn.so.5 && sudo ldconfig

编译Caffe

$ cd ~/Github/caffe
$ make all -j4 #-j根据自己电脑配置决定
$ make test
$ make runtest
$ make distribute

pycaffe

$ cd ~/Github/caffe/python
$ for req in $(cat requirements.txt); do pip install $req -i https://pypi.douban.com/simple; done
$ cd ~/Github/caffe
$ make pycaffe
$ cd ~/Github/caffe/python
$ pwd
/home/tzx/Github/caffe/python # 替换tzx为你自己的用户名
$ sudo gedit ~/.bashrc

在.bashrc最后，添加以下代码

export PYTHONPATH=/home/tzx/Github/caffe/python:$PYTHONPATH  # 替换tzx为你自己的用户名

然后，让修改立即生效$ source ~/.bashrc。

验证pycaffe

tzx@jt:~$ python
Python 2.7.12 (default, Nov 19 2016, 06:48:10) 
[GCC 5.4.0 20160609] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import caffe
>>> caffe.__version__
'1.0.0-rc3'
>>>

运行MNIST测试

$ cd ~/Github/caffe
$ ./data/mnist/get_mnist.sh        # 下载MNIST数据库并解压缩
$ ./examples/mnist/create_mnist.sh # 将其转换成Lmdb数据库格式
$ ./examples/mnist/train_lenet.sh  # 训练网络

最后的训练结果

I0104 16:56:03.183193 28155 solver.cpp:317] Iteration 10000, loss = 0.00346665
I0104 16:56:03.183210 28155 solver.cpp:337] Iteration 10000, Testing net (#0)
I0104 16:56:03.324321 28155 solver.cpp:404]     Test net output #0: accuracy = 0.9908
I0104 16:56:03.324347 28155 solver.cpp:404]     Test net output #1: loss = 0.0303717 (* 1 = 0.0303717 loss)
I0104 16:56:03.324350 28155 solver.cpp:322] Optimization Done.
I0104 16:56:03.324353 28155 caffe.cpp:254] Optimization Done.