ubuntu下安裝caffe筆記

General dependencies

sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libhdf5-serial-dev protobuf-compiler
sudo apt-get install --no-install-recommends libboost-all-dev

CUDA: Install by apt-get or the NVIDIA .run package. The NVIDIA package tends to follow more recent library and driver versions, but the installation is more manual. If installing from packages, install the library and latest driver separately; the driver bundled with the library is usually out-of-date. This can be skipped for CPU-only installation.

BLAS: install ATLAS by sudo apt-get install libatlas-base-dev or install OpenBLAS by sudo apt-get install libopenblas-dev or MKL for better CPU performance.

Python (optional): if you use the default Python you will need to sudo apt-get install the python-dev package to have the Python headers for building the pycaffe interface.

 

先按照官網的步驟把環境安裝一下(我這裏安裝僅供CPU使用，跳過CUDA)：

具體命令以下：

　　　

sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libhdf5-serial-dev protobuf-compiler
sudo apt-get install --no-install-recommends libboost-all-dev
sudo apt-get install libatlas-base-dev

進入源碼的python目錄下：

for req in $(cat requirements.txt); do pip install $req; done

 

Compilation with Make

Configure the build by copying and modifying the example Makefile.config for your setup. The defaults should work, but uncomment the relevant lines if using Anaconda Python.

cp Makefile.config.example Makefile.config
# Adjust Makefile.config (for example, if using Anaconda Python, or if cuDNN is desired)
make all
make test
make runtest

• For CPU & GPU accelerated Caffe, no changes are needed.
• For cuDNN acceleration using NVIDIA’s proprietary cuDNN software, uncomment the USE_CUDNN := 1 switch in Makefile.config. cuDNN is sometimes but not always faster than Caffe’s GPU acceleration.
• For CPU-only Caffe, uncomment CPU_ONLY := 1 in Makefile.config.

To compile the Python and MATLAB wrappers do make pycaffe and make matcaffe respectively. Be sure to set your MATLAB and Python paths in Makefile.config first!

Distribution: run make distribute to create a distribute directory with all the Caffe headers, compiled libraries, binaries, etc. needed for distribution to other machines.

Speed: for a faster build, compile in parallel by doing make all -j8 where 8 is the number of parallel threads for compilation (a good choice for the number of threads is the number of cores in your machine).

Now that you have installed Caffe, check out the MNIST tutorial and the reference ImageNet model tutorial.

 

我這裏使用make編譯：

按如上步驟編譯主要有兩個問題：

　　

1.fatal error: hdf5.h: No such file or directory
2./usr/bin/ld: cannot find -lhdf5_hl 　
　/usr/bin/ld: cannot find -lhdf5


主要是Makefile.config中的INCLUDE_DIRS，LIBRARY_DIRS兩個變量找不到以前安裝的libhdf5-serial-dev,把相應的路徑添加進去就能夠啦
添加以後是這樣的：

　　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/hdf5/serial

 

而後再按照官網的流程就能夠啦

 

參考帖子：　http://blog.csdn.net/lkj345/article/details/51280369

 

順便裝下pycaffe

在源碼根目錄下運行以下命令：

　　make pycaffe

錯誤以下：

‘numpy/arrayobject.h' file not found

找不到numpy

在python shell 運行以下命令查看numpy的目錄：　　

>>> import numpy as np
>>> np.get_include()

將此路徑追加在以前的Makefile.config文件中PYTHON_INCLUDE變量後面，再次make pycaffe就編譯ok

而後就是將caffe/python目錄添加到環境變量中的問題了

～/.bashrc添加以下代碼於最後一行：

　　export PYTHONPATH={your path}:$PYTHONPATH

再source ~/.bashrc 一下，至此就ok了