- Linux installation----Download ubuntu14.04 installation is possible.
- GCC downgrade
sudo apt-get install-y gcc-4.7//download GCC 4.7sudo apt-get install-y g++-4.7//download g++4.7Cd/use/binsudo rm gcc//Remove GCCsudo ln-s gcc-4.7 gcc//linksudo rm g++sudo ln-s g++-4.7 g++sudo ldconfig
- Git installation and Caffe code download
sudo apt-get install build-essential cmake git pkg-config
sudo git clone https://github.com/BVLC/caffe.git
1 CUDA Repository
To obtain CUDA installation package, please go to NVIDIA official website to download the installation package yourself. (https://developer.nvidia.com/cuda-downloads)
$ sudo dpkg-i cuda-repo-ubuntu1504-7-5-local_7.5-18_amd64
$ sudo apt-get update
2 CUDA Toolkit
$ sudo apt-get install-y cuda
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
BLAS
sudo apt-get install libatlas-base-dev
Everything is packaged in 14.04
sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev
Google Logging Library(glog)
:https://code.google.com/p/google-glog/,然后解压安装:
Tar zxvf glog-0.3.3.tar.gz
./configure
Make
$ sudo make install
Installing CUDNN
Go to the official website to download CUDNN, after the download, unzip, get a Cuda folder, and then enter the following statement: CD Cuda/includesudo CP *.H/USR/LOCAL/INCLUDE/CD. /lib64sudo CP lib*/usr/local/lib/cd/usr/local/libsudo chmod +r libcudnn.so.4.0.4sudo ln-sf libcudnn.so.4.0.4 libcudnn. So.4sudo ln-sf libcudnn.so.4 Libcudnn.sosudo ldconfig
Make uses
cp Makefile.config.example Makefile.config
make all
make test
make runtest
CMake use
mkdir buildcd buildcmake ..make allmake installmake runtest
ubuntu14.04 installation Caffe+matlab