cd $CAFFE_ROOT
./data/cifar10/get_cifar10.sh
./examples/cifar10/create_cifar10.sh
./examples/cifar10/train_quick.sh
cd $CAFFE_ROOT
./data/mnist/get_mnist.sh
./examples/mnist/create_mnist.sh
./examples/mnist/train_lenet.sh
Check failed: registry.count(type) == 1 (0 vs. 1) Unknown solver type: SGD (known types: )
重新编译即可。
make clean
make all -j8
CUDA
boost > 1.5
sudo apt-get install libboost-all-dev
/sbin/ldconfig.real: /usr/local/cuda/lib64/libcudnn.so.7 is not a symbolic link
sudo ln -s libcudnn.so.7.1.3 libcudnn.so.7
sudo ln -s libcudnn.so.7 libcudnn.so
sudo apt-get install libatlas-base-dev
sudo apt-get install libopenblas-dev
如果不安装依赖包，会出现各种编译问题。
sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev python-dev libgflags-dev libatlas-base-dev libhdf5-serial-dev protobuf-compiler libgoogle-glog-dev liblmdb-dev
fatal error: opencv2/core/core.hpp caffe framework installation
但是通过 sudo apt-get install libopencv-dev 安装的是 opencv 3 ，可以通过 dpkg -l | grep libopencv 得出来， 因此需要在 Makefile.config 中取消 OPENCV_VERSION=3 的注释。参见caffe fails with error undefined reference to cv::imread(cv::String const&, int)' and undefined reference to cv::imdecode(cv::_InputArray const&, int)
git clone https://github.com/BVLC/caffe.git
cd caffe
cp Makefile.config.example Makefile.config
将 Makefile.config 中以下条目修改。
USE_CUDNN := 1
WITH_PYTHON_LAYER := 1
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
For CUDA >= 9.0, comment the *_20 and *_21 lines for compatibility.
make all -j8
make test -j8
make runtest -j8