Shulou(Shulou.com)06/01 Report--

This article mainly introduces how to install the convolution neural network framework caffe under Linux related knowledge, the content is detailed and easy to understand, the operation is simple and fast, has a certain reference value, I believe that everyone read this Linux how to install the convolution neural network framework caffe article will have a harvest, let's take a look.

Caffe is a clear and efficient deep learning framework, is a pure C++/CUDA architecture, supports the command line, python and matlab interfaces; can switch seamlessly between CPU and GPU, by Jia Yangqing, caffe2 has been released.

Download caffe and save it to a directory (recommended to / home directory)

Install dependencies: sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libhdf5-serial-dev protobuf-compiler (this string is a command) sudo apt-get install--no-install-recommends libboost-all-devsudo apt-get install libatlas-base-devsudo apt-get install python-devsudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-devsudo apt-get install python-lmdbsudo apt-get install python-pip1234567cd caffe, and modify the configuration file cp Makefile.config.example Makefile.config (if you run with CPU Comment out the cuda correlation of these two files) 1 modify the options to be compiled (for ubuntu15.04 and later versions, the previous version does not need to be modified Used to solve the problem that files such as hdf5.h cannot be found) INCLUDE_DIRS: = $(PYTHON_INCLUDE) / usr/local/include\ / usr/lib/x86_64-linux-gnu/hdf5/serial/includeLIBRARY_DIRS: = $(PYTHON_LIB) / usr/local/lib / usr/lib\ / usr/lib/x86_64-linux-gnu/hdf5/serial123456

If there is an error in step 4 make runtest, error while loading shared libraries: libcudart.so.8.0: cannot open shared object file: cannot open (the file libcudart.so.8.0 cannot be found in the / usr/local/lib/ directory, it can be found in the usr/local/cuda-8.0/lib64/ directory), and the solution:

Sudo cp / usr/local/cuda-8.0/lib64/libcudart.so.8.0 / usr/local/lib/libcudart.so.8.0 & & sudo ldconfigsudo cp / usr/local/cuda-8.0/lib64/libcublas.so.8.0 / usr/local/lib/libcublas.so.8.0 & & sudo ldconfigsudo cp / usr/local/cuda-8.0/lib64/libcurand.so.8.0 / usr/local/lib/libcurand.so.8.0 & & sudo ldconfigsudo cp / usr/local/cuda-8.0/lib64/libcudnn.so.6 / usr/local/lib/libcudnn.so.6 & & sudo ldconfig1234

Or can you add this path after LIBRARY_DIRS? Usr/local/cuda-8.0/lib64/ (not tried, should be OK)

Reference link: https://blog.csdn.net/qq_38451119/article/details/81126692

Remove the comments before the CPU-only option to make it effective. If you need Python layer, you also need to remove the comments before WITH_PYTHON_LAYER: = 1. If you need to compile the MATLAB interface, you need to modify MATLAB_DIR to change it to the MATLAB path under the current system.

Compile caffe:make all-j4make test-j4make runtest123

Compile the python interface

Sudo pip install-r python/requirements.txtmake pycaffe12

Compile the matlab interface

Make matcaffegedit / .bashrc12

Add at the end: export PYTHONPATH=/home/francis/caffe/python

Rm * ~ you can consider deleting backup files automatically generated due to modification. Cpu_only can be configured successfully.

To install the GPU version, CUDA should be installed first. If the conditions permit, you can join cudnn to accelerate the convolution operation. The installation steps of CUDA will be improved and CPU-only will be commented out. The rest is the same as before.

This is the end of the article on "how to install convolution neural network framework caffe under Linux". Thank you for reading! I believe that everyone has a certain understanding of "how to install convolution neural network framework caffe under Linux". If you want to learn more, you are welcome to follow the industry information channel.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.