Caffe branch for R-FCN
C++ Python Cuda CMake Protocol Buffer Matlab Other
Permalink
Failed to load latest commit information.
cmake Add support for matcaffe build in Windows Mar 7, 2016
data [data] get_mnist.sh rewrite; prevents prompt in tutorial notebooks Feb 24, 2016
docker Pin the base image version for the GPU Dockerfile May 2, 2016
docs Typo in docs/installation.md Apr 18, 2016
examples Merge master@{2016-05-09} into windows May 9, 2016
include/caffe R-FCN change sets Jun 17, 2016
matlab R-FCN change sets Jun 17, 2016
models [examples] switch examples + models to Input layers Feb 26, 2016
python Merge master@{2016-05-09} into windows May 9, 2016
scripts Read the data as a binary Apr 24, 2016
src Update: fix a bug in scale_layer.cpp Sep 30, 2016
tools Merge master@{2016-05-09} into windows May 9, 2016
windows R-FCN change sets Jun 17, 2016
.Doxyfile update doxygen config to stop warnings Sep 3, 2014
.gitattributes Add support for windows build Feb 11, 2016
.gitignore R-FCN change sets Jun 17, 2016
.travis.yml Separate IO dependencies Sep 17, 2015
CMakeLists.txt Merge pull request #3451 from intelcaffe/cmake-clang-fix Feb 11, 2016
CONTRIBUTING.md [docs] add CONTRIBUTING.md which will appear on GitHub new Issue/PR p… Jul 30, 2015
CONTRIBUTORS.md clarify the license and copyright terms of the project Aug 7, 2014
INSTALL.md installation questions -> caffe-users Oct 19, 2015
LICENSE copyright 2015 Jun 23, 2015
Makefile Fix bug Jun 19, 2016
Makefile.config.example [build] note that `make clean` clears build and distribute dirs May 4, 2016
README.md fix indent and update download link Sep 9, 2016
appveyor.yml Remove optimization in Appveyor build Mar 28, 2016
caffe.cloc [fix] stop cloc complaint about cu type Sep 4, 2014

README.md

Caffe branch for R-FCN

This is a branch of Caffe supporting R-FCN, which has been tested under Windows (Windows 7, 8, Server 2012 R2) and Linux (Ubuntu 14.04).

Linux Setup

Pre-Build Steps

Copy Makefile.config.example to Makefile.config

We need to modify Makefile.config to specify some software PATHS, you may view my Makefile.config for reference.

CUDA

Download CUDA Toolkit 7.5 from nVidia website.

Matlab

Uncomment MATLAB_DIR and set MATLAB_DIR accordingly to build Caffe Matlab wrapper. Matlab 2014a and later versions are supported.

cuDNN (optional)

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.

Download cuDNN v3 or cuDNN v4 from nVidia website. And unpack downloaded zip to $CUDA_PATH (It typically would be /usr/local/cuda/include and /usr/local/cuda/lib64).

Build

Simply type

make -j8 && make matcaffe

Windows Setup

Requirements: Visual Studio 2013

Pre-Build Steps

Copy .\windows\CommonSettings.props.example to .\windows\CommonSettings.props

3rd party dependencies required by Caffe are automatically resolved via NuGet.

CUDA

Download CUDA Toolkit 7.5 from nVidia website.

Matlab

Set MatlabSupport to true and MatlabDir to the root of your Matlab installation in .\windows\CommonSettings.props to build Caffe Matlab wrapper. Matlab 2014a and later versions are supported.

cuDNN (optional)

Download cuDNN v3 or cuDNN v4 from nVidia website. Unpack downloaded zip to %CUDA_PATH% (environment variable set by CUDA installer). Alternatively, you can unpack zip to any location and set CuDnnPath to point to this location in .\windows\CommonSettings.props. CuDnnPath defined in .\windows\CommonSettings.props. By default, cuDNN is not enabled. You can enable cuDNN by setting UseCuDNN to true in the property file.

Build

Now, you should be able to build .\windows\Caffe.sln

Remark

After you have built solution with Matlab support, copy all files in .\Build\x64\Release to R-FCN\external\caffe\matlab\caffe_rfcn.

Further Details

Refer to the BVLC/caffe master branch README for all other details such as license, citation, and so on.