Here is the linux/windows compatible version of caffe forked from https://github.com/BVLC/caffe in 04/10/2016. Multi-GPU is supported in this version.

I also have a talk on brief introduction of Deep Learning,  part1,  part2,  slides.

Besides, I shared some practical tricks for deep learning(quite general),   DeepLearnigTricks.

Windows

Tools:

1. Visual Studio 2013

2. Cuda 7.5 ( you should install cuda after the installation of Visual Studio 2013 to incorporate cuda vs integration into VS)

3. OpenCV 2.4.9

4. Boost

Steps:

1. Copy folder 3rdparty (http://pan.baidu.com/s/1ge3nKRp) and bin (http://pan.baidu.com/s/1jIyEjKq) to the caffe root directory

2. Configure the environment variables: BOOST_1_56_0 (e.g. C:\local\boost_1_56_0), OPENCV_2_4_9 (e.g. D:\apps\opencv\build)

3. Compile the caffe.sln in build-windows by VS2013

Notes:

1. Currently Caffe works with cuDNN_v3 or cuDNN_v4 (The current settings in caffe.sln do not use cuDNN)

You need copy More details at https://github.com/BVLC/caffe/tree/windows

Linux

Please follow the official tutorials here: http://caffe.berkeleyvision.org/installation.html

License and Citation

The official turtorial is here: caffe tutorial

Caffe is released under the BSD 2-Clause license. The BVLC reference models are released for unrestricted use.

Please cite Caffe in your publications if it helps your research:

@article{jia2014caffe,
  Author = {Jia, Yangqing and Shelhamer, Evan and Donahue, Jeff and Karayev, Sergey and Long, Jonathan and Girshick, Ross and Guadarrama, Sergio and Darrell, Trevor},
  Journal = {arXiv preprint arXiv:1408.5093},
  Title = {Caffe: Convolutional Architecture for Fast Feature Embedding},
  Year = {2014}
}