SWCaffe is a deep learning framework on SW26010 based by Caffe with MPI supported. Compiling method is the same with Caffe. This repo is maintained by National Supercomputer Center in Wuxi. More details about optimization tricks to convolutional layers please see our Technical Report and our IPDPS-17's paper swDNN: A Library for Accelerating Deep Learning Applications on Sunway TaihuLight. We also have a conference paper on swCaffe in submission.
Jiarui Fang (fang_jiarui@163.com), Liandeng Li(liandeng_li@163.com)
Caffe is released under the BSD 2-Clause license. The BAIR/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}
}