GSKNN (General Stride K-Nearest Neighbor) README ---

Thank you for deciding to give GSKNN a try! This code can be used for exact K-Nearest Neighbor search.

See INSTALL on how to install it.

Nearest Neighbor search appears in machine learning, manifold learning and other scientific computing problems.

GSKNN (General Stride K-Nearest Neighbor) is a portable framework that provides a high performance memory efficient nearest neighbors search based on the BLIS framework.

GSKNN has several features. For further details of this project, please check the Githup repo:

https://github.com/ChenhanYu/rnn

To reuse the material of, please cite GSKNN in the following form:

@article{

yu2015performance, title={Performance Optimization for the K Nearest-Neighbor Kernel on x86 Architectures}, author={Yu, Chenhan D. and Huang, Jianyu and Austin, Woody and Xiao, Bo and Biros, George}, year={2015}

}

Thank you again for being intersted in GSKNN!

Best regards,

Chenhan D. Yu chenhan@cs.utexas.edu