This is the LIBASKIT set of scalable machine learning and data analysis tools. Currently we provide codes for kernel sums, nearest-neighbors, kmeans clustering, kernel regression, and multiclass kernel logistic regression. All codes use OpenMP and MPI for shared memory and distributed memory parallelism.
- ASKIT : (Approximate Skeletonization Kernel Independent Treecode) Code for fast approximate kernel summation. It finds applications in kernel machines. It supports treecode and fast-multipole versions.
- RKDT : (Randomized KD-trees) Set of core algorithms for nearest-neighbor searches
- GSKS : (General Stride Kernel summation) X86 optimized libraries for direct, O(N^2), kernel summation.
- GSKNN : (General Stride K-nearest-neighbors) X86 optimized libraries for direct, O(N^2) nearest-neighbor searches.
- PNYSTR : (Parallel Nystrom method) Code for kernel summation using the Nystrom method.
All software is available under the the GNU General Public License
GPL license
Downloads:
- ASKIT : askit.tar.gz (requires RKDT and GSKS)
- RKDT : rkdt.tar.gz
- GSKS : GSKS on github
- GSKNN: GSKNN on github
- PNYSTR: in preparation