NN-apps is an implementation of near neighbor searching using LSH (locality sensitive hashtables) with additional applications in clustering. NN-apps accept input in the form of vectors. Dimensionality reduction is performed in order to reduce the complexity of the dataset. Vector proximity can be calculated in the form of Euclidean or Frechet distance. The clustering is performed using Lloyd's, reverse LSH as well as reverse hypercube algorithms.
pantach/NN-apps
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|