nptsne 1.0.0

The nptsne package is designed to export a number of python classes that wrap tSNE.

nptsne - A numpy compatible python extension for GPGPU linear complexity tSNE

The nptsne package is designed to export a number of python classes that wrap GPGPU linear complexity tSNE implementations based on the following publications DOI: 10.1109/TVCG.2019.2934307 or the arXiv preprint

When using nptsne please include the following citation:

N. Pezzotti et al., "GPGPU Linear Complexity t-SNE Optimization," in IEEE Transactions on Visualization and Computer Graphics.
doi: 10.1109/TVCG.2019.2934307
keywords: {Minimization;Linear programming;Computational modeling;Approximation algorithms;Complexity theory;Optimization;Data visualization;High Dimensional Data;Dimensionality Reduction;Progressive Visual Analytics;Approximate Computation;GPGPU},
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8811606&isnumber=4359476

Attributions

The tSNE implementations are the original work of the authors named in the literature.

Full documentation

Full documentation is available at the nptsne doc pages

Wrapper classes

Class	Description	Doc link
TextureTsne	Linear time tSNE reliant on GPU textures	https://biovault.github.io/nptsne/nptsne.html#nptsne.TextureTsne
TextureTsneExtended	Linear time tSNE reliant on GPU textures, extended API	https://biovault.github.io/nptsne/nptsne.html#nptsne.TextureTsneExtended

Linux support

Linux: (only Ubuntu 16.06 and upward is supported). Download the correct file (see below) for your python version and install using pip install <file>.whl

py36	py37
Ubuntu py36 wheel	Ubuntu py37 wheel