NOT MAINTAINED

• I have not used this project myself on a long time
• No issues or PRs can be created
• At the time of writting this April 6, 2023 - The lastest release on PyPI (0.3.1) works as intented
• I recommend moving to a native solution to other frameworks such as Tensorflow or PyTorch

Python-TSNE

Python library containing T-SNE algorithms.

Note: Scikit-learn v0.17 includes TSNE algorithms and you should probably be using that instead.

Installation

Requirements

• cblas or openblas. Tested version is v0.2.5 and v0.2.6 (not necessary for OSX).

From PyPI:

pip install tsne

From conda:

conda install -c maxibor tsne

Usage

Basic usage:

from tsne import bh_sne
X_2d = bh_sne(X)

Examples

• Iris
• MNIST
• word2vec on presidential speeches via @prateekpg2455

Algorithms

Barnes-Hut-SNE

A python (cython) wrapper for Barnes-Hut-SNE aka fast-tsne.

I basically took osdf's code and made it pip compliant.

Additional resources

• See Barnes-Hut-SNE (2013), L.J.P. van der Maaten. It is available on arxiv.