pavlin-policar/openTSNE

Extensible, parallel implementations of t-SNE

67
/ 100
Established

Implements both FIt-SNE (fast interpolation-based) and Barnes-Hut tree algorithms with C/C++ kernels accelerated by OpenMP and optional FFTW3 for FFT computation. Supports incremental embedding of new points into existing reference embeddings and multiscale kernel tricks for improved global structure preservation. Integrates with scikit-learn's data loading ecosystem and targets single-cell transcriptomics and large-scale visualization workflows.

1,617 stars and 68,472 monthly downloads. Used by 3 other packages. Available on PyPI.

Maintenance 6 / 25
Adoption 23 / 25
Maturity 18 / 25
Community 20 / 25

How are scores calculated?

Stars

1,617

Forks

174

Language

Python

License

BSD-3-Clause

Last pushed

Nov 13, 2025

Monthly downloads

68,472

Commits (30d)

0

Dependencies

3

Reverse dependents

3

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/pavlin-policar/openTSNE"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.