Skip to main content

Implementation of the gradient-based t-SNE attribution method described in our GLBIO oral presentation: 'Towards Computing Attributions for Dimensionality Reduction Techniques'

Project description

Interpretable t-SNE

License

interpretable_tsne is an implementation of our gradient-based attributiont technique described in our GLBIO oral presentation: 'Towards Computing Attributions for Dimensionality Reduction Techniques'.

To replicate the experiments we performed in our presentation, go to: MattScicluna/interpretable_tsne_experiment

Synthetic data experiment demonstrating that our t-SNE attributions can identify the ground truth significant features. See our pre-print for more details.

https://github.com/MattScicluna/interpretable_tsne/assets/19255250/789c075c-44fd-48ef-9264-e7dca60a59ce


Installation

interpretable_tsne requires Python 3.8 or higher to run. It is only available on Linux operating systems.

PyPi

This code can be installed via pip using the following command:

pip install interpretable_tsne

Source

You can install interpretable_tsne from source. Just clone this repository and run the following line in the root directory:

pip install .

Run the unittests

python -m unittest -v tests/*.py

Citation

If you use interpretable_tsne in your work, please cite us:

    @article {Scicluna2023.05.12.540592,
        author = {Matthew Crispin Scicluna and Jean-Christophe Grenier and Raphael Poujol and Sebastien Lemieux and Julie Hussin},
        title = {Towards Computing Attributions for Dimensionality Reduction Techniques},
        elocation-id = {2023.05.12.540592},
        year = {2023},
        doi = {10.1101/2023.05.12.540592},
        publisher = {Cold Spring Harbor Laboratory},
        URL = {https://www.biorxiv.org/content/early/2023/05/14/2023.05.12.540592},
        eprint = {https://www.biorxiv.org/content/early/2023/05/14/2023.05.12.540592.full.pdf},
        journal = {bioRxiv}
    }

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

interpretable_tsne-0.0.4.tar.gz (725.0 kB view details)

Uploaded Source

Built Distribution

interpretable_tsne-0.0.4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.7 MB view details)

Uploaded CPython 3.10 manylinux: glibc 2.17+ x86-64

File details

Details for the file interpretable_tsne-0.0.4.tar.gz.

File metadata

  • Download URL: interpretable_tsne-0.0.4.tar.gz
  • Upload date:
  • Size: 725.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.8.10

File hashes

Hashes for interpretable_tsne-0.0.4.tar.gz
Algorithm Hash digest
SHA256 658bf8dae86585dda4ed4de28b59f0dd37ad7a63029cc48c47ec3e5d2e68bb2f
MD5 0192bb5ec0e0a38c42ede3592521d576
BLAKE2b-256 a9fef36144d2c552d6df78e753912ff6edcd564e916c0dc7b6ee30c2d80817f2

See more details on using hashes here.

File details

Details for the file interpretable_tsne-0.0.4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.

File metadata

File hashes

Hashes for interpretable_tsne-0.0.4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm Hash digest
SHA256 9149bc17406fbe61e9dabee31e1ae0a77a80ff37c0d4ed77f6f103e212bbeadc
MD5 67c60cdea73e688493de6397a6d454ba
BLAKE2b-256 66f02e20cacbde00191033295c0ec0cb7b7f41c182ea58b67fa4bd249181c187

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page