Data

Tools

Indexes

Applications

Experiments

Open Source Science

Surjectors

Surjectors: surjection layers for density estimation with normalizing flows - Published in JOSS (2024)

https://github.com/dirmeier/surjectors

Science Score: 93.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ✓
    DOI references
    Found 5 DOI reference(s) in README and JOSS metadata
  • ✓
    Academic publication links
    Links to: joss.theoj.org
  • ○
    Committers with academic emails
  • ○
    Institutional organization owner
  • ✓
    JOSS paper metadata
    Published in Journal of Open Source Software

Keywords

bijections density-estimation normalizing-flows python surjections

Scientific Fields

Artificial Intelligence and Machine Learning Computer Science - 83% confidence
Last synced: 6 months ago · JSON representation

Repository

Surjection layers for density estimation with normalizing flows

Basic Info
  • Host: GitHub
  • Owner: dirmeier
  • License: apache-2.0
  • Language: Python
  • Default Branch: main
  • Homepage: https://surjectors.rtfd.io
  • Size: 478 KB
Statistics
  • Stars: 4
  • Watchers: 2
  • Forks: 1
  • Open Issues: 2
  • Releases: 5
Topics
bijections density-estimation normalizing-flows python surjections
Created over 3 years ago · Last pushed 6 months ago
Metadata Files
Readme License

README.md

surjectors

active ci version doi

Surjection layers for density estimation with normalizing flows

About

Surjectors is a light-weight library for density estimation using inference and generative surjective normalizing flows, i.e., flows can that reduce or increase dimensionality. Surjectors makes use of

  • Haiku`s module system for neural networks,
  • Distrax for probability distributions and some base bijectors,
  • Optax for gradient-based optimization,
  • JAX for autodiff and XLA computation.

Examples

You can, for instance, construct a simple normalizing flow like this:

```python import distrax import haiku as hk from jax import numpy as jnp, random as jr from surjectors import Slice, LULinear, Chain from surjectors import TransformedDistribution from surjectors.nn import make_mlp

def decoderfn(ndim): def fn(z): params = makemlp([32, 32, ndim * 2])(z) means, logscales = jnp.split(params, 2, -1) return distrax.Independent(distrax.Normal(means, jnp.exp(log_scales))) return _fn

@hk.withoutapplyrng @hk.transform def flow(x): basedistribution = distrax.Independent( distrax.Normal(jnp.zeros(5), jnp.ones(5)), 1 ) transform = Chain([Slice(5, decoderfn(5)), LULinear(5)]) pushforward = TransformedDistribution(basedistribution, transform) return pushforward.logprob(x)

x = jr.normal(jr.PRNGKey(1), (1, 10)) params = flow.init(jr.PRNGKey(2), x) lp = flow.apply(params, x) ```

More self-contained examples can be found in examples.

Documentation

Documentation can be found here.

Installation

Make sure to have a working JAX installation. Depending whether you want to use CPU/GPU/TPU, please follow these instructions.

To install the package from PyPI, call:

bash pip install surjectors

To install the latest GitHub , just call the following on the command line:

bash pip install git+https://github.com/dirmeier/surjectors@<RELEASE>

Contributing

Contributions in the form of pull requests are more than welcome. A good way to start is to check out issues labelled good first issue.

In order to contribute:

1) Clone Surjectors and install hatch via pip install hatch, 2) create a new branch locally git checkout -b feature/my-new-feature or git checkout -b issue/fixes-bug, 3) implement your contribution and ideally a test case, 4) test it by calling hatch run test on the (Unix) command line, 5) submit a PR 🙂

Citing Surjectors

If you find our work relevant to your research, please consider citing:

@article{dirmeier2024surjectors, author = {Simon Dirmeier}, title = {Surjectors: surjection layers for density estimation with normalizing flows}, year = {2024}, journal = {Journal of Open Source Software}, publisher = {The Open Journal}, volume = {9}, number = {94}, pages = {6188}, doi = {10.21105/joss.06188} }

Author

Simon Dirmeier sfyrbnd @ pm me

Owner

  • Name: Simon Dirmeier
  • Login: dirmeier
  • Kind: user
  • Location: Zurich, Switzerland
  • Company: @SwissDataScienceCenter

ML research @SwissDataScienceCenter

JOSS Publication

Surjectors: surjection layers for density estimation with normalizing flows
Published
February 29, 2024
DOI
Volume 9, Issue 94, Page 6188
Authors
Simon Dirmeier
Swiss Data Science Center, Zurich, Switzerland, ETH Zurich, Zurich, Switzerland
Editor
Arfon Smith ORCID
Tags
JAX Density estimation Normalizing flow Machine learning Statistics
View PDF Review Thread Software Archive

GitHub Events

Total
  • Issues event: 1
  • Watch event: 1
  • Push event: 2
Last Year
  • Issues event: 1
  • Watch event: 1
  • Push event: 2

Committers

Last synced: 7 months ago

All Time
  • Total Commits: 45
  • Total Committers: 1
  • Avg Commits per committer: 45.0
  • Development Distribution Score (DDS): 0.0
Past Year
  • Commits: 1
  • Committers: 1
  • Avg Commits per committer: 1.0
  • Development Distribution Score (DDS): 0.0
Top Committers
Name Email Commits
Simon Dirmeier s****3@p****m 45

Issues and Pull Requests

Last synced: 6 months ago

All Time
  • Total issues: 10
  • Total pull requests: 26
  • Average time to close issues: 8 months
  • Average time to close pull requests: 7 days
  • Total issue authors: 3
  • Total pull request authors: 1
  • Average comments per issue: 1.1
  • Average comments per pull request: 0.35
  • Merged pull requests: 23
  • Bot issues: 0
  • Bot pull requests: 0
Past Year
  • Issues: 1
  • Pull requests: 0
  • Average time to close issues: N/A
  • Average time to close pull requests: N/A
  • Issue authors: 1
  • Pull request authors: 0
  • Average comments per issue: 0.0
  • Average comments per pull request: 0
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
View more stats
Top Authors
Issue Authors
  • dirmeier (8)
  • sandeshkatakam (1)
  • animikhaich (1)
Pull Request Authors
  • dirmeier (31)
Top Labels
Issue Labels
Pull Request Labels

Packages

  • Total packages: 1
  • Total downloads:
    • pypi 31 last-month
  • Total dependent packages: 1
  • Total dependent repositories: 1
  • Total versions: 5
  • Total maintainers: 1
pypi.org: surjectors

Surjection layers for density estimation with normalizing flows

  • Versions: 5
  • Dependent Packages: 1
  • Dependent Repositories: 1
  • Downloads: 31 Last month
Rankings
Dependent packages count: 10.1%
Dependent repos count: 21.6%
Average: 25.4%
Downloads: 44.5%
Maintainers (1)
dirmeier
Last synced: 6 months ago

Dependencies

.github/workflows/ci.yaml actions
  • actions/checkout v3 composite
  • actions/setup-python v3 composite
  • pre-commit/action v3.0.0 composite
.github/workflows/release.yaml actions
  • actions/checkout v2 composite
  • actions/setup-python v2 composite
  • pypa/gh-action-pypi-publish release/v1 composite
pyproject.toml pypi
  • distrax >=0.1.2
  • dm-haiku >=0.0.9
  • optax >=0.1.3
docs/requirements.txt pypi
  • ipykernel *
  • ipython *
  • matplotlib *
  • nbsphinx *
  • pandas *
  • scikit-learn *
  • seaborn *
  • session_info *
  • sphinx *
  • sphinx-autobuild *
  • sphinx-book-theme >=1.1.0
  • sphinx-copybutton *
  • sphinx-math-dollar *
  • sphinx_autodoc_typehints *
  • sphinx_design *
  • sphinx_fontawesome *
  • sphinx_gallery *
  • sphinxcontrib-fulltoc *
  • tqdm *
.github/workflows/examples.yaml actions
  • actions/checkout v3 composite
  • actions/setup-python v3 composite