scikit-learn-contrib/sklearn-ann

Integration with (approximate) nearest neighbors libraries for scikit-learn + clustering based on with kNN-graphs.

scikit-learn-contrib

scikit-learn-contrib is a github organization for gathering high-quality scikit-learn compatible projects. It also provides a template for establishing new scikit-learn compatible projects.

Vision

With the explosion of the number of machine learning papers, it becomes increasingly difficult for users and researchers to implement and compare algorithms. Even when authors release their software, it takes time to learn how to use it and how to apply it to one's own purposes. The goal of scikit-learn-contrib is to provide easy-to-install and easy-to-use high-quality machine learning software. With scikit-learn-contrib, users can install a project by pip install sklearn-contrib-project-name and immediately try it on their data with the usual fit, predict and transform methods. In addition, projects are compatible with scikit-learn tools such as grid search, pipelines, etc.

Projects

If you would like to include your own project in scikit-learn-contrib, take a look at the workflow.

DenMune: Density-peak clustering using mutual nearest neighbors

A simple-but-efficient density-based clustering algorithm that can find clusters of arbitrary size, shapes and densities in two-dimensions. Higher dimensions are first reduced to 2-D using the t-sne. The algorithm relies on a single parameter K, the number of nearest neighbors.

Read The Docs, Read the Paper

Maintained by: Mohamed Abbas

lightning

Large-scale linear classification, regression and ranking.

Maintained by Mathieu Blondel and Fabian Pedregosa.

skglm

Fast and modular Generalized Linear Models with support for models missing in scikit-learn.

Maintained by Mathurin Massias, Pierre-Antoine Bannier, Quentin Klopfenstein and Quentin Bertrand.

py-earth

A Python implementation of Jerome Friedman's Multivariate Adaptive Regression Splines.

Maintained by Jason Rudy and Mehdi.

imbalanced-learn

Python module to perform under sampling and over sampling with various techniques.

Maintained by Guillaume Lemaitre, Fernando Nogueira, Dayvid Oliveira and Christos Aridas.

polylearn

Factorization machines and polynomial networks for classification and regression in Python.

Maintained by Vlad Niculae.

forest-confidence-interval

Confidence intervals for scikit-learn forest algorithms.

Maintained by Ariel Rokem, Kivan Polimis and Bryna Hazelton.

hdbscan

A high performance implementation of HDBSCAN clustering.

Maintained by Leland McInnes, jc-healy, c-north and Steve Astels.

categorical-encoding

A library of sklearn compatible categorical variable encoders.

Maintained by Will McGinnis and Paul Westenthanner

boruta_py

Python implementations of the Boruta all-relevant feature selection method.

Maintained by Daniel Homola

sklearn-pandas

Pandas integration with sklearn.

Maintained by Israel Saeta Pérez

skope-rules

Machine learning with logical rules in Python.

Maintained by Florian Gardin, Ronan Gautier, Nicolas Goix and Jean-Matthieu Schertzer.

stability-selection

A Python implementation of the stability selection feature selection algorithm.

Maintained by Thomas Huijskens

metric-learn

Metric learning algorithms in Python.

Maintained by CJ Carey, Yuan Tang, William de Vazelhes, Aurélien Bellet and Nathalie Vauquier.

sklearn-ann's Issues

Maintenance status

Hi, is this still maintained?

We’d be interested to include something like it into scanpy.

In case you’re no longer interested in maintenance, maybe we could transfer this project to scverse and take over, publish it to PyPI, …

Add ScaNN

This is a top scoring system on ann-benchmarks

Add NGT KNeighbors transformer

https://github.com/yahoojapan/NGT

Specialised parameter search

Can get k = max knn graph and then iteratively truncate it and rerun

Can produce histogram as well as validation indices

Installation instructions

It would be nice if there were documented installation instructions, including information about what "extras" were available.

Likely downstream of #21 and getting a name on pypi.

Document parameters of FAISS

It would be nice if the parameters of the transformers were documented more fully.

In particular, FAISS's index_key argument is not particularly intuitive, and I think can have fairly complex variations.

Add transformers based on radius queries for libraries that support them

Publish to pypi?

AFAICT this isn't available on pypi. Could it get published?

Fix pre-commit check on main

Seems like the no-commit-to-branch hook gets run on main. I assume pre-commit.ci skips it somehow. We should do that too for that nice green checkmark

Make sure all KNeighbors transformers return sorted/zeroed outputs

May require postprocessing. Possibly there should be an option to disable this.

Fix readme to make releases possible

The release workflows currently fail with

 Checking dist/sklearn_ann-0.1.1-py3-none-any.whl: FAILED
ERROR    `long_description` has syntax errors in markup and would not be        
         rendered on PyPI.          

https://github.com/scikit-learn-contrib/sklearn-ann/actions/runs/8285153041/job/22672248085

maybe we simply missed adding readme = "README.rst" in pyproject.toml?

Add high dimensional million plus point example for RNN-DBSCAN

Fix readme for hatch

Investigate whether it's possible to make some of the knn indices picklable

It is possible to mmap within a pickle: https://github.com/UniNE-CHYN/mmappickle so there might be hope