mlsauce

---

Miscellaneous Statistical/Machine learning stuff.

Contents

Installation for Python and R | Package description | Quick start | Contributing | Tests | Dependencies | Citing mlsauce | API Documentation | References | License

Installation (for Python and R)

Python

• 1st (preferred) method: from Github, for the development version

pip install git+https://github.com/Techtonique/mlsauce.git --verbose

• 2nd method: using conda

conda install -c conda-forge mlsauce 

(Note to self or developers: https://github.com/conda-forge/mlsauce-feedstock and https://conda-forge.org/docs/maintainer/adding_pkgs.html#step-by-step-instructions)

R

Only for Linux, for now. Windows users can envisage using WSL, the Windows Subsystem for Linux.

From GitHub

remotes::install_github("Techtonique/mlsauce_r") # the repo is in this organization

From R-universe

install.packages('mlsauce', repos = c('https://techtonique.r-universe.dev',
'https://cloud.r-project.org'))

General rule for using the package in R: object accesses with .'s are replaced by $'s. R Examples can be found in the package, once installed, by typing (in R console):

?mlsauce::AdaOpt

For a list of available models, visit https://techtonique.github.io/mlsauce/.

Package description

Miscellaneous Statistical/Machine learning stuff. See next section.

Quick start

Examples can be found here on GitHub. You can also read about this package here, and in particular for LSBoost: https://thierrymoudiki.github.io/blog/#LSBoost.

Contributing

Your contributions are welcome, and valuable. Please, make sure to read the Code of Conduct first. If you're not comfortable with Git/Version Control yet, please use this form to provide a feedback.

In Pull Requests, let's strive to use black for formatting files:

pip install black
black --line-length=80 file_submitted_for_pr.py

A few things that we could explore are:

• Enrich the tests
• Continue to make mlsauce available to R users --> here
• Any benchmarking of mlsauce models can be stored in demo (notebooks) or examples (flat files), with the following naming convention: yourgithubname_ddmmyy_shortdescriptionofdemo.[py|ipynb|R|Rmd]

Tests

Ultimately, tests for mlsauce's features will be located here. In order to run them and obtain tests' coverage (using nose2), you'll do:

• Install packages required for testing:

pip install nose2
pip install coverage

• Run tests and print coverage:

git clone https://github.com/thierrymoudiki/mlsauce.git
cd mlsauce
nose2 --with-coverage

• Obtain coverage reports:

At the command line:

coverage report -m

or an html report:

coverage html

Note to self and developpers: https://conda-forge.org/docs/maintainer/adding_pkgs.html#step-by-step-instructions

API Documentation

• https://techtonique.github.io/mlsauce/

Dependencies

• Numpy
• Scipy
• scikit-learn
• querier

Citation

@misc{moudiki2019mlsauce,
author={Moudiki, Thierry},
title={\code{mlsauce}, {M}iscellaneous {S}tatistical/{M}achine {L}earning stuff},
howpublished={\url{https://github.com/thierrymoudiki/mlsauce}},
note={BSD 3-Clause Clear License. Version 0.x.x.},
year={2019--2020}
}

References

• Moudiki, T. (2020). LSBoost, gradient boosted penalized nonlinear least squares. Available at: https://www.researchgate.net/publication/346059361_LSBoost_gradient_boosted_penalized_nonlinear_least_squares

• Moudiki, T. (2020). AdaOpt: Multivariable optimization for classification. Available at: https://www.researchgate.net/publication/341409169_AdaOpt_Multivariable_optimization_for_classification

License

BSD 3-Clause © Thierry Moudiki, 2019.

Credits

This package was created with Cookiecutter and the project template.