Repository files navigation

Functional Data Analysis, or FDA, is the field of Statistics that analysesdata that depend on a continuous parameter.

This package offers classes, methods and functions to give support to FDAin Python. Includes a wide range of utils to work with functional data, and itsrepresentation, exploratory analysis, or preprocessing, among other taskssuch as inference, classification, regression or clustering of functional data.See documentation for further information on the features included in thepackage.

The documentation is available atfda.readthedocs.io/en/stable/, whichincludes detailed information of the different modules, classes and methods ofthe package, along with severalexamples showing different functionalities.

The documentation of the latest version, corresponding with the developversion of the package, can be found atfda.readthedocs.io/en/latest/.

If you want a quick overview of the package, we recommend you to try thenew:doc:`tutorial <auto_tutorial/index>`. For articles about specifictopics, feel free to explore the:doc:`examples <auto_examples/index>`. Wantto check the documentation of a particular class or function? Try searchingfor it in the:doc:`API list <apilist>`.

Currently,scikit-fda is available in Python versions above 3.8, regardless of theplatform.The stable version can be installed viaPyPI:

pip install scikit-fda

It is also available fromconda-forge:

conda install -c conda-forge scikit-fda

It is possible to install the latest version of the package, available in thedevelop branch, by cloning this repository and doing a manual installation.

git clone https://github.com/GAA-UAM/scikit-fda.gitpip install ./scikit-fda

Make sure that your default Python version is currently supported, or changethe python and pip commands by specifying a version, such aspython3.8:

git clone https://github.com/GAA-UAM/scikit-fda.gitpython3.8 -m pip install ./scikit-fda

scikit-fda depends on the following packages:

• fdasrsf - SRSF framework
• findiff - Finite differences
• matplotlib - Plotting with Python
• multimethod - Multiple dispatch
• numpy - The fundamental package for scientific computing with Python
• pandas - Powerful Python data analysis toolkit
• rdata - Reader of R datasets in .rda format in Python
• scikit-datasets - Scikit-learn compatible datasets
• scikit-learn - Machine learning in Python
• scipy - Scientific computation in Python
• setuptools - Python Packaging

The dependencies are automatically installed.

Please, if you find this software useful in your work, reference it citing the following paper:

@article{ramos-carreno++_2024_scikit-fda,  author = {Ramos-Carreño, Carlos and Torrecilla, José L. and Carbajo Berrocal, Miguel and Marcos Manchón, Pablo and Suárez, Alberto},  doi = {10.18637/jss.v109.i02},  journal = {Journal of Statistical Software},  month = may,  number = {2},  pages = {1--37},  title = {{scikit-fda: A Python Package for Functional Data Analysis}},  url = {https://www.jstatsoft.org/article/view/v109i02},  volume = {109},  year = {2024}}

You can additionally cite the software repository itself using:

@misc{ramos-carreno++_2024_scikit-fda-repo,  author = {The scikit-fda developers},  doi = {10.5281/zenodo.3468127},  month = feb,  title = {scikit-fda: Functional Data Analysis in Python},  url = {https://github.com/GAA-UAM/scikit-fda},  year = {2024}}

If you want to reference a particular version for reproducibility, check the version-specific DOIs available in Zenodo.

All contributions are welcome. You can help this project grow in multiple ways,from creating an issue, reporting an improvement or a bug, to doing arepository fork and creating a pull request to the development branch.

The people involved at some point in the development of the package can befound in thecontributorsfile.

The package is licensed under the BSD 3-Clause License. A copy of thelicense can be found along with the code.

The project has received financial support from projects PID2019-109387GB-I00,PID2019-106827GB-I00, and PID2022-139856NB-I00, funded byMCIN/ AEI / 10.13039/501100011033 / FEDER, UE,project IDEA-CM (TEC-2024/COM-89) from the Autonomous Community of Madrid,and from the ELLIS Unit Madrid. The authors acknowledge computational supportfrom the Centro de Computación Científica-Universidad Autónoma de Madrid(CCC-UAM).