susi 1.4.2

Python package for unsupervised, supervised and semi-supervised self-organizing maps (SOM)

SuSi: Supervised Self-organizing maps in Python

Python package for unsupervised, supervised and semi-supervised self-organizing maps (SOM)

Description

We present the SuSi package for Python. It includes a fully functional SOM for unsupervised, supervised and semi-supervised tasks:

• SOMClustering: Unsupervised SOM for clustering

• SOMRegressor: (Semi-)Supervised Regression SOM

• SOMClassifier: (Semi-)Supervised Classification SOM

License:

3-Clause BSD license

Author:

Felix M. Riese

Citation:

see Citation and in the bibtex file

Documentation:

Documentation

Installation:

Installation guidelines

Paper:

F. M. Riese, S. Keller and S. Hinz in Remote Sensing, 2020

Installation

Pip

pip3 install susi

Conda

conda install -c conda-forge susi

More information can be found in the installation guidelines.

Examples

A collection of code examples can be found in the documentation. Code examples as Jupyter Notebooks can be found here:

• examples/SOMClustering

• examples/SOMRegressor

• examples/SOMRegressor_semisupervised

• examples/SOMRegressor_multioutput

• examples/SOMClassifier

• examples/SOMClassifier_semisupervised

FAQs

• How should I set the initial hyperparameters of a SOM? For more details on the hyperparameters, see in documentation/hyperparameters.

• How can I optimize the hyperparameters? The SuSi hyperparameters can be optimized, for example, with scikit-learn.model_selection.GridSearchCV, since the SuSi package is developed according to several scikit-learn guidelines.

---

Citation

The bibtex file including both references is available in bibliography.bib.

Paper:

F. M. Riese, S. Keller and S. Hinz, “Supervised and Semi-Supervised Self-Organizing Maps for Regression and Classification Focusing on Hyperspectral Data”, Remote Sensing, vol. 12, no. 1, 2020. DOI:10.3390/rs12010007

@article{riese2020supervised,
    author = {Riese, Felix~M. and Keller, Sina and Hinz, Stefan},
    title = {{Supervised and Semi-Supervised Self-Organizing Maps for
              Regression and Classification Focusing on Hyperspectral Data}},
    journal = {Remote Sensing},
    year = {2020},
    volume = {12},
    number = {1},
    article-number = {7},
    URL = {https://www.mdpi.com/2072-4292/12/1/7},
    ISSN = {2072-4292},
    DOI = {10.3390/rs12010007}
}

Code:

Felix M. Riese, “SuSi: SUpervised Self-organIzing maps in Python”, Zenodo, 2019. DOI:10.5281/zenodo.2609130

@misc{riese2019susicode,
    author = {Riese, Felix~M.},
    title = {{SuSi: Supervised Self-Organizing Maps in Python}},
    year = {2019},
    DOI = {10.5281/zenodo.2609130},
    publisher = {Zenodo},
    howpublished = {\href{https://doi.org/10.5281/zenodo.2609130}{doi.org/10.5281/zenodo.2609130}}
}

---

License

This project is published under the 3-Clause BSD license.

Change Log

[1.4.2] - 2024-09-17

• [ADDED] Numpy types as placeholders

[1.4.1] - 2024-08-04

• [ADDED] Numpy 2 compatibility

[1.4.0] - 2024-03-29

• [ADDED] Official support for Python 3.12.

• [REMOVED] Official support for Python 3.8 (will be deprecated in 10/2024). It might still work, but will not be maintained on this version.

[1.3.0] - 2023-07-08

• [ADDED] Official support for Python 3.11.

• [FIXED] Quantization error axis bug.

• [REMOVED] Official support for Python 3.7 (deprecated in 2023). It might still work, but will not be maintained on this version.

[1.2.2] - 2021-12-11

• [ADDED] Official support for Python 3.10.

• [REMOVED] Official support for Python 3.6 (will be deprecated end of 2021 anyways). It might still work, but will not be maintained on this version.

[1.2.1] - 2021-10-19

• [ADDED] Quantization error get_quantization_error()

[1.2] - 2021-04-04

• [ADDED] Landing page with vuepress.

• [ADDED] Conda-forge recipe.

• [ADDED] Function SOMClassifier.predict_proba()

• [ADDED] Example notebook for multi-output regression

• [CHANGED] Code formatting to black.

• [CHANGED] CI from travis to GitHub actions.

• [FIXED] Requirements in setup.py

[1.1.2] - 2021-02-18

• [ADDED] Python 3.9 support. Python 3.6 support will be removed soon.

• [CHANGED] Function names for private use now start with an underscore.

[1.1.1] - 2020-11-18

• [ADDED] New distance metric “spectralangle”.

• [ADDED] FAQs.

• [ADDED] Separate between positional and keyword parameters.

• [ADDED] Plot script for neighborhood distance weight matrix.

• [FIXED] Added inherited members to code documentation.

[1.1.0] - 2020-08-31

• [ADDED] Logo.

• [ADDED] SOMPlots documentation.

• [REMOVED] Python 3.5 support. Now, only 3.6-3.8 are supported.

• [FIXED] Scikit-learn warnings regarding validation of positional arguments.

• [FIXED] Sphinx documentation warnings.

[1.0.10] - 2020-04-21

• [ADDED] Support for Python 3.8.x.

• [ADDED] Test coverage and MultiOutput test.

• [CHANGED] Function setPlaceholder to _set_placeholder.

• [FIXED] Documentation links

[1.0.9] - 2020-04-07

• [ADDED] Documentation of the hyperparameters.

• [ADDED] Plot scripts.

• [CHANGED] Structure of the module files.

[1.0.8] - 2020-01-20

• [FIXED] Replaced scikit-learn sklearn.utils.fixes.parallel_helper, see #12.

[1.0.7] - 2019-11-28

• [ADDED] Optional tqdm visualization of the SOM training

• [ADDED] New init_mode_supervised called random_minmax.

• [CHANGED] Official name of package changes from SUSI to SuSi.

• [CHANGED] Docstrings for functions are now according to guidelines.

• [FIXED] Semi-supervised classification handling, sample weights

• [FIXED] Supervised classification SOM initalization of n_iter_supervised

• [FIXED] Code refactored according to prospector

• [FIXED] Resolved bug in get_datapoints_from_node() for unsupervised SOM.

[1.0.6] - 2019-09-11

• [ADDED] Semi-supervised abilities for classifier and regressor

• [ADDED] Example notebooks for semi-supervised applications

• [ADDED] Tests for example notebooks

• [CHANGED] Requirements for the SuSi package

• [REMOVED] Support for Python 3.4

• [FIXED] Code looks better in documentation with sphinx.ext.napoleon

[1.0.5] - 2019-04-23

• [ADDED] PCA initialization of the SOM weights with 2 principal components

• [ADDED] Variable variance

• [CHANGED] Moved installation guidelines and examples to documentation

[1.0.4] - 2019-04-21

• [ADDED] Batch algorithm for unsupervised and supervised SOM

• [ADDED] Calculation of the unified distance matrix (u-matrix)

• [FIXED] Added estimator_check of scikit-learn and fixed recognized issues

[1.0.3] - 2019-04-09

• [ADDED] Link to arXiv paper

• [ADDED] Mexican-hat neighborhood distance weight

• [ADDED] Possibility for different initialization modes

• [CHANGED] Simplified initialization of estimators

• [FIXED] URLs and styles in documentation

• [FIXED] Colormap in Salinas example

[1.0.2] - 2019-03-27

• [ADDED] Codecov, Codacy

• [CHANGED] Moved decreasing_rate() out of SOM classes

• [FIXED] Removed duplicate constructor for SOMRegressor, fixed fit() params

[1.0.1] - 2019-03-26

• [ADDED] Config file for Travis

• [ADDED] Requirements for read-the-docs documentation

[1.0.0] - 2019-03-26

• Initial release