susi 1.0.9

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. The class structure is set up as follows:

• SOMClustering: Unsupervised SOM for clustering

  • SOMEstimator: Base class for supervised and semi-supervised SOMs

    • SOMRegressor: Regression SOM

    • SOMClassifier: 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 install 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/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? Because the SuSi package is developed according to several scikit-learn guidelines, it can be used with, for example, scikit-learn.model_selection.GridSearchCV.

---

Citation

The bibtex file including both references is available here.

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.

@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”, 10.5281/zenodo.2609130, 2019.

@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}}
}

Change Log

[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