PyCVI 0.1.5

2024/10/23

Python versions

This version supports Python versions 3.8 to 3.11.

New Features

• pycvi.compute_scores.compute_all_scores, CVIs and their corresponding functions can now use a specific numpy Random Generator for reproducibility purpose.

Changes

• pycvi.cvi.select now raises a SelectionError if no clustering could be selected (was None previously)

Fixes

• Fix Davies-Bouldin in order to ignore comparing the distance between a centroid to itself when computing the index.

Documentation

• Added "how to cite" section
• Fixed truncated examples

Contributors

• Natacha Galmiche (@nglm)

PyCVI 0.1.4

2024/09/24

Python versions

This version supports Python versions 3.8 to 3.11.

Fixes

• Some package compatibility issues fixed
• Minor fixes in the documentation
• JOSS paper fixed
• GitHub actions updated

Contributors

• Natacha Galmiche (@nglm)

PyCVI 0.1.3

2024/08/20

Python versions

This version supports Python versions 3.8 to 3.11.

New Features

• The class pycvi.cvi.CVIAggregator was added in order to be able to easily combine several CVIs to select the best clustering. An example featuring pycvi.cvi.CVIAggregator has been added to the documentation: CVIAggregator: Combining CVIs
• pycvi.cvi.compute_all_scores can now take either a CVI or a CVIAggregator as an argument.

Changes

• pycvi.cvi.select method now raises a SelectionError when all cvi values were equal instead of choosing the clustering with the smallest number of clusters.

Fixes

• Some package compatibility issues fixed
• Minor fixes in the documentation

Contributors

• Natacha Galmiche (@nglm)

PyCVI 0.1.2

2024/07/02

Python versions

This version supports Python versions 3.8 to 3.11.

Fixes

• Some dependency incompatibilities have been fixed.

Features

• Example scripts can now be run with the set of extra dependencies "examples". This extra set of dependency can be installed by running:

  # using poetry
  poetry add pycvi-lib[examples]
  # using pip or anaconda
  pip install pycvi-lib[examples]

• All instructions to run examples scripts are now available in the documentation: Running example scripts on your computer.

Contributors

• Natacha Galmiche (@nglm)

PyCVI 0.1.1

PyCVI 0.1.1 release notes

Changes

• Added Selection error

PyCVI 0.1.0

PyCVI 0.1.0 release notes

2023/12/22

First release of PyCVI.

Python versions

This version supports Python versions 3.8 to 3.11.

Features

• 12 internal CVIs implemented: Hartigan, Calinski-Harabasz, GapStatistic, Silhouette, ScoreFunction, Maulik-Bandyopadhyay, SD, SDbw, Dunn, Xie-Beni, XB* and Davies-Bouldin.
• Compute CVI values and select the best clustering based on the results.
• Compatible with time-series, Dynamic Time Warping (DTW) and Dynamic time warping Barycentric Average (DBA).
• Compatible with scikit-learn, scikit-learn extra, aeon and sktime, for an easy integration into any clustering pipeline in python.
• Can compute the clusterings beforehand if provided with a sklearn-like clustering class.
• Enable users to define custom CVIs.
• Variation of Information implemented (distances between clustering).
• counterparts of pdist and cdist in scipy, for time series and non time-series data.
• DTW and DBA based on aeon.

Contributors

• Natacha Galmiche (@nglm)