Release Notes (What's New)

Version 2.0.0 (December 7, 2024)

For a list of all changes in this release, see the full changelog. Below are the changes we think users may wish to be aware of.

Breaking Changes

The function scores.probability.tw_crps_for_ensemble previously took an optional (mis-spelled) argument chainging_func_kwargs. The spelling has been corrected and the argument is now chaining_func_kwargs. See PR #780 and PR #772.
For those who develop on scores, you will need to update your installation of the scores package with pip install -e .[all], to get updated versions of black, pylint and mypy. See PR #768, PR #769 and PR #771.

Features

Added three new metrics:
- Brier score for ensembles: scores.probability.brier_score_for_ensemble. See PR #735.
- Negative predictive value: scores.categorical.BasicContingencyManager.negative_predictive_value. See PR #759.
- Positive predictive value: scores.categorical.BasicContingencyManager.positive_predictive_value. See PR #761 and PR #756.
Also added one new emerging metric and two supporting functions:
- Risk matrix score: scores.emerging.risk_matrix_scores.
- Risk matrix score - matrix weights to array: scores.emerging.matrix_weights_to_array.
- Risk matrix score - warning scaling to weight array: scores.emerging.weights_from_warning_scaling.
  See PR #724 and PR #794.
A new method called format_table was added to the class BasicContingencyManager to improve visualisation of 2x2 contingency tables. The tutorial Binary_Contingency_Scores was updated to demonstrate the use of this function. See PR #775.
The functions scores.processing.comparative_discretise, scores.processing.binary_discretise and scores.processing.binary_discretise_proportion now accept either a string indicating the choice of operator to be used, or an operator from the Python core library operator module. Using one of the operators from the Python core module is recommended, as doing so is more reliable for a variety of reasons. Support for the use of a string may be removed in future. See PR #740 and PR #758.

Documentation

Added "The Risk Matrix Score" tutorial. See PR #724 and PR #794.
Updated the "Brier Score" tutorial to include a new section about the Brier score for ensembles. See PR #735.
Updated the "Binary Categorical Scores and Binary Contingency Tables (Confusion Matrices)"
tutorial:
- Included "positive predictive value" in the list of binary categorical scores.
- Included "negative predictive value" in the list of binary categorical scores.
- Demonstrated the use of the new format_table method for visualising 2x2 contingency tables.
  See PR #759 and PR #775.
Updated the “Contributing Guide”:
- Added a new section: "Creating Your Own Fork of scores for the First Time".
- Updated the section: "Workflow for Submitting Pull Requests".
- Added a new section: "Pull Request Etiquette".
  See PR #787.
Updated the README:
- Added a link to a video of a PyCon AU 2024 conference presentation about scores. See PR #783.
- Added a link to the archives of scores on Zenodo. See PR #784.
Added Scoringrules to "Related Works". See PR #746, PR #766 and PR #789.

Internal Changes

Removed scikit-learn as a dependency. scores has replaced the use of scikit-learn with a similar function from SciPy (which was an existing scores dependency). This change was manually tested and found to be faster. See PR #774.
Version pinning of dependencies in release files (the wheel and sdist files used by PyPI and conda-forge) is now managed and set by the hatch_build script. This allows development versions to be free-floating, while being more specific about dependencies in releases. The previous process also aimed to do this, but was error-prone. A new entry called pinned_dependencies was added to pyproject.toml to specify the release dependencies. See PR #760.

Contributors to this Release

Arshia Sharma* (@arshiaar), A.J. Fisher* (@AJTheDataGuy), Liam Bluett* (@lbluett), Jinghan Fu* (@JinghanFu), Sam Bishop* (@techdragon), Robert J. Taggart (@rob-taggart), Tennessee Leeuwenburg (@tennlee), Stephanie Chong (@Steph-Chong) and Nicholas Loveday (@nicholasloveday).

* indicates that this release contains their first contribution to scores.

Climate Change - Earth and Climate Modeling - Jupyter Notebook
Published by tennlee 5 months ago

Release Notes (What's New)

Version 1.3.0 (November 15, 2024)

For a list of all changes in this release, see the full changelog. Below are the changes we think users may wish to be aware of.

Introduced Support for Python 3.13 and Dropped Support for Python 3.9

In line with other scientific Python packages, scores has dropped support for Python 3.9 in this release.
scores has added support for Python 3.13. See PR #710.

Features

Added four new metrics:
- Quantile Interval Score: scores.continuous.quantile_interval_score. See PR #704, PR #733 and PR #738.
- Interval Score: scores.continuous.interval_score. See PR #704, PR #733 and PR #738.
- Kling-Gupta Efficiency (KGE): scores.continuous.kge. See PR #679, PR #700 and PR #734.
- Interval threshold weighted continuous ranked probability score (twCRPS) for ensembles: scores.probability.interval_tw_crps_for_ensemble. See PR #682 and PR #734.
Added an optional include_components argument to several continuous ranked probability score (CRPS) functions for ensembles. If supplied, the include_components argument will return the underforecast penalty, the overforecast penalty and the forecast spread term, in addition to the overall CRPS value. This applies to the following CRPS functions:
- continuous ranked probability score (CRPS) for ensembles: scores.probability.crps_for_ensemble
- threshold weighted continuous ranked probability score (twCRPS) for ensembles: scores.probability.tw_crps_for_ensemble
- tail threshold weighted continuous ranked probability score (twCRPS) for ensembles: scores.probability.tail_tw_crps_for_ensemble
- interval threshold weighted continuous ranked probability score (twCRPS) for ensembles: scores.probability.interval_tw_crps_for_ensemble)
  See PR #708 and PR #734.

Documentation

Added "Kling–Gupta Efficiency (KGE)" tutorial. See PR #679, PR #700 and PR #734.
Added "Quantile Interval Score and Interval Score" tutorial. See PR #704, PR #736 and PR #738.
Added "Threshold Weighted Continuous Ranked Probability Score (twCRPS) for ensembles" tutorial. See PR #706 and PR #722.
Updated the title in the "Binary Categorical Scores and Binary Contingency Tables (Confusion Matrices)" tutorial and the description for the corresponding thumbnail in the tutorial gallery. See PR #741 and PR #743.
Updated the pull request template. See PR #719.

Internal Changes

Sped up (improved the computational efficiency of) the continuous ranked probability score (CRPS) for ensembles. This also addresses memory issues when a large number of ensemble members are present. See PR #694.

Contributors to this Release

Mohammadreza Khanarmuei (@reza-armuei), Nicholas Loveday (@nicholasloveday), Durga Shrestha (@durgals), Tennessee Leeuwenburg (@tennlee), Stephanie Chong (@Steph-Chong) and Robert J. Taggart (@rob-taggart).

Climate Change - Earth and Climate Modeling - Jupyter Notebook
Published by tennlee 5 months ago

Scores - 1.2.0

Release Notes (What's New)

Version 1.2.0 (September 13, 2024)

For a list of all changes in this release, see the full changelog. Below are the changes we think users may wish to be aware of.

Features

Added three new metrics:
- Percent bias (PBIAS): scores.continuous.pbias. See PR #639 and PR #655.
- Threshold weighted continuous ranked probability score (twCRPS) for ensembles: scores.probability.tw_crps_for_ensemble. See PR #644.
- Tail threshold weighted continuous ranked probability score (twCRPS) for ensembles: scores.probability.tail_tw_crps_for_ensemble. See PR #644.
The FIxed Risk Multicategorical (FIRM) score (scores.categorical.firm) can now take a sequence of mulitdimensional arrays (xr.DataArray) of thresholds. This allows the FIRM score to be used with categorical thresholds that vary across the domain. See PR #661.

Documentation

Added information about percent bias to the "Additive Bias and Multiplicative Bias" tutorial. See PR #639 and PR #656.
Updated documentation to say there are now over 60 metrics, statistical techniques and data processing tools contained in scores. See PR #659.
In the "Contributing Guide", updated instructions for installing a conda-based virtual environment. See PR #654.

Internal Changes

Modified automated tests to work with NumPy 2.1. Incorporated a union type of array and generic in assert statements for Dask operations. See PR #643.

Contributors to this Release

Durga Shrestha* (@durgals), Maree Carroll (@mareecarroll), Nicholas Loveday (@nicholasloveday), Tennessee Leeuwenburg (@tennlee), Stephanie Chong (@Steph-Chong) and Robert J. Taggart (@rob-taggart).

* indicates that this release contains their first contribution to scores.

Climate Change - Earth and Climate Modeling - Jupyter Notebook
Published by tennlee 8 months ago

Scores - 1.1.0

Release Notes (What's New)

Version 1.1.0 (August 9, 2024)

For a list of all changes in this release, see the full changelog. Below are the changes we think users may wish to be aware of.

Features

scores is now available on conda-forge.
Added five new metrics
- threshold weighted squared error: scores.continuous.tw_squared_error
- threshold weighted absolute error: scores.continuous.tw_absolute_error
- threshold weighted quantile score: scores.continuous.tw_quantile_score
- threshold weighted expectile score: scores.continuous.tw_expectile_score
- threshold weighted Huber loss: scores.continuous.tw_huber_loss.
  See PR #609.

Documentation

Added "Threshold Weighted Scores" tutorial. See PR #609.
Removed nbviewer link from documentation. See PR #615.

Internal Changes

Modified numpy.trapezoid call to work with either NumPy 1 or 2. See PR #610.

Contributors to this Release

Nicholas Loveday (@nicholasloveday), Tennessee Leeuwenburg (@tennlee), Stephanie Chong (@Steph-Chong) and Robert J. Taggart (@rob-taggart).

Climate Change - Earth and Climate Modeling - Jupyter Notebook
Published by tennlee 9 months ago

Scores - 1.0.0

Release Notes (What's New)

Version 1.0.0 (July 10, 2024)

We are happy to have reached the point of releasing “Version 1.0.0” of scores. While we look forward to many version increments to come, version 1.0.0 represents a milestone. It signifies a stabilisation of the API, and marks a turning point from the initial construction period. We have also published a paper in the Journal of Open Source Software (see citation further below).

From this point forward, scores will be following the Semantic Versioning Specification (SemVer) in its release management.

This is a good moment to acknowledge and thank the contributors that helped us reach this point. They are: Tennessee Leeuwenburg, Nicholas Loveday, Elizabeth E. Ebert, Harrison Cook, Mohammadreza Khanarmuei, Robert J. Taggart, Nikeeth Ramanathan, Maree Carroll, Stephanie Chong, Aidan Griffiths and John Sharples.

Please consider a citation of our paper if you use our code. The citation is:

Leeuwenburg, T., Loveday, N., Ebert, E. E., Cook, H., Khanarmuei, M., Taggart, R. J., Ramanathan, N., Carroll, M., Chong, S., Griffiths, A., & Sharples, J. (2024). scores: A Python package for verifying and evaluating models and predictions with xarray. Journal of Open Source Software, 9(99), 6889. https://doi.org/10.21105/joss.06889

BibTeX:

@article{Leeuwenburg_scores_A_Python_2024,
author = {Leeuwenburg, Tennessee and Loveday, Nicholas and Ebert, Elizabeth E. and Cook, Harrison and Khanarmuei, Mohammadreza and Taggart, Robert J. and Ramanathan, Nikeeth and Carroll, Maree and Chong, Stephanie and Griffiths, Aidan and Sharples, John},
doi = {10.21105/joss.06889},
journal = {Journal of Open Source Software},
month = jul,
number = {99},
pages = {6889},
title = {{scores: A Python package for verifying and evaluating models and predictions with xarray}},
url = {https://joss.theoj.org/papers/10.21105/joss.06889},
volume = {9},
year = {2024}
}

For the full details of all changes in this release, see the GitHub commit history.

Climate Change - Earth and Climate Modeling - Jupyter Notebook
Published by tennlee 10 months ago

Scores - 0.9.3

Release Notes (What's New)

Version 0.9.3 (July 9, 2024)

For the full details of all changes in this release, see the GitHub commit history. Below are the changes we think users may wish to be aware of.

Breaking Changes

Renamed and relocated function scores.continuous.correlation to scores.continuous.correlation.pearsonr. See PR #583 by @nicholasloveday.

Documentation

Added "Dimension Handling" tutorial, which describes reducing and preserving dimensions. See PR #589 by @nicholasloveday.
Updated "Detailed Installation Guide" with information on installing kernels in a Jupyter environment. See PR #586 by @tennlee and PR #587 by @Steph-Chong.

Internal Changes

Introduced pinned versions for dependencies on main. See PR #580 by @tennlee.

Climate Change - Earth and Climate Modeling - Jupyter Notebook
Published by tennlee 10 months ago

Scores - 0.9.2

What's Changed

Add Badges to the README for CodeQL, code coverage, and binder link by @tennlee in https://github.com/nci/scores/pull/555
Substantially update "Data Sources" page in documentation by @Steph-Chong in https://github.com/nci/scores/pull/544
Add a Key Features page to docs by @Steph-Chong in https://github.com/nci/scores/pull/567
Addition of consistent scoring rules by @nicholasloveday in https://github.com/nci/scores/pull/540
Release 0.9.2 by @tennlee in https://github.com/nci/scores/pull/570

Full Changelog: https://github.com/nci/scores/compare/0.9.1...0.9.2

Climate Change - Earth and Climate Modeling - Jupyter Notebook
Published by tennlee 10 months ago