v0.9.3

This release contains the usual dependency compatibility updates, as well as:

• new impute_nan setting to many featurizers, that allows missing features to be imputed automatically as the mean of the remainder of that column. This is particularly helpful when e.g., trying to use a feature that uses elemental data on a dataset where maybe one row contains an element not present in the dataset (#892)
• added a featurizer based on elemental pseudo-inverse of tabulated optical/transport data (#892)

What's Changed

• Optical and transport data as elemental pseudo-inverse contributions by @gbrunin in #892
• Do not run test PyPI build on forks by @ml-evs in #933
• Added missing impute_nan for WenAlloys and removed unnecessary warning by @gbrunin in #937
• Suppress pymatgen element data warnings when imputing by @ml-evs in #934
• Disable release creation workflow on forks by @gbrunin in #940
• Bug fixes in pseudo inverse computations by @gbrunin in #942
• Disable test PyPI build by @ml-evs in #941
• Automatic deps upgrades by @ml-evs in #944
• spglib 2.5 compatibility by @esoteric-ephemera in #943
• Support NumPy 2 by @DanielYang59 in #949

New Contributors

• @gbrunin made their first contribution in #892
• @esoteric-ephemera made their first contribution in #943

Full Changelog: v0.9.2...v0.9.3

v0.9.2

This release simply removes the upper bound on pandas v2, which, since the changes implemented in v0.9.1, is no longer necessary.

What's Changed

• Attempt to release the pandas pin by @ml-evs in #929
• Automated dependency upgrades by @tschaume in #930

Full Changelog: v0.9.1...v0.9.2

v0.9.1

This release contains several compatibility fixes across tests and dependencies, including support for the latest numpy and pymatgen versions.
Dependencies have been tightened to make sure that any future incompatibilities do not render this package entirely unusable.
Users are encouraged to install the pinned dependencies from requirements/*.txt if they run into any issues.

What's Changed

• Linting and test updates by @ml-evs in #912
• Prevent warning in fingerprint.py during yaml.safe_load by @afonari in #914
• Fix maximum element failure by @JaGeo in #921
• Begin unpicking dependency issues by @ml-evs in #922
• Correct valence electron numbers by @DanielYang59 in #917
• Reintroduce pandas pin and update code for numpy 1.24+ by @ml-evs in #925
• More dependency upgrades: support for latest pymatgen and numpy by @ml-evs in #928

New Contributors

• @afonari made their first contribution in #914
• @JaGeo made their first contribution in #921
• @DanielYang59 made their first contribution in #917

Full Changelog: v0.9.0...v0.9.1

v0.9.0

What's Changed

• Add hint how to avoid OOM errors to BaseFeaturizer.set_n_jobs() doc str by @janosh in #894
• Add ability to customize BaseFeaturizer pbar by passing dict by @janosh in #893
• Assign __init__ arg to instance attribute to fix sklearn base __repr__ by @ml-evs in #896
• Fixes for CI and tests by @ml-evs in #901
• Optimize BandCenter and IntersticeDistribution featurizers by @ml-evs in #897
• Prequel to dependency upgrades by @ml-evs in #903
• Skip the thermo tests rather than crash if MP API is missing by @ml-evs in #905
• Defer MPRester import to allow it to act more like an optional dep by @ml-evs in #906

New Contributors

• @ml-evs made their first contribution in #896

Full Changelog: v0.8.0...v0.9.0

v0.8.0

Version 0.8.0

Commits

• [97a0bc6] trying again ... (#890)
• [6a34b63] rm publishPackage workflow_dispatch
• [2ef3a56] update docs
• [c67904f] yet another description change (#889)
• [5ef7d3c] try explicitly checking out main
• [48e652d] minor description change (#888)
• [1fdd698] update docs
• [d639200] release workflow_dispatch
• [0c53b60] set git user
• [d129826] add description (#887)
• [0d65adc] Automated dependency upgrades (#886)
• [aabebc3] add sphinx dep
• [f4f450a] work on release through github actions (#885)
• [a0ae13d] also pull tags for scm version
• [78ce6d6] merge docs build into release
• [34c569e] skip tests for now
• [c762580] only run tests on changes in matminer subdir
• [08f1dae] fix action typo
• [cb8634d] add docs and release to github actions
• [e59bb20] echo tests
• [ca7fa1e] no-deps
• [ff38fea] Automated dependency upgrades (#884)
• [bf1d37c] cache-dependency-path 2
• [fc5d9b8] dependabot not needed anymore
• [7f6a1c1] cache-dependency-path
• [b770dac] update all test actions
• [10b3d9a] move jsonschema to mpds extra
• [26b0e37] update github actions
• [5d9c661] include dev dependencies
• [61c5f3e] add requirements files (pip-compile)
• [3f742f3] setup.py: explicitly list dependencies
• [a0f754a] delete old requirements files
• [8f1742e] add upgrade-dependencies action
• [7f8520b] reformat to make linter happy
• [e8bacc1] reformat
• [0f6f62c] allow for automatic retries of dataset downloads, plus alter tests (attempt to fix hash problem that fails tests randomly)
• [4a86c0a] Merge pull request #839 from hackingmaterials/dependabot/pip/ujson-5.4.0
• [66b2163] format fix
• [618f6f7] fix linting
• [ead30fe] Merge pull request #809 from jacksund/main
• [886524a] Merge pull request #841 from janosh/crystal-nn-fingerprint-docs
• [2dd8761] explain meaning of 'cn' | 'ops' in CrystalNNFingerprint.from_preset doc string
• [19e7319] Bump ujson from 5.2.0 to 5.4.0
• [b3e2954] fix psitestats tests
• [d471b57] Merge branch 'hackingmaterials:main' into main

v0.7.8

manual base release for new github actions

v0.7.6

v0.7.6-release

Update to v0.7.6

v0.7.6

Update reqs

v0.7.4

Update reqs

v0.7.3

Update reqs