Published November 19, 2026 | Version v0.3.0

OpenADMET/openadmet-models: v0.3.0

Description

What's Changed

  • [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci[bot] in https://github.com/OpenADMET/openadmet-models/pull/570
  • Make shuffle and drop_last training-only across deep-learning workflows by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/572
  • Consolidate random seeding under random_seed (global + per-section) by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/573
  • Free runner disk space before Docker build by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/574
  • Set persistent_workers=True in ChemProp DataLoader to prevent fd exhaustion by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/575
  • [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci[bot] in https://github.com/OpenADMET/openadmet-models/pull/578
  • [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci[bot] in https://github.com/OpenADMET/openadmet-models/pull/581
  • Migrate ReadTheDocs conda tool from mambaforge to miniforge3 by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/583
  • [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci[bot] in https://github.com/OpenADMET/openadmet-models/pull/585
  • Fix for boto3 dependency in Docker containers by @tlhr in https://github.com/OpenADMET/openadmet-models/pull/586
  • Fix UnscaleTransform identity when no scaler is provided by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/577
  • Make shape handling consistent across package for #509 by @dwwest in https://github.com/OpenADMET/openadmet-models/pull/512
  • Fix ChemProp scheduler configuration, API semantics, and model defaults by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/576
  • Add TabICL v2 architecture with regression and classification models by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/590
  • Rename butina cutoff kwarg to dist_cutoff by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/599
  • Add CheMeleon embedding featurizer and model predict_embedding by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/588
  • Add PCA dimensionality reduction to the anvil pipeline by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/592
  • Add TrainedModelFeaturizer for featurizing with a pretrained model by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/598
  • fix: resolve accelerator to device in TabPFN model classes by @li-lizhe in https://github.com/OpenADMET/openadmet-models/pull/601
  • Fix GPU CI workflow by @cynthiaxu04 in https://github.com/OpenADMET/openadmet-models/pull/606
  • [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci[bot] in https://github.com/OpenADMET/openadmet-models/pull/593
  • [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci[bot] in https://github.com/OpenADMET/openadmet-models/pull/611
  • Add MonroeFeaturizer for Monroe foundation model embeddings by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/602
  • fix: use 'auto' accelerator (not 'gpu') as default for predict methods by @li-lizhe in https://github.com/OpenADMET/openadmet-models/pull/604
  • Add featurizer alias for keying same-type feature blocks by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/603
  • [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci[bot] in https://github.com/OpenADMET/openadmet-models/pull/613
  • [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci[bot] in https://github.com/OpenADMET/openadmet-models/pull/619
  • Exclude fit_time and score_time from cross-validation metrics by @aaronkanzer in https://github.com/OpenADMET/openadmet-models/pull/621
  • Add a line to the pull request template by @dwwest in https://github.com/OpenADMET/openadmet-models/pull/618
  • Hook post-hoc binary metrics and plots into standard evaluation framework by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/584
  • [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci[bot] in https://github.com/OpenADMET/openadmet-models/pull/623
  • Fix loading gpu-saved tabpfn model loading by @dwwest in https://github.com/OpenADMET/openadmet-models/pull/616
  • Fix pairwise inference task index selection by @SergeiNikolenko in https://github.com/OpenADMET/openadmet-models/pull/505
  • Honour ye olde configured aggregation whence building from a foundation model by @dwwest in https://github.com/OpenADMET/openadmet-models/pull/624
  • Support pixi, uv, pip, and conda installs with locked environments by @smcolby in https://github.com/OpenADMET/openadmet-models/pull/617

New Contributors

  • @tlhr made their first contribution in https://github.com/OpenADMET/openadmet-models/pull/586
  • @li-lizhe made their first contribution in https://github.com/OpenADMET/openadmet-models/pull/601
  • @aaronkanzer made their first contribution in https://github.com/OpenADMET/openadmet-models/pull/621

Full Changelog: https://github.com/OpenADMET/openadmet-models/compare/v0.2.0...v0.3.0

Files

OpenADMET/openadmet-models-v0.3.0.zip

Files (20.0 MB)

Name Size
md5:1089a86480ccf89c43f179401535fe65
20.0 MB Preview Download

Additional details

Related works