v0.9.6

v0.9.6 (2024-04-18)

Full Changelog

This is a minor release.

Changed

• Removed explicit dependency on Pydantic (#1002)

Development

• Bump tj-actions/changed-files from 1.1.2 to 41.0.0 in /.github/workflows (#992)
• Update README.md (#996)
• Update numba requirement from !=0.54.0,<0.59.0,>=0.50.0 to >=0.50.0,!=0.54.0,<0.60.0 (#999)
• Update twine requirement from <5.0.0,>3.2.0 to >3.2.0,<6.0.0 (#1001)
• build(ci): Migrate actions to later Node version (#1003)

v0.9.5

v0.9.5 (2024-01-22)

Full Changelog

This is a patch release fixing several bugs, updating dependencies and a change of license.

Fixed

• Fix torch version bound in setup.py extras_require (#950)
• Fix DistributedExplainer import errors that arise when ray absent(#951)
• Fix memory limit issue in tox ci jobs (#956)
• Fix E721 linting errors (#958)
• Fix plot_pd function to work with matplotlib 3.8.0 changes (#965)
• Fix typechecking with matplotlib 3.8.0 (#969)
• fix typechecking for matplotlib 3.8.1 (#981)
• Fix typechecking for mypy 1.7.0 (#983)
• Fix test models to output logits and work with default loss functions (#975)
• Fix dtype type in helper method for AnchorText samplers (#980)

Changed

• Alibi License change from Apache to Business Source License 1.1 (#995)

Development

• Update myst-parser requirement upper bound from 2.0 to 3.0 (#931)
• Update pillow requirement upper bound from 10.0 to 11.0 (#939)
• Add notebooks tests for python 3.11 (#948) & (#949)
• Update sphinxcontrib-apidoc requirement upper bound from 0.4.0 to 0.5.0 (#962)
• Update numba requirement upper bound from 0.58.0 to 0.59.0 (#967)
• Update shap requirement upper bound from 0.43.0 to 0.44.0 (#974)
• Update tensorflow requirement upper bound from 2.14.0 to 2.15.0 (#968)
• Update Alibi_Explain_Logo_rgb image with white stroked letters (#979)
• Remove macos from ci (#995)
• Add security scans to CI (#995)

v0.9.4

v0.9.4 (2023-07-07)

Full Changelog

This is a patch release to support numpy >= 1.24 and scikit-learn >= 1.3.0, and drop official support for Python 3.7.

Development

• Drop official support for Python 3.7 (#942).
• Maximum supported version of shap bumped to 0.42.x, to give numpy >= 1.24 support (#944).
• Fix handling of scikit-learn partial_dependence kwarg's in tests, for compatibility with scikit-learn 1.3.0 (#940).

v0.9.3

v0.9.3 (2023-06-21)

Full Changelog

This is a patch release to officially enable support for Python 3.11.
This is the last release with official support for Python 3.7.

Development

• Test library on Python 3.11 (#932).
• Separate code quality into its own Github Action and only run against the main development version of Python, currently Python 3.10 (#925).
• Check and remove stale mypy ignore commands (#874).
• Bump torch version to 2.x (#893).
• Bump scikit-image version to 0.21.x (#928).
• Bump numba version to 0.57.x (#922).
• Bump sphinx version to 7.x (#921).

v0.9.2

v0.9.2 (2023-04-28)

Full Changelog

This is a patch release fixing several bugs, updating dependencies and adding some small extensions.

Added

• Allow IntegratedGradients layer selection to be specified with a custom callable (#894).
• Implement reset_predictor method for PartialDependence explainer (#897).
• Extend GradientSimilarity explainer to allow models of any input type (#912).

Fixed

• AnchorText auto-regressive language model sampler updating input_ids tensor (#895).
• AnchorTabular length discrepancy between feature and names fields (#902).
• AnchorBaseBeam unintended coverage update during the multi-armed bandit run (#919, #914).

Changed

• Maximum supported version of tensorflow bumped to 2.12.x (#896).
• Supported version of pandas bumped to >1.0.0, <3.0.0 (#899).
• Update notebooks to account for pandas version 2.x deprecations (#908, #910).
• Maximum supported version of scikit-image bumped to 0.20.x (#882).
• Maximum supported version of attrs bumped to 23.x (#905).

Development

• Migrate codecov to use Github Actions and don't fail CI on coverage report upload failure due to rate limiting (#901, #913).
• Bumpy mypy version to >=1.0, <2.0 (#886).
• Bump sphinx version to 6.x (#852).
• Bump sphinx-design version to 0.4.1 (#904).
• Bump nbsphinx version to 0.9.x (#889).
• Bump myst-parser version to >=1.0, <2.0 (#887).
• Bump twine version to 4.x (#620).
• Bump pre-commit version to 3.x and update the config (#866).

v0.9.1

v0.9.1 (2023-03-13)

Full Changelog

This is a patch release fixing several bugs.

Fixed

• Replace deprecated usage of np.object in the codebase which was causing errors with numpy>=1.24 (#872, #890).
• Fix a bug/typo in cfrl_base.py of the tensorflow backend (#891).
• Correctly handle calls to .reset_predictor for KernelShap and TreeShap explainers (#880).
• Update saving of KernelShap to avoid saving the internal _explainer object (#881).

Development

• Show text execution times (#849).
• Replace Python 2 style type comments with type annotations (#870).
• Bump the mypy version to ~1.0 (#871).
• Bump the default development version of Python to 3.10 (#876, #877).

v0.9.0

v0.9.0 (2023-01-11)

Full Changelog

Added

• New feature PermutationImportance explainer implementing the permutation feature importance global explanations. Also included is a plot_permutation_importance utility function for flexible plotting of the resulting feature importance scores (docs, #798).
• New feature PartialDependenceVariance explainer implementing partial dependence variance global explanations. Also included is a plot_pd_variance utility function for flexible plotting of the resulting PD variance plots (docs, #758).

Fixed

• GradientSimilarity explainer now automatically handles sparse tensors in the model by converting the gradient tensors to dense ones before calculating similarity. This used to be a source of bugs when calculating similarity for models with embedding layers for which gradients tensors are sparse by default. Additionally, it now filters any non-trainable parameters and doesn't consider those in the calculation as no gradients exist. A warning is raised if any non-trainable layers or parameters are detected (#829).
• Updated the discussion of the interpretation of ALE. The previous examples and documentation had some misleading claims; these have been removed and reworked with an emphasis on the mostly qualitative interpretation of ALE plots (#838, #846).

Changed

• Deprecated the use of the legacy Boston housing dataset in examples and testing. The new examples now use the California housing dataset (#838, #834).
• Modularized the computation of prototype importances and plotting for ProtoSelect, allowing greater flexibility to the end user (#826).
• Roadmap documentation page removed due to going out of date (#842).

Development

• Tests added for tensorflow models used in CounterfactualRL (#793).
• Tests added for pytorch models used in CounterfactualRL (#799).
• Tests added for ALE plotting functionality (#816).
• Tests added for PartialDependence plotting functionality (#819).
• Tests added for PartialDependenceVariance plotting functionality (#820).
• Tests added for PermutationImportance plotting functionality (#824).
• Tests addef for ProtoSelect plotting functionality (#841).
• Tests added for the datasets subpackage (#814).
• Fixed optional dependency installation during CI to make sure dependencies are consistent (#817).
• Synchronize notebook CI workflow with the main CI workflow (#818).
• Version of pytest-cov bumped to 4.x (#794).
• Version of pytest-xdist bumped to 3.x (#808).
• Version of tox bumped to 4.x (#832).

v0.8.0

v0.8.0 (2022-09-26)

Full Changelog

Added

• New feature PartialDependence and TreePartialDependence explainers implementing partial dependence (PD) global explanations. Also included is a plot_pd utility function for flexible plotting of the resulting PD plots (docs, #721).
• New exceptions.NotFittedError exception which is raised whenever a compulsory call to a fit method has not been carried out. Specifically, this is now raised in AnchorTabular.explain when AnchorTabular.fit has been skipped (#732).
• Various improvements to docs and examples (#695, #701, #698, #703, #717, #711, #750, #784).

Fixed

• Edge case in AnchorTabular where an error is raised during an explain call if the instance contains a categorical feature value not seen in the training data (#742).

Changed

• Improved handling of custom grid_points for the ALE explainer (#731).
• Renamed our custom exception classes to remove the verbose Alibi* prefix and standardised the *Error suffix. Concretely:
  • exceptions.AlibiPredictorCallException is now exceptions.PredictorCallError
  • exceptions.AlibiPredictorReturnTypeError is now exceptions.PredictorReturnTypeError. Backwards compatibility has been maintained by subclassing the new exception classes by the old ones, but these will likely be removed in a future version (#733).
• Warn users when TreeShap is used with more than 100 samples in the background dataset which is due to a limitation in the upstream shap package (#710).
• Minimum version of scikit-learn bumped to 1.0.0 mainly due to upcoming deprecations (#776).
• Minimum version of scikit-image bumped to 0.17.2 to fix a possible bug when using the slic segmentation function with AnchorImage (#753).
• Maximum supported version of attrs bumped to 22.x (#727).
• Maximum supported version of tensorflow bumped to 2.10.x (#745).
• Maximum supported version of ray bumped to 2.x (#740).
• Maximum supported version of numba bumped to 0.56.x (#724).
• Maximum supported version of shap bumped to 0.41.x (#702).
• Updated shap example notebooks to recommend installing matplotlib==3.5.3 due to failure of shap plotting functions with matplotlib==3.6.0 (#776).

Development

• Extend optional dependency checks to ensure the correct submodules are present (#714).
• Introduce pytest-custom_exit_code to let notebook CI pass when no notebooks are selected for tests (#728).
• Use UTF-8 encoding when loading README.md in setup.py to avoid a possible failure of installation for some users (#744).
• Updated guidance for class docstrings (#743).
• Reinstate ray tests (#756).
• We now exclude test files from test coverage for a more accurate representation of coverage (#751). Note that this has led to a drop in code covered which will be addressed in due course (#760).
• The Python 3.10.x version on CI has been pinned to 3.10.6 due to typechecking failures, pending a new release of mypy (#761).
• The test_changed_notebooks workflow can now be triggered manually and is run on push/PR for any branch (#762).
• Use codecov flags for more granular reporting of code coverage (#759).
• Option to ssh into Github Actions runs for remote debugging of CI pipelines (#770).
• Version of sphinx bumped to 5.x but capped at <5.1.0 to avoid CI failures (#722).
• Version of myst-parser bumped to 0.18.x (#693).
• Version of flake8 bumped to 5.x (#729).
• Version of ipykernel bumped to 6.x (#431).
• Version of ipython bumped to 8.x (#572).
• Version of pytest bumped to 7.x (#591).
• Version of sphinx-design bumped to 0.3.0 (#739).
• Version of nbconvert bumped to 7.x (#738).

v0.7.0

v0.7.0 (2022-05-18)

Full Changelog

This release introduces two new methods, a GradientSimilarity explainer and a ProtoSelect data summarisation algorithm.

Added

• New feature GradientSimilarity explainer for explaining predictions of gradient-based (PyTorch and TensorFlow) models by returning the most similar training data points from the point of view of the model (docs).
• New feature We have introduced a new subpackage alibi.prototypes which contains the ProtoSelect algorithm for summarising datasets with a representative set of "prototypes" (docs).
• ALE explainer now can take a custom grid-point per feature to evaluate the ALE on. This can help in certain situations when grid-points defined by quantiles might not be the best choice (docs).
• Extended the IntegratedGradients method target selection to handle explaining any scalar dimension of tensors of any rank (previously only rank-1 and rank-2 were supported). See #635.
• Python 3.10 support. Note that PyTorch at the time of writing doesn't support Python 3.10 on Windows.

Fixed

• Fixed a bug which incorrectly handled multi-dimensional scaling in CounterfactualProto (#646).
• Fixed a bug in the example using CounterfactualRLTabular (#651).

Changed

• tensorflow is now an optional dependency. To use methods that require tensorflow you can install alibi using pip install alibi[tensorflow] which will pull in a supported version. For full instructions for the recommended way of installing optional dependencies please refer to Installation docs.
• Updated sklearn version bounds to scikit-learn>=0.22.0, <2.0.0.
• Updated tensorflow maximum allowed version to 2.9.x.

Development

• This release introduces a way to manage the absence of optional dependencies. In short, the design is such that if an optional dependency is required for an algorithm but missing, at import time the corresponding public (or private in the case of the optional dependency being required for a subset of the functionality of a private class) algorithm class will be replaced by a MissingDependency object. For full details on developing alibi with optional dependencies see Contributing: Optional Dependencies.
• The CONTRIBUTING.md has been updated with further instructions for managing optional dependencies (see point above) and more conventions around docstrings.
• We have split the Explainer base class into Base and Explainer to facilitate reusability and better class hierarchy semantics with introducing methods that are not explainers (#649).
• mypy has been updated to ~=0.900 which requires additional development dependencies for type stubs, currently only types-requests has been necessary to add to requirements/dev.txt.
• Fron this release onwards we exclude the directories doc/ and examples/ from the source distribution (by adding prune directives in MANIFEST.in). This results in considerably smaller file sizes for the source distribution.

v0.6.5

v0.6.5 (2022-03-18)

Full Changelog

This is a patch release to correct a regression in CounterfactualProto introduced in v0.6.3.

Added

• Added a Frequently Asked Questions page to the docs.

Fixed

• Fix a bug introduced in v0.6.3 which prevented CounterfactualProto working with categorical features (#612).
• Fix an issue with the LanguageModelSampler where it would sometimes sample punctuation (#585).

Development

• The maximum tensorflow version has been bumped from 2.7 to 2.8 (#588).