https://github.com/ibm/lale - Dependency Update Release

Maintenance release that supports newer versions of dependencies.

• Add support for Python 3.12
• Add support for Numpy>=2.0
• Support newer versions of dependencies:
  • imbalanced-learn: 0.13
  • XGBoost: 3.0
  • lightgbm: 4.6
  • SnapML: 1.16
  • scikit-learn: 1.6
• Improvements to test datasets
  • Replace use of no longer available dataset
  • Enable support for prefetching and caching datasets in CI
• Update all static tools

Removals/Known Issues: - Remove support for Python 3.8 and Python 3.9 - Python 3.9 is still believed to work , but it is no longer tested as part of CI and hence no longer supported. - Disable testing for RASL subsystem. - It should still work with older versions of Numpy and Python, however it currently breaks with newer versions - This breakage will hopefully be fixed, and tests re-enabled, in a future release. - Testing now generally happens against Numpy>=2.0. Numpy 1.X should still work, however it is no longer well-tested or supported.

- Python
Published by shinnar 11 months ago

https://github.com/ibm/lale - Version 0.8.4

Adds support for hyperopt 0.2.7

- Python
Published by shinnar over 1 year ago

https://github.com/ibm/lale - Version 0.8.3

• Visualizations as SVG with links and tooltips in jupyter notebook >=7.
• Eliminated dependency on numpy<2 directly in Lale, but note that some packages that are used with Lale still depend on numpy<2.

- Python
Published by hirzel over 1 year ago

https://github.com/ibm/lale - Version 0.8.2

• Improvements to the release deployment scripts

- Python
Published by shinnar almost 2 years ago

https://github.com/ibm/lale - Version 0.8.1

This release includes minor bug fixes

• Fix for xgb classifier without a label encoder
• URL links fix to open them in new tab
• Use newer AIF360 version that does not require an upper bound on numpy version

- Python
Published by shinnar almost 2 years ago

https://github.com/ibm/lale - Version 0.8.0

• Add support for Python 3.11
  • The SMAC hyperoptimization backend is not currently supported on 3.11
• Add support for scikit-learn 1.3 and 1.4
• Update supported versions of numpy, scipy, pandas, xgboost, lightgbm, snapml, and tensorflow
  • AIF360 does not currently support numpy>=1.24, so testing for aif360 libraries is done using an older version of numpy
• Fix readthedocs build so our documentation is correctly updated
• Add label encoding to our wrapper for XGBClassifier, since it longer does this.
• Shuffle data in some batching tests to (mostly) avoid problems with encoding
• Remove support for Python 3.7 and scikit-learn < 1.0
• Remove autoailibs wrappers and tests (they have been moved into the autoailibs package).

- Python
Published by shinnar almost 2 years ago

https://github.com/ibm/lale - Version 0.7.11

Fix a bug where links from pipeline visualizations to operator documentation did not work.

- Python
Published by hirzel about 2 years ago

https://github.com/ibm/lale - Version 0.7.10

• guard import of mystic
• add upper bound to tensorflow dependency
• remove protobuf installation version constraint.

- Python
Published by shinnar about 2 years ago

https://github.com/ibm/lale - Version 0.7.9

• Relax some dependency upper bounds
• Add RandomUnderSampler from imblearn
• Add Orbis and Urbis Mystic implementations
• Add 7 more fairness datasets
• Documentation improvements
• Add wrapper for autoai_libs.nsfa transformer

- Python
Published by shinnar over 2 years ago

https://github.com/ibm/lale - Version 0.7.8

• Updates to autoai_libs schemas to improve pretty-printing
• Update version of static checkers

- Python
Published by shinnar almost 3 years ago

https://github.com/ibm/lale - Version 0.7.7

• Improvements to CI
  • more static checking
  • only using trusted actions
• Improvements to aif360 schemas and derived operators
• Update SnapML schemas for new version
• loosen scipy release bounds

- Python
Published by shinnar almost 3 years ago

https://github.com/ibm/lale - 0.7.6

• suppress warnings from AIF360 when tempeh or fairlearn are not installed
• support np.bool_ in JSON schema validation and lale.lib.aif360
• eliminate duplicates in lale.lib.autogen to make it easier to maintain

- Python
Published by hirzel almost 3 years ago

https://github.com/ibm/lale - 0.7.5

• Rewrite importfromsklearnpipeline, reducing copying and calls to setparams
• update black version used for formatting code to 2023 standard
• Improvements to monoidal TargetEnoder implementation
• minor improvement to task graph code

- Python
Published by shinnar about 3 years ago

https://github.com/ibm/lale - 0.7.4

• Add support for scikit-learn 1.2.0
• Add a monoidal version of TargetEncoder
• Simplify importfromsklearn_pipeline and improve how it works with higher order operators
• Fix spark schema inspection so it uses only metadata
• Refactor schemas in autoai_libs

- Python
Published by shinnar about 3 years ago

https://github.com/ibm/lale - 0.7.3

1. Test-suite improvements such as adding sphinx build
2. Documentation improvements: fixed all sphinx warnings
3. New operators: TargetEncoder and fairSMOTE
4. Bug fixes and support of new schema features such as "transient":"alwaysPrint" to always print a hyperparameter value in pretty_print even if it has default value.

- Python
Published by kiran-kate about 3 years ago

https://github.com/ibm/lale - 0.7.2

CI and RASL fixes.

- Python
Published by kiran-kate over 3 years ago

https://github.com/ibm/lale - 0.7.1

• fixes to autogen schemas
• fix to autoai_libs DateTransformer

- Python
Published by hirzel over 3 years ago

https://github.com/ibm/lale - 0.7.0

• Improves support for partial_fit
• Improves the pretty printer
• Improves support for typed users
• Adds lale.lib.sklearn.perceptron (wrapping sklearn.linear_model.Perceptron)
• RASL (experimental):
  • Removes support for Spark Dataframes that don't have an index
  • Moves HashingEncoder to category_encoders and improved documentation

- Python
Published by shinnar over 3 years ago

https://github.com/ibm/lale - 0.6.19

Updated version of aif360 during installation.

- Python
Published by kiran-kate over 3 years ago

https://github.com/ibm/lale - 0.6.16

Adding py.typed marker to enable MyPy on packages that use Lale.

- Python
Published by hirzel over 3 years ago

https://github.com/ibm/lale - 0.6.17

1. fit_transform for lale operators
2. partial_fit for xgboost and lightgbm
3. Minor fixes and updates to README.

- Python
Published by kiran-kate over 3 years ago

https://github.com/ibm/lale - 0.6.16

Changed the version of black in setup.py compared to 0.6.15.

- Python
Published by kiran-kate over 3 years ago

https://github.com/ibm/lale - 0.6.15

• Add support for scikit-learn 1.1
• Add lower and upper bound constraints for scikit-learn to help suggest recommended versions
• Add support for newer versions of XGBoost

- Python
Published by shinnar over 3 years ago

https://github.com/ibm/lale - 0.6.14

Updated metrics to handle y as DataFrame.

- Python
Published by hirzel over 3 years ago

https://github.com/ibm/lale - 0.6.13

Release for the KDD'22 tutorial

- Python
Published by kiran-kate over 3 years ago

https://github.com/ibm/lale - 0.6.12

Hands-on tutorials for KDD'22: https://github.com/IBM/lale/tree/master/examples/kdd22

- Python
Published by kiran-kate over 3 years ago

https://github.com/ibm/lale - 0.6.11

1. RASL: balanced accuracy, balancedaccuracyand_di
2. Documentation improvements
3. lale.lib.autoai_libs.DateTransformer

- Python
Published by kiran-kate over 3 years ago

https://github.com/ibm/lale - 0.6.10

Fixes and changes to RASL, lale.lib.aif360 and import and export from sklearn.

- Python
Published by kiran-kate over 3 years ago

https://github.com/ibm/lale - 0.6.9

1. rasl fixes
2. a fix for autoaitslibs
3. a change to Hyperopt's fit to accept a validation dataset.

- Python
Published by kiran-kate almost 4 years ago

https://github.com/ibm/lale - 0.6.8

1. Batching can handle an iterable or data loader without knowing n_batches.
2. XGBoost 1.6
3. prettyprint lists a list of external modules in `wrapimported_operators`.

- Python
Published by kiran-kate almost 4 years ago

https://github.com/ibm/lale - 0.6.7

1. Batching changes to use task graphs
2. Removed autoaitslibs operators
3. BatchedTreeEnsemble estimators from SnapML
4. New rasl operators such as BatchedBaggingClassifier and HashingEncoder
5. Spilling in task graphs

- Python
Published by kiran-kate almost 4 years ago

https://github.com/ibm/lale - 0.6.6

1. Bug fixes
2. Improved interface for Monoids
3. Spilling in task graphs
4. multi-column index in SparkWithIndex

- Python
Published by kiran-kate almost 4 years ago

https://github.com/ibm/lale - 0.6.5

1. Fixes a regression (https://github.com/IBM/lale/commit/33d897218edd404ea5ddc4757c719f46fadf4bd8)
2. New lale.lib.rasl operators.

- Python
Published by kiran-kate almost 4 years ago

https://github.com/ibm/lale - 0.5.11

New release that delivers the string_indexer fix for 0.5.x.

- Python
Published by kiran-kate about 4 years ago

https://github.com/ibm/lale - 0.6.4

Added a new operator lale.lib.autoai_libs.ColumnSelector.

- Python
Published by kiran-kate about 4 years ago

https://github.com/ibm/lale - 0.6.3

Release with correct schema updates for xgboost 1.5.1.

- Python
Published by kiran-kate about 4 years ago

https://github.com/ibm/lale - 0.6.2

A version that is fully tested (almost, without static checks) on Python 3.9. Contains minor fixes compared to the previous version.

- Python
Published by kiran-kate about 4 years ago

https://github.com/ibm/lale - 0.5.10

1. New RASL operators: MinMaxScaler, OrdinalEncoder and OneHotEncoder
2. Fixes and changes for autoai-ts-libs
3. Scikit-learn compatibility by creating a steps property on lale pipelines and a mechanism to forward attribute access.

- Python
Published by kiran-kate about 4 years ago

https://github.com/ibm/lale - 0.6.1

1. New RASL operators: MinMaxScaler, OrdinalEncoder and OneHotEncoder
2. Fixes and changes for autoai-ts-libs
3. Scikit-learn compatibility by creating a steps property on lale pipelines and a mechanism to forward attribute access.

- Python
Published by kiran-kate about 4 years ago

https://github.com/ibm/lale - 0.5.9

Simplified combined fairness and predictive accuracy metrics to use a linear combination.

- Python
Published by hirzel about 4 years ago

https://github.com/ibm/lale - 0.5.8

• schema changes for autoaitslibs.
• partial_fit for a pipeline.
• diff of pipelines.
• some fixes and other changes.
• fixes for autoaitslibs.

- Python
Published by kiran-kate about 4 years ago

https://github.com/ibm/lale - 0.6.0

1. Schema changes for autoaitslibs.
2. partial_fit for a pipeline.
3. diff of pipelines.
4. Some fixes and other changes.

- Python
Published by kiran-kate about 4 years ago

https://github.com/ibm/lale - 0.5.7

• Making pretty_print() more robust.
• Making fairness support more robust.

- Python
Published by hirzel over 4 years ago

https://github.com/ibm/lale - 0.5.6

1. RASL operator implementation such as Filter, Aggregate, GroupBy, OrderBy etc.
2. Changes for ensembling experiments with lale.lib.aif360.
3. Refactoring of lale.lib.aif360 and creation of a new setup target fairness.
4. Customize schemas if the environment has sklearn 1.0.
5. Update of schema constraints based on the "weakest precondition" work.
6. Other changes and bug fixes.

- Python
Published by kiran-kate over 4 years ago

https://github.com/ibm/lale - 0.5.5

1. Access to 2 multi-table datasets: go_sales and imdb.
2. Improvement in error messages
3. Support for predictlogproba
4. Bug fixes

- Python
Published by kiran-kate over 4 years ago

https://github.com/ibm/lale - 0.5.4

1. New StackingClassifier and StackingRegressor operators
2. Lale operators for autoai-ts-libs
3. New Join operator
4. Bug fixes

- Python
Published by kiran-kate over 4 years ago

https://github.com/ibm/lale - 0.5.3

1. New datasets to lale.lib.aif360: SpeedDating, Boston housing, MEPS, Titanic, COMPAS, Teaching Assistant Evaluation, and Nursery
2. Improvements in documentation and schemas
3. Minor bug fixes
4. New curated schemas for IsoMap and IsolationForest

- Python
Published by kiran-kate over 4 years ago

https://github.com/ibm/lale - 0.5.2

Added an option for timeout to grid search and halving grid search. Also removed codecov from our build due to a reported security issue with codecov.

- Python
Published by kiran-kate almost 5 years ago

https://github.com/ibm/lale - 0.5.1

Fixes and more changes for sklearn compatibility.

- Python
Published by kiran-kate almost 5 years ago

https://github.com/ibm/lale - 0.5.0

Added LinearSVR, removed BaseClassifier and regressor, and added support for get_params(deep=True).

- Python
Published by kiran-kate almost 5 years ago

https://github.com/ibm/lale - 0.4.15

Support for HalvingGridSearchCV and score. Improved documentation and scikit-learn wrappers.

- Python
Published by kiran-kate almost 5 years ago

https://github.com/ibm/lale - 0.4.14

Improvements to cloning, new SnapML estimators, improvements to AIF360 wrappers.

- Python
Published by hirzel about 5 years ago

https://github.com/ibm/lale - 0.4.12

Disabled data schema validation by default with an option to control it via lale/settings.py, new SnapML estimators, changes to aif360 datasets.

- Python
Published by kiran-kate about 5 years ago

https://github.com/ibm/lale - 0.4.11

New changes for aif360 support and replacement of pai4sk by publicly available SnapML package.

- Python
Published by kiran-kate about 5 years ago

https://github.com/ibm/lale - 0.4.10

Support for new mitigators and metrics from aif360, minor bug fixes and schema changes.

- Python
Published by kiran-kate about 5 years ago

https://github.com/ibm/lale - 0.4.9

Improvements to AIF360 support, and implemented Map operator for pandas.

- Python
Published by hirzel about 5 years ago

https://github.com/ibm/lale - Infrastructure improvements

Add support for auto-deployment of releases to PyPi

- Python
Published by shinnar over 5 years ago

https://github.com/ibm/lale - 0.4.7

Support for categorical attributes in using aif360 with Lale, scorers that combine performance and disparate impact score, change to use the latest hyperopt version, switch from travis to github actions and minor bug fixes.

- Python
Published by kiran-kate over 5 years ago

https://github.com/ibm/lale - 0.4.6

More checks with flake, schema simplification bug fixes, upgrade to new versions of aif360, pandas and support for numpy ndarrays and categorical values while using disparate_impact scorer with Lale.

- Python
Published by kiran-kate over 5 years ago

https://github.com/ibm/lale - 0.4.5

Schema changes to some estimators and changes after applying black and isort.

- Python
Published by kiran-kate over 5 years ago

https://github.com/ibm/lale - Lale 0.4.4

Improvements to pretty-printing and error messages.

- Python
Published by hirzel over 5 years ago

https://github.com/ibm/lale - 0.4.3

AutoPipeline operator and bug fixes for sklearn 0.23 support and pretty print of autoai_libs operators.

- Python
Published by kiran-kate over 5 years ago

https://github.com/ibm/lale - 0.4.2

Upgraded the default supported version of sklearn to be >=0.20.3, and minor bug fixes related to the previous release.

- Python
Published by kiran-kate over 5 years ago

https://github.com/ibm/lale - 0.4.1

New operators for data joins and aggregations, schema changes to many operators to support sklearn version >=0.22 and minor fixes.

- Python
Published by kiran-kate over 5 years ago

https://github.com/ibm/lale - Lale 0.4.0

Improved type checking and type-based search space pruning. Moved XGBoost and LightGBM from required to full dependencies. Improved sklearn_compat and added three Snap ML operators.

- Python
Published by hirzel over 5 years ago

https://github.com/ibm/lale - Minor updates

- Python
Published by shinnar over 5 years ago

https://github.com/ibm/lale - 0.3.19

Added test coverage, more test cases and some bug fixes.

- Python
Published by kiran-kate over 5 years ago

https://github.com/ibm/lale - 0.3.18

Minor fixes.

- Python
Published by kiran-kate over 5 years ago

https://github.com/ibm/lale - 0.3.16

Minor fixes.

- Python
Published by kiran-kate almost 6 years ago

https://github.com/ibm/lale - 0.3.15

Minor fixes.

- Python
Published by kiran-kate almost 6 years ago

https://github.com/ibm/lale - 0.3.14

Minor updates to README and fixes to 3 operators.

- Python
Published by kiran-kate almost 6 years ago

https://github.com/ibm/lale - Lale 0.3.13

Minor fixes.

- Python
Published by hirzel almost 6 years ago

https://github.com/ibm/lale - Lale 0.3.12

This release adds operator lale.lib.autoai_libs.TNoOp.

- Python
Published by hirzel almost 6 years ago

https://github.com/ibm/lale - 0.3.11

Schema changes to remove hyper-parameter constraints which do not raise any errors from scikit-learn.

- Python
Published by kiran-kate almost 6 years ago

https://github.com/ibm/lale - 0.3.10

This release contains schema changes for some lale.lib.sklearn operators.

- Python
Published by kiran-kate almost 6 years ago

https://github.com/ibm/lale - Lale 0.3.9

This release contains some improvements to pretty-printing and error messages.

- Python
Published by hirzel almost 6 years ago

https://github.com/ibm/lale - 0.3.8

Bug fixes and documentation improvements. Addition of BaselineClassifier and BaselineRegressor.

- Python
Published by kiran-kate almost 6 years ago

https://github.com/ibm/lale - 0.3.7

Bug fixes from the previous release.

- Python
Published by kiran-kate almost 6 years ago

https://github.com/ibm/lale - 0.3.6

pretty-print bug fixes, added TopKVotingClassifier, added support for new libraries such as autoai_libs and imbalanced-learn, updated autopen schemas for documentation and more documentation enhancements.

- Python
Published by kiran-kate almost 6 years ago

https://github.com/ibm/lale - Lale 0.3.5

Besides the capabilities from v0.3.4.6, this release makes it easier to use classifiers in the middle of a pipeline by supporting decision_function.

- Python
Published by hirzel almost 6 years ago

https://github.com/ibm/lale - v0.3.4.6

This release includes new features such as support for lists and tuples for nested higher order operators, schema validation changes to expect separate schemas for transform and predict and to allow more flexible schemas ('laleType':'any'), documentation updates and ability to generate doc strings from schemas, fixes and updates to dependencies, fixes and updates to import and export from and to scikit-learn pipelines. It also includes bug fixes and enhancements to visualization, pretty_print, operator schemas, optimizer wrappers etc.

- Python
Published by kiran-kate almost 6 years ago

https://github.com/ibm/lale - Lale 0.3.3

This release makes convenience methods (removelast, visualize, prettyprint, transform_schema) available on all operators (individual, pipeline, choice). Also, this release makes improvements to higher-order operators (such as ensembles) and grammars.

- Python
Published by hirzel about 6 years ago

https://github.com/ibm/lale - Lale 0.3.2

This release turns search tools into proper Lale operators, so they work more smoothly with wrapimportedoperators and auto_configure. Also, this release expands the use of (sub-)schema checking on intermediate datasets.

- Python
Published by hirzel about 6 years ago

https://github.com/ibm/lale - Fix dependencies

Changes dependencies on liac-arff and mypy.

- Python
Published by gbdrt about 6 years ago

https://github.com/ibm/lale - Latest

Latest with new features: AIF360, Batching, automatically generated schemas, Glove embeddings and Universal Sentence Encoder and other fixes and updates.

- Python
Published by kiran-kate over 6 years ago