Tags: salesforce/TransmogrifAI
Tags
0.7.0 release (#481) * Revert "Revert back to Spark 2.3 (#399)" This reverts commit 95a77b1. * Update to Spark 2.4.3 and XGBoost 0.90 * special double serializer fix * fix serialization * fix serialization * docs * fixed missng value for test * meta fix * Updated DecisionTreeNumericMapBucketizer test to deal with the change made to decision tree pruning in Spark 2.4. If nodes are split, but both child nodes lead to the same prediction then the split is pruned away. This updates the test so this doesn't happen for feature 'b' * fix params meta test * FIxed failing xgboost test * ident * cleanup * added dataframe reader and writer extensions * added const * cherrypick fixes * added xgboost params + update models to use public predict method * blarg * double ser test * update mleap and spark testing base * Update README.md * type fix * bump minor version * Update Spark version in the README * bump version * Update build.gradle * Update pom.xml * set correct json4s version * upgrade helloworld deps * upgrade notebook deps on TMog and Spark * bump to version 0.7.0 for Spark update * align helloworld dependencies * align helloworld dependencies * get -> getOrElse with exception * fix helloworld compilation * style * WIP release notes * TMog version bump * update release notes * update release notes * updates to changelog * updates to changelog * updates to changelog * updates to changelog * updates to changelog * updates to changelog * fix changelog * fix changelog * keep helloworld on 0.6.1 until release Co-authored-by: Matthew Tovbin <[email protected]> Co-authored-by: Matthew Tovbin <[email protected]> Co-authored-by: Christopher Suchanek <[email protected]> Co-authored-by: Kevin Moore <[email protected]> Co-authored-by: Matthew Tovbin <[email protected]>
0.6.1 release (#403) Bug fixes: - Ensure correct metrics despite model failures on some CV folds [#404](#404) - Fix flaky `ModelInsight` tests [#395](#395) - Avoid creating `SparseVector`s for LOCO [#377](#377) New features / updates: - Model combiner [#385](#399) - Added new sample for HousingPrices [#365](#365) - Test to verify that custom metrics appear in model insight metrics [#387](#387) - Add `FeatureDistribution` to `SerializationFormat`s [#383](#383) - Add metadata to `OpStandadrdScaler` to allow for descaling [#378](#378) - Improve json serde error in `evalMetFromJson` [#380](#380) - Track mean & standard deviation as metrics for numeric features and for text length of text features [#354](#354) - Making model selectors robust to failing models [#372](#372) - Use compact and compressed model json by default [#375](#375) - Descale feature contribution for Linear Regression & Logistic Regression [#345](#345) Dependency updates: - Update tika version [#382](#382)