All Versions
170
Latest Version
Avg Release Cycle
13 days
Latest Release
-

Changelog History
Page 1

  • v3.34.0.1 Changes

    πŸš€ Download at: http://h2o-release.s3.amazonaws.com/h2o/rel-zizler/1/index.html

    πŸ› Bug

    • πŸ›  [PUBDEV-8326] - Fixed matplotlib 3.4 compatibility issues with partial_plot.
    • πŸ—„ [PUBDEV-8316] - Deprecated is_supervised parameter for h2o.grid method in R.
    • πŸ›  [PUBDEV-8314] - Fixed AutoML NPE by ensuring that models without metrics are not added to the leaderboard.
    • [PUBDEV-8295] - Redistributed the time budget for AutoML.
    • πŸ›  [PUBDEV-8290] - Fixed and reorganized the H2O Explain leaderboard and fixed the confusion matrix.
    • [PUBDEV-8289] - Decreased the number of displayed features in the heatmap for AutoML inside H2O Explain.
    • πŸ›  [PUBDEV-8276] - Fixed NPE raised from weight_column not being in the training model.
    • πŸ“š [PUBDEV-8274] - Fixed the weight=0 documentation change error.
    • βœ… [PUBDEV-8271] - Fixed failing rotterdam tests.
    • πŸ›  [PUBDEV-8267] - Fixed GAM NPE from multiple runs with knots specified in a frame.
    • [PUBDEV-8266] - Fixed col_sample_rate not sampling for XGBoost when set to a value lower than 1.0.
    • πŸ›  [PUBDEV-8257] - Fixed wrong column type on MOJO models for Cross-Validation Metrics Summary.
    • [PUBDEV-8245] - Prevented R connect from starting H2O locally.
    • [PUBDEV-8233] - Added StackedEnsembles to AutoML’s time budget to prevent unexpected training times.
    • [PUBDEV-8210] - Fixed the failing pyunit_scale_pca_rf.py test.
    • [PUBDEV-8175] - Improved AutoML behavior when multiple instances are created in parallel.
    • [PUBDEV-7855] - Solved corner cases involving mapping between encoded varimps and predictor columns for H2O Explain by making the varimp feature consolidation more robust.

    πŸ‘Œ Improvement

    • [PUBDEV-8273] - Ensured that AutoML uses the entire time budget for max_runtime.
    • [PUBDEV-8196] - Implemented custom progress widgets for Wave apps using H2O-3.
    • πŸ–¨ [PUBDEV-8189] - Allowed users to convert floats to doubles with PrintMojo to prevent possible parsing issues.
    • ⚑️ [PUBDEV-8185] - Updated GBM cross validation with early_stopping to use ntrees that produce the best score.
    • πŸ–¨ [PUBDEV-8184] - Enabled print_mojo to produce .png outputs.
    • ⚑️ [PUBDEV-8180] - Updated Python API for all algorithms and AutoML to retrieve the trained model or leader.
    • 🚚 [PUBDEV-8174] - Removed algorithm-specific logic from base classes.
    • πŸ‘ [PUBDEV-8172] - Added support for scoreContributions for imported MOJOs in Java.
    • [PUBDEV-8170] - Exposed AutoML args as writeable properties until first called to train.
    • ⚑️ [PUBDEV-8168] - Updated XGBoost print_mojo to now output weights.
    • 🚚 [PUBDEV-8152] - Removed the Python client dependency on colorama.
    • [PUBDEV-8146] - Added the parameters and their default values to the _init_ function of the Py code generator.
    • [PUBDEV-8114] - Reduced the workspace of the validation frame in GBM by sharing it with the training frame in cross validation.
    • [PUBDEV-8085] - Slightly reduced precision of predictions stored in holdout frames to significantly save on memory.
    • 🚚 [PUBDEV-8015] - Removed warning in the Stacked Ensemble prediction function about missing fold_column frame.
    • [PUBDEV-7958] - Enabled returning data from Explain’s varimp_heatmap and model_correlation_matrix.
    • [PUBDEV-7937] - Exposed the top n and bottom n reason codes in Python/R and MOJO.
    • πŸ— [PUBDEV-5300] - Fixed nightly build version mismatch that prevented the H2OCluster timezone being set to America/Denver.

    πŸ†• New Feature

    • βœ… [PUBDEV-8319] - Implemented a java-self-check to allow users to run on latest Java.
    • ⚑️ [PUBDEV-8312] - Sped up GBM by optimizing the building of histograms.
    • ⚑️ [PUBDEV-8287] - Added a warning to the TreeSHAP reweighting feature if there are 0 weights and updated the API.
    • [PUBDEV-8235] - Added Maximum R Square Improvement (MAXR) algorithm to GLM.
    • ⚠ [PUBDEV-8229] - Added warning for when H2O doesn’t have enough memory to run XGBoost.
    • [PUBDEV-8221] - Added the ability to specify a custom file name when saving a MOJO.
    • πŸ–¨ [PUBDEV-8203] - Added output version number of genmodel.jar when printing usage for PrintMojo.
    • [PUBDEV-8113] - Added MOJO to Rulefit.
    • [PUBDEV-8099] - Implemented ability to calculate Shapley values on a re-weighted tree.
    • [PUBDEV-8088] - Implemented H2O ANOVA GLM algorithm for GLM.
    • [PUBDEV-7354] - Improved and consolidated the handling of version mismatch between Python and Backend.
    • [PUBDEV-7139] - Implemented permutation feature importance for black-box models.
    • [PUBDEV-7138] - Implemented Extended Isolation Forest algorithm.
    • πŸ‘ [PUBDEV-6364] - Added support for saving a model directly to S3.

    Task

    • πŸ”€ [PUBDEV-8292] - Fixed the time limits for the Merge/Sort benchmark.
    • 🚚 [PUBDEV-8197] - Switched removed pandas as_matrix method to .values and exposed the interim pandas.DataFrame object.
    • [PUBDEV-8116] - Fixed S3 credential for pyunit_s3_model_save.py test.
    • [PUBDEV-8084] - Connected XGBoost aggregation functionality with sorting functionality.

    Technical task

    • [PUBDEV-8202] - Replaced subsampling in Extended Isolation Forest.

    πŸ“„ Docs

    • ⚑️ [PUBDEV-8307] - Updated the AutoML FAQ.
    • [PUBDEV-8304] - Corrected the ignored_columns example.
    • πŸ“š [PUBDEV-8299] - Added RMarkdown, Jupyter Notebook, and HTML output example files to H2O Explain documentation.
    • πŸ“š [PUBDEV-8282] - Added Maximum R Improvements (MAXR) GLM documentation.
    • [PUBDEV-8261] - Added the loss function equations for each distribution and link type.
    • πŸ“š [PUBDEV-8248] - Updated the documentation about StackedEnsembles time constraints in AutoML.
    • [PUBDEV-8205] - Clarified that the Explain function only works for supervised models.
    • πŸ“š [PUBDEV-8179] - Added Examine Models section to AutoML documentation.
    • πŸ“š [PUBDEV-8166] - Added documentation for H2O ANOVA GLM algorithm.
    • πŸ“š [PUBDEV-8123] - Fixed the H2O Explain example in the documentation.
    • ⚑️ [PUBDEV-8053] - Updated and gathered Java links to a singular place in the User Guide.
  • v3.32.1.7 Changes

    πŸš€ Download at: http://h2o-release.s3.amazonaws.com/h2o/rel-zipf/7/index.html

    πŸ› Bug

    • πŸ›  [PUBDEV-8234] - Fixed predicting issues with imported MOJOs trained with an offset-column.
    • πŸ— [PUBDEV-8247] - Fixed slow tree building by implementing a switch to turn off the generation of plain language rules.
    • [PUBDEV-8298] - Fixed potential NPE thrown by setting _orig_projection_array=[].
    • πŸ›  [PUBDEV-8309] - Fixed generic model deserialization.
    • πŸ›  [PUBDEV-8310] - Fixed predictions for splits NA vs REST with monotone constraints.

    πŸ†• New Feature

    • [PUBDEV-8293] - H2O Standalone now uses log4j2 as the logger implementation.
  • v3.32.1.6 Changes

    πŸš€ Download at: http://h2o-release.s3.amazonaws.com/h2o/rel-zipf/6/index.html

    πŸ› Bug

    • πŸ‘€ [PUBDEV-8263] - Fixed the POJO mismatch from MOJO and in-H2O scoring for an unseen categorical value.
    • [PUBDEV-8260] - Simplified duplicated XGBoost parameters in Flow.
    • πŸ›  [PUBDEV-8239] - Fixed broken data frame conversion behavior.

    πŸ‘Œ Improvement

    • βž• Added security updates.

    πŸ†• New Feature

    • [PUBDEV-8284] - Exposed the scale_pos_weight parameter in XGBoost.

    Task

    • [PUBDEV-8241] - Clarified the anomaly score formula used for score calculation within Isolation Forest and Extended Isolation Forest.

    πŸ“„ Docs

    • [PUBDEV-8096] - Added a note on memory usage when using XGBoost to User Guide.
  • v3.32.1.5 Changes

    πŸš€ Download at: http://h2o-release.s3.amazonaws.com/h2o/rel-zipf/5/index.html

    πŸ› Bug

    • 🐳 [PUBDEV-8254] - Modified legacy Dockerfile to add a non-root user.
    • πŸ›  [PUBDEV-8253] - Fixed an issue where running java -jar h2o.jar -version failed.
    • πŸ›  [PUBDEV-8250] - Fixed an issue where monotone constraints in GBM caused issues when reproducing the model.
    • πŸ›  [PUBDEV-8246] - Fixed an issue that caused DRF to create incorrect leaf nodes due to rounding errors.
    • πŸ›  [PUBDEV-8244] - Fixed an issue that caused CoxPH MOJO import to fail.
    • πŸ›  [PUBDEV-8242] - Fixed an issue where categorical splits NAvsREST were not represented correctly.
    • πŸ›  [PUBDEV-8240] - Fixed GBM reproducibility for correlated columns with NAs.
    • πŸ›  [PUBDEV-8237] - Fixed h2odriver so that it no longer uses invalid GC options.
    • πŸ›  [PUBDEV-8230] - Fixed GenericModel predictions for non-AUTO categorical encodings.
    • πŸ›  [PUBDEV-8218] - Fixed H2O interaction outcomes.
    • [PUBDEV-8190] - When remove_collinear_columns=True, fixed an issue where the dimension of gradient and coefficients changed when predictors were removed.

    πŸ“„ Docs

  • v3.32.1.4 Changes

    πŸš€ Download at: http://h2o-release.s3.amazonaws.com/h2o/rel-zipf/4/index.html

    πŸ› Bug

    • πŸ›  [PUBDEV-8226] - Fixed h2odriver invalid argument error on Java 11.
    • [PUBDEV-8224] - Fixed GLM GRADIENT_DESCENT_SQERR Solver validation.
    • ⬆️ [PUBDEV-8219] - Upgraded to latest version of Javassist (3.28).
    • πŸ›  [PUBDEV-8207] - Fixed H statistic gpu assertion error.
    • πŸ›  [PUBDEV-8195] - Fixed predict contributions failure in multi-MOJO environments.
    • πŸ›  [PUBDEV-8194] - Fixed bug in ordinal GLM class predictions.
    • πŸ›  [PUBDEV-8188] - Fixed Partial Dependent Plot not working with Flow.
    • ⚑️ [PUBDEV-8181] - Updated to current Python syntax.
    • πŸ›  [PUBDEV-8167] - Fixed bug in ordinal GLM class predictions.

    πŸ‘Œ Improvement

    • πŸ‘ [PUBDEV-8141] - Added support for refreshing HDFS delegation tokens for standalone H2O.

    πŸ†• New Feature

    • [PUBDEV-8109] - Obtained Friedman’s H statistic for XGBoost and GBM.

    Task

    • ⚠ [PUBDEV-8150] - Added a warning message when using alpha as a hyperparameter for GLM

    πŸ“„ Docs

    • [PUBDEV-8158] - Added section on how to delete objects in Flow.
    • πŸ“„ [PUBDEV-8151] - Added a note to the productionizing docs that C++ is only available with additional support.
  • v3.32.1.3 Changes

    πŸš€ Download at: http://h2o-release.s3.amazonaws.com/h2o/rel-zipf/3/index.html

    πŸ› Bug

    • [PUBDEV-8136] - Fixed the printing for auc_pr and pr_auc in cross-validation summaries.

    πŸ†• New Feature

    • 🐎 [PUBDEV-8131] - Added parameter auc_type to performance method to compute multiclass AUC.

    Task

    • ⬆️ [PUBDEV-8147] - Upgraded XGBoost predictor to 0.3.18.
    • πŸ“¦ [PUBDEV-8145] - Increased the timeout duration on the R package jar download.

    πŸ“„ Docs

    • πŸ— [PUBDEV-8119] - Fixed formatting errors for local builds.
    • ⚑️ [PUBDEV-8091] - Updated docs examples for baseline hazard, baseline survival, and concordance.
  • v3.32.1.2 Changes

    πŸš€ Download at: http://h2o-release.s3.amazonaws.com/h2o/rel-zipf/2/index.html

    πŸ› Bug

    [PUBDEV-7842] - Stacked Ensemble will no longer ignore a column if any base model uses it. [PUBDEV-7851] - Added a user-friendly reminder that the new explainability functions require newer versions of ggplot2 in R. [PUBDEV-7948] - NullPointerException error no longer thrown when used a saved and reloaded RuleFit model. [PUBDEV-7955] - Can now extract metrics from the validation dataset with a Rulefit Model. βœ… [PUBDEV-8076] - Fixed failures from Stacked Ensemble with Multinomial GLM within tests. πŸ›  [PUBDEV-8077] - Fixed AutoML error when an alpha array is used for GLM. πŸ›  [PUBDEV-8079] - Fixed β€œRollup not possible" stats failure in GLM. [PUBDEV-8097] - H2O will now still start despite system properties that begin with β€˜ai.h2o.’. 🌲 [PUBDEV-8098] - H2O exits without logging any buffered messages instead of throwing a NullPointerException when starting H2O with an invalid argument. [PUBDEV-8100] - ModelDescriptor field in MOJO is now Serializable. πŸ— [PUBDEV-8102] - AutoML no longer crashes if model builder produces H2OIllegalArgumentException in the parameter validation phase. [PUBDEV-8106] - Weights in GLM grid search is no longer used as features. πŸ›  [PUBDEV-8120] - Fixed Stacked Ensemble MOJO for cases when sub-model doesn’t have the same columns as the metalearner. [PUBDEV-8125] - Efron-method now fully deterministic in CoxPH.

    πŸ‘Œ Improvement

    πŸ“œ [PUBDEV-8087] - User now allowed to specify the escape character for parsing CSVs. [PUBDEV-8092] - Added H2O reconnection script for intermittent 401 errors to R. πŸ™‹ [PUBDEV-8101] - Added β€˜ice_root’ error documented in FAQ. [PUBDEV-8118] - Added further regularization to the GLM metalearner.

    πŸ†• New Feature

    [PUBDEV-6249] - Warning now issued against irreproducible model when early stopping is enabled but neither score_tree_interval or score_each_iteration are defined. [PUBDEV-8023] - Encrypted files that contain CSVs can now be imported. [PUBDEV-8072] - Added guidelines for correct use of remove_collinear_columns for GLM. πŸ‘ [PUBDEV-8111] - Support added for CDP 7.2.

    πŸ“„ Docs

    [PUBDEV-8067] - Added information about the path argument for exporting .xlsx files.

  • v3.32.1.1 Changes

    πŸš€ Download at: http://h2o-release.s3.amazonaws.com/h2o/rel-zipf/1/index.html

    πŸ› Bug

    [PUBDEV-6356] - GBM histograms now ignore rows with NA responses. [PUBDEV-7606] - Variable Importances added to GLM Generic model. πŸ›  [PUBDEV-7782] - Fixed the ArrayIndexOutOfBoundsException issue with GLM CV. 🐎 [PUBDEV-7825] - CoxPH performance no longer fails when a factor is used for the event_column. [PUBDEV-7841] - Existing frame no longer overwritten when data with the same query is loaded. πŸ›  [PUBDEV-7909] - Fixed how gain is calculated in XGBFI for GBM. [PUBDEV-7934] - Improved the error messages for save_to_hive_table. βœ… [PUBDEV-7963] - Added missing argument ’test’ for h2o.explain_row(). πŸ–¨ [PUBDEV-7979] - All trees now supported for XGBoost Print MOJO in Java. [PUBDEV-7987] - CoxPH prediction no longer fails when offset_column is specified. [PUBDEV-7998] - Added keys for Individual Conditional Expectation (ICE) plot in H2OExplanation class. [PUBDEV-8013] - [email protected]$parameters$x now reports actual feature names instead of names. [PUBDEV-8016] - h2o.explain no longer errors when AutoML object is trained with a fold_column. πŸ›  [PUBDEV-8046] - Fixed issues with python’s explanation plots not displaying fully.

    πŸ†• New Feature

    [PUBDEV-7706] - Ignored columns that are actually used for model training are unignored and no longer prevent model training to start in Flow. [PUBDEV-7735] - Added baseline hazard function estimate to CoxPH model. πŸ‘ [PUBDEV-7748] - Target Encoding now supports feature interactions. [PUBDEV-7805] - Added CoxPH concordance to both Flow and R/Python CoxPH summaries. [PUBDEV-7820] - Added a topbasemodel attribute to AutoML. [PUBDEV-7831] - Added new learning curve plotting function to R/Python. [PUBDEV-7854] - Added script for estimating the memory usage of a dataset. [PUBDEV-7859] - Added fault protections to grid search allowing saving of data and parameters, model checkpointing, and auto-recovery. πŸ‘ [PUBDEV-7884] - Added support for Java 15. πŸ‘ [PUBDEV-7969] - Added CDP7.1 support. πŸ–¨ [PUBDEV-7978] - Added support for XGBoost to Print MOJO as JSON. πŸ‘ [PUBDEV-8021] - Added support for refreshing HDFS delegation tokens. βͺ [PUBDEV-8035] - Reverted XGBoost categorical encodings for contributions.

    Task

    [PUBDEV-7637] - max_hit_ratio_k deprecated and removed. πŸ“¦ [PUBDEV-7894] - Added upper bound cap to supported Java version in H2O CRAN package requirements.

    πŸ‘Œ Improvement

    [PUBDEV-7473] - Users now allowed to include categorical column name in beta constraints. [PUBDEV-7579] - Multinomial PDP can now be plotted for more than one target class in Flow. [PUBDEV-7736] - Sped up CoxPH concordance score by using tree instead of the direct approach. [PUBDEV-7819] - XGBoost no longer fails when specifying custom fold_column. [PUBDEV-7843] - XGBoost CV models now built on multiple GPUs in parallel. [PUBDEV-7968] - Missing metrics added to GLM scoring history. [PUBDEV-8017] - Added validation checks for sampling rates for XGBoost for the R/Python clients. [PUBDEV-8024] -
    No longer errors when trying to use a fold column where not all folds are represented. [PUBDEV-8032] - Added the metalearner_transform option to Stacked Ensemble. [PUBDEV-8057] - GBM main model now built in parallel to the CV models. 🚚 [PUBDEV-8060] - Removed redundant extraction weights from GBM/DRF histogram. [PUBDEV-8061] - GBM now avoids scoring the last iteration twice when early stopping is enabled. [PUBDEV-8063] - POJO predictions for XGBoost now even closer to in-H2O predictions. [PUBDEV-8064] - Double-scoring of CV models in AutoML now avoided thus speeding up AutoML. [PUBDEV-8070] - AutoML now uses fewer neurons in DL grids and has improved the metalearner for Stacked Ensemble.

    Technical task

    [PUBDEV-7860] - Thin plate regression splines added to GAM.

    πŸ“„ Docs

    [PUBDEV-7917] - Added checkpoint description to GLM. πŸ“š [PUBDEV-7976] - Added thin plate regression spline documentation to GAM algorithm page. [PUBDEV-7988] - Added missing parameters to XGBoost algorithm page. 🌲 [PUBDEV-7992] - Added more information about log files to User Guide.

  • v3.32.0.5 Changes

    πŸš€ Download at: http://h2o-release.s3.amazonaws.com/h2o/rel-zermelo/5/index.html

    πŸ› Bug

    [PUBDEV-7798] - GAM no longer creates multiple knots at the same coordinates when the cardinality of the gam_columns is less than the number of knots specified by the user.

    πŸ‘Œ Improvement

    [PUBDEV-7954] -
    πŸ”‹ Feature interactions can now be save as .xlxs files. πŸ‘· [PUBDEV-8034] - Job polling will retry connecting to h2o nodes if connection fails.

  • v3.32.0.4 Changes

    πŸš€ Download at: http://h2o-release.s3.amazonaws.com/h2o/rel-zermelo/4/index.html

    πŸ› Bug

    [PUBDEV-7949] - Partial Dependence Plot no longer failing for High Cardinality even when user_splits is defined. πŸ›  [PUBDEV-7951] - Fixed failing Delta Lake import for Python API. [PUBDEV-7962] - Fix Stacked Ensemble’s incorrect handling of fold column.

    πŸ‘Œ Improvement

    πŸ‘ [PUBDEV-7737] - Added MOJO support for CoxPH. 0️⃣ [PUBDEV-7953] - Escape all quotes by default when writing CSV.

    πŸ“„ Docs

    πŸ“„ [PUBDEV-7945] - Added to docs that AUCPR can be plotted. ⚑️ [PUBDEV-7964] - Updated the Customer Algorithm graphic for the Architecture section of the User Guide. ⚑️ [PUBDEV-7983] - Updated the copyright year to 2021.