Skip to content

Latest commit

 

History

History
910 lines (638 loc) · 66 KB

NEWS.md

File metadata and controls

910 lines (638 loc) · 66 KB

datarobot v2.29.0.9002

The datarobot package is now dependent on R >= 3.5.

The datarobot package now covers the entirety of the DataRobot Public API. It is now dependent on the datarobot.apicore package, which provides auto-generated functions to access the Public API. The datarobot package provides a number of "API wrapper functions" around the apicore package to make it easier to use.

DataRobot recommends starting with package documentation: try ?datarobot and ?datarobot.apicore in your R session. Also, take a look at the vignettes.

New API Functions:

  • Generated API wrapper functions are organized into categories based on their tags from the OpenAPI specification, which were themselves redone for the entire DataRobot Public API in v2.27.
  • These functions use camel-cased argument names, to be consistent with the rest of the package.
  • Most function names follow a VerbObject pattern based on the OpenAPI specification.
  • Some function names match "legacy" functions that existed in v2.18 of the R Client, if they invoked the same underlying endpoint. For example, the wrapper function is called GetModel, not RetrieveProjectsModels, since the latter is what was implemented in the R client for the endpoint /projects/{mId}/models/{mId}.
  • Similarly, these functions use the same arguments as the corresponding "legacy" functions, to ensure that DataRobot does not break existing code that calls those functions.

Other New Features:

  • The R client (both datarobot and datarobot.apicore packages) will output a warning when you attempt to access certain resources (projects, models, deployments, etc.) that are deprecated or disabled by the DataRobot platform migration to Python 3.
  • Added the helper function EditConfig that allows you to interactively modify drconfig.yaml.
  • Added the DownloadDatasetAsCsv function to retrieve dataset as CSV using catalogId.
  • Added the GetFeatureDiscoveryRelationships function to get the feature discovery relationships for a project.
  • Added support for comprehensive autopilot: use mode = AutopilotMode.Comprehensive.

Enhancements:

  • The function RequestFeatureImpact now accepts a rowCount argument, which will change the sample size used for Feature Impact calculations.
  • The internal helper function ValidateModel was renamed to ValidateAndReturnModel and now works with model classes from the apicore package.

Bugfixes:

  • The enum ModelCapability has been properly exported.
  • Fixed FullAverageDataset function in the PartialDependence vignette to ignore NA when calculating the min and max of the data range.
  • Fixed RetrieveAutomatedDocuments function to accept filename argument that is used to specify where to save the automated document.
  • Fixed datarobot.apicore file upload functions to properly encode the payload as "multipart".
  • Fixed datarobot.apicore JSON serialization bugs.
  • Fixed some tests exercising the BuildPath helper function.

API Changes:

  • The helper function isApicoreModel no longer checks for datarobot.apicore::FrozenModelRetrieveResponse since it no longer exists in the Public API. It was replaced by the class datarobot.apicore::ModelDetailsResponse.
  • The functions ListProjects and as.data.frame.projectSummaryList no longer return fields related to recommender models, which were removed in v2.5.0.
  • The function SetTarget now sets autopilot mode to Quick by default. Additionally, when Quick is passed, the underlying /aim endpoint will no longer be invoked with Auto.
  • The functions CreateDatasetsFromHDFS, CreateDatasetsVersionsFromHDFS, and CreateHdfsProjects have been deprecated.

Deprecated and Defunct:

  • quickrun argument is removed from the function SetTarget. Users should set mode = AutopilotMode.Quick instead.
  • Compliance Documentation got deprecated in favor of Automated Documentation API.
  • The Transferable Models family of functions (ListTransferableModels, GetTransferableModel, RequestTransferableModel, DownloadTransferableModel, UploadTransferableModel, UpdateTransferableModel, DeleteTransferableModel) have been removed. The underlying endpoints -- long deprecated -- were removed from the Public API with the removal of the Standalone Scoring Engine (SSE).
  • Removed files (code, tests, doc) representing parts of the Public API not present in v2.27-2.29.

Dependency Changes:

  • The datarobot package is now dependent on R >= 3.5 due to changes in the updated "Introduction to DataRobot" vignette.
  • Added dependency on AmesHousing package for updated "Introduction to DataRobot" vignette.
  • Removed dependency on MASS package.
  • Removed dependency on R6 package; it is already a dependency of datarobot.apicore but is not used in datarobot itself.
  • Client documentation is now explicitly generated with Roxygen2 v7.2.1.

Documentation Changes:

  • Package-level documentation for both packages has been updated to explain how to use package options.
  • Updated "Introduction to DataRobot" vignette to use Ames, Iowa housing data instead of Boston housing dataset.
  • Compressed extdata/Friedman1.csv and updated vignettes dependent on that dataset.
  • Removed extdata/anomFrame.csv as it was unused.

datarobot v2.27.0.9000

This was a Private Preview release of the R API Client. Earlier preview releases of this package were versioned v3.0.0 but are subsumed by this one.

datarobot v2.18.1

New Features:

  • Added the ability to retrieve and restore features that have been reduced using the time series feature generation and reduction functionality. Discarded features can be retrieved using a RetrieveDiscardedFeaturesInformation and restored using RestoreDiscardedFeatures
  • Added the ability to create and retrieve DataEngineQueryGenerators and create a Dataset from a DataEngineQueryGenerator for time series data prep.
  • Added support to upload a prediction dataset from the AI catalog.
  • Added an ability to calculate and retrieve Datetime trend plots for datetime aware model. This includes Accuracy over Time, Forecast vs Actual, and Anomaly over Time.
    • Plots can be calculated using a common ComputeDatetimeTrendPlots function.
    • Metadata for plots can be retrieved using the GetAccuracyOverTimePlotsMetadata, GetForecastVsActualPlotsMetadata, and GetAnomalyOverTimePlotsMetadata functions.
    • Plots can be retrieved using the GetAccuracyOverTimePlot, GetForecastVsActualPlot, and GetAnomalyOverTimePlot functions.
    • Preview plots can be retrieved using the GetAccuracyOverTimePlotPreview, GetForecastVsActualPlotPreview, and GetAnomalyOverTimePlotPreview functions.

Enhancements:

  • GetProject now returns all available output from the API.
  • Many functions no longer use the internal function ApplySchema to filter a response from the DataRobot API to a given schema. This ensures that we return all information the API provides whenever necessary. Only a handful of functions, such as IsBlenderEligible still use this function where it is primarily useful.

Bugfixes:

  • The function GetBlenderModelFromJobId now returns an object with the field modelId rather than id, consistent with GetBlenderModel.

API Changes:

Deprecated and Defunct:

  • The deprecated BlendMethod$FORECAST_DISTANCE has been removed. Use BlendMethod$FORECAST_DISTANCE_ENET instead.

Dependency Changes:

  • Client documentation is now explicitly generated with Roxygen2 v7.1.2.

Documentation Changes:

datarobot v2.18.0

This release brings the R Client to parity with DataRobot API v2.18 (DataRobot 5.2), but also includes a number of features from API v2.19 (DataRobot 5.3) as well as Anomaly Assessment, a DataRobot 7.1 feature.

New Features:

  • Residuals Chart data for models can be retrieved using GetLiftCharts and GetAllLiftCharts functions. This is valid only for regression models that are not time-aware.
  • Added "Average by Forecast Distance" blender for time series projects configured with more than one Forecast Distance. The blender blends the selected models, selecting the best 3 models based on the backtesting score for each Forecast Distance and averaging their predictions. The new blender method FORECAST_DISTANCE_AVG has been added as BlendMethod$FORECAST_DISTANCE_AVG.
  • Added functions for tracking the health and status of a deployment. GetDeploymentServiceStats retrieves metrics that track deployment utilization and performance, while GetDeploymentAccuracy retrieves metrics that track the accuracy of a deployment's predictions. GetDeploymentServiceStatsOverTime and GetDeploymentAccuracyOverTime will track changes to those metrics over a specified time interval.
  • SubmitActuals can now be used to submit data about actual results from a deployed model, which can be used to calculate accuracy metrics.
  • Projects can be cloned using CloneProject. The clone will be post-EDA1 and ready for setting targets and modeling options.
  • CreateCalendar now supports series-specific events via the multiSeriesIdColumn argument. An example of a series-specific event: some but not all stores being affected by a holiday.
  • GetDeploymentAssociationId and UpdateDeploymentAssociationId can be used to manage a deployment's association ID for use with SubmitActuals and the Deployment Accuracy functions.
  • GetDeploymentSettings can be used to retrieve any and all settings related to a deployed model. UpdateDeploymentSettings will allow you to make piecemeal changes as well. The convenience functions GetDeploymentDriftTrackingSettings and GetDeploymentAssociationId use these methods internally.
  • Time series model exports also support prediction intervals: RequestTransferableModel now has a predictionIntervalsSize parameter.
  • Added support for Anomaly Assessment insight. This insight is available for anomaly detection models in time series unsupervised projects which also support calculation of Shapley values. It is possible to:
    • InitializeAnomalyAssessment initializes an anomaly assessment insight for the specified subset
    • ListAnomalyAssessmentRecords retrieves records
    • GetAnomalyAssessmentExplanations retrieves shap explanations
    • GetAnomalyAssessmentPredictionsPreview retrieves predictions preview
    • DeleteAnomalyAssessmentRecord deletes records

Enhancements:

  • Monotonic constraints are now supported for OTV projects. To that end, the parameters monotonicIncreasingFeaturelistId and monotonicDecreasingFeaturelistId can be specified in calls to RequestNewDatetimeModel.
  • You can now get the model associated with a model job by getting the modelId field on the GetModelJob or ListModelJobs response objects.
  • Added the new field recommendedFeaturelistId to the Blueprint response object. If absent, there is no recommended feature list for this blueprint.
  • The Model S3 class now exposes the modelNumber field. This field is also exposed in the responses to GetFrozenModel, GetDatetimeModel, GetBlenderModel, and GetRatingTableModel.
  • The method GetModelCapabilities has been extended to return supportsCodeGeneration, supportsShap, and other newly-added capabilities. See ModelCapability for more details.
  • GetFeatureInfo will now return descriptive statistics on summarized categorical features in the field keySummary.
  • ListDeployments now supports sorting and searching the results using the new orderBy and search parameters.
  • GetResidualsChart and ListResidualsCharts are now backwards-compatible with DataRobot 5.2, which does not return rowNumber.
  • GetWordCloud now includes a variable field that represents the source of each ngram, as well as a class field that represents values of the target class.
  • Performance improvements for GetPredictions and Predict when retrieving probabilities for large prediction datasets on multiclass projects, i.e. Predict(irisModel, largeDataset, type = "probability")
  • Unit tests can now be written against testthat edition 3. This is an opt-in feature; all tests are run against edition 2 by default.

Bugfixes:

  • Calls to ListDeployments will now return more than 20 deployments when available.
  • ListPrimeModels now returns an empty data frame when the API returns zero results, consistent with its documentation. Previously it would return an empty list. This response is also classed as dataRobotPrimeModels.
  • CreateCalendar now terminates properly when DataRobot is unable to create the calendar. Previously, it would hang due to the R package not checking for the right error response.
  • formatRFC3339Timestamp now works for vectors of length > 1.

API Changes:

  • The first argument of CreateCalendar and CreateRatingTable is changed from file to dataSource to reflect that the functions can process data frames as well as CSV files.
  • The helper method ProjectFromAsyncUrl is replaced with ProjectFromJobResponse; this change allowed us to simplify the package's dependency on httr.

Deprecated and Defunct:

  • BlendMethod$FORECAST_DISTANCE is deprecated and will be removed in 2.19. Use BlendMethod$FORECAST_DISTANCE_ENET instead.

Dependency Changes:

  • Client documentation is now explicitly generated with Roxygen2 v7.1.1.
  • To support new features, testthat@>3.0.0 and devtools@>2.4.0 is now required. The test suites are being updated to meet testthat 3e requirements.
  • Removed Suggests: rex as it is no longer needed for package development.

Documentation Changes:

  • This NEWS file was renamed to NEWS.md and formatted as Markdown.
  • Added unit test guidelines for developers to the README.
  • Documentation for GetBlenderModel and GetBlenderModelFromJobId are now more consistent.
  • Parameter documentation for StartAutopilot and SetTarget is clarified.
  • Organized some functions into families for easier reference.
  • Tweaked documentation related to predictions and time series projects.
  • Fixed some spelling mistakes, typos, and Roxygen syntax errors.

datarobot v2.17.1

  • Removed dependency on V8 package by removing code that used the colormap package. The V8 package was flagged by the CRAN maintainers as not building so this removal was necessary to keep the datarobot package on CRAN.
  • Removed curl from Imports since it was causing a NOTE when devtools::check_win_devel() was run.

datarobot v2.17.0

New Features:

  • You can now deploy models via the API! Use CreateDeployment to create a deployment against a particular prediction server. Use ListPredictionServers to list all the available prediction servers. Use GetDeployment and ListDeployments to see particular deployments that you have. You can delete a deployment with DeleteDeployment.
  • The model backing a deployment can be replaced with ReplaceDeployedModel. Use ValidateReplaceDeployedModel first to test that the deployment replacement is valid, if desired.
  • Deployments support drift tracking. Use GetDeploymentDriftTrackingSettings to get drift tracking settings for a deployment. You can update the drift tracking using UpdateDeploymentDriftTrackingSettings.
  • Information on feature clustering and the association strength between pairs of numeric or categorical features is now available with GetFeatureAssociationMatrix. Relative pairwise feature association statistics can be retrieved with GetFeatureAssociationMatrixDetails.
  • Multiple feature type transformations can now be executed in a single batch request using BatchFeaturesTypeTransform.

Enhancements:

  • You can now use doNotDerive in the featureSettings of CreateDatetimePartition to disable DataRobot's automatic time series feature engineering for a particular feature (e.g., so you can derive lags yourself manually).
  • Users can now embed DataRobot-generated content in compliance doc templates (see UploadComplianceDocTemplate) using keyword tags.
  • Prediction intervals are now supported for start-end retrained models in a time series project.
  • Previously, all backtests had to be run before prediction intervals for a time series project could be requested with predictions. Now, backtests will be computed automatically if needed when prediction intervals are requested.

Bugfixes:

  • Calls to GetPredictionExplanationsRowsAsDataFrame previously did not work with numeric labels. This has been fixed.

API Changes:

Deprecated and Defunct:

  • The defaultToAPriori parameter in CreateDatetimePartitionSpecification has been renamed to defaultToKnownInAdvance. defaultToAPriori has now been fully removed.
  • The aPriori flag in the featureSettings parameter in CreateDatetimePartitionSpecification as been renamed to knownInAdvance. aPriori has now been fully removed.
  • The deprecated SetupProjectFromMySQL, SetupProjectFromOracle and SetupProjectFromPostgreSQL have now been removed. Use SetupProjectFromDataSource instead.
  • GetTransferrableModel, ListTransferrableModels, UpdateTransferrableModel, DeleteTransferrableModel, DownloadTransferrableModel, and UploadTransferrableModel have been removed and replaced with their correctly spelled counterparts (GetTransferableModel, ListTransferableModels, UpdateTransferableModel, DeleteTransferableModel, DownloadTransferableModel, and UploadTransferableModel).

Dependency Changes:

Documentation Changes:

datarobot v2.16.0

New Features:

  • You can now retrieve series accuracy information, showing accuracy metrics for each series for a multiseries project. Use GetSeriesAccuracy to retrieve the accuracy. You can also download it as a CSV with DownloadSeriesAccuracy.

Enhancements:

  • Prediction intervals can now be returned for predictions with datetime models. Use includePredictionIntervals = TRUE in calls to Predict. For each model, prediction intervals estimate the range of values DataRobot expects actual values of the target to fall within. They are similar to a confidence interval of a prediction, but are based on the residual errors measured during the backtesting for the selected model.
  • ListPredictions now returns metadata on prediction intervals. includesPredictionIntervals is TRUE if there are prediction intervals in the predictions and FALSE otherwise. predictionIntervals specifies the size (in percent) of intervals or is NULL if there are no intervals.
  • For time series projects, the effective Feature Derivation Window, specifying the full span of historical data required at predict time, is now available through the API. It may be longer than the feature derivation window of the project depending on the differencing settings used.
  • More of the project partitioning settings are also available in the metadata for datetime models (see GetDatetimeModel). The new attributes are effectiveFeatureDerivationWindowStart, effectiveFeatureDerivationWindowEnd, forecastWindowStart, forecastWindowEnd, and windowsBasisUnit.
  • DownloadComplianceDocumentation and GetSeriesAccuracy now support a maxWait parameter to customize the amount of time to wait before raising a timeout error.

Deprecated and Defunct:

  • RecommendedModelType$Recommended type for GetModelRecommendation and GetRecommendedModel has been removed and replaced with RecommendedModelType$RecommendedForDeployment.

Documentation Changes:

  • Fixed more spelling mistakes in the documentation.

datarobot v2.15.0

New Features:

  • Advanced tuning can now be done on any model. See StartTuningSession for details.
  • DataRobot time series now supports calendar files, which allow specifying special events like holidays. See CreateCalendar, GetCalendar, ListCalendars, UpdateCalendar, and DeleteCalendar.
  • Projects now can be shared with other users. See Share for details.
  • Calendars can be shared with other users. See Share for details.

Enhancements:

  • UploadPredictionDataset and UploadPredictionDatasetFromDataSource will now return dataQualityWarnings that mention any potential problems with the uploaded dataset.
  • UploadPredictionDataset and UploadPredictionDatasetFromDataSource now have a parameter relaxKIAFeaturesCheck. If TRUE, uploaded datasets for time series projects will allow missing values for the Known in Advance features in the forecast window at prediction time.
  • ROC Curve information retrieval has been extended to contain four new fields (fractionPredictedAsPositive, fractionPredictedAsNegative, liftPositive, and liftNegative) with cumulative gains and lift data.
  • Added Forecast Distance blender for time series projects configured with more than one Forecast Distance. It blends the selected models creating separate linear models for each Forecast Distance.
  • Added a filter option to ListProjects that supports filtering retrieval of project lists by name using the projectName filter.
  • GetCalendarFromProject can be used to get the calendar associated with a project.
  • Data source objects can now be used in StartProject to quickly create a project from a data source.
  • Data source objects can now be used in addition to data source IDs in SetupProjectFromDataSource.
  • crossSeriesGroupByColumns has been added to datetime partitioning to allow users the ability to indicate how to further split series in to related groups.
  • The prediction explanations workflow is now ~3x faster for most use cases.

Bugfixes:

  • Time series windowBasisUnit has been renamed to the correct windowsBasisUnit.

Deprecated and Defunct:

  • Reason codes have been renamed to Prediction Explanations to provide increased clarity and accessibility. DeleteReasonCodes, DeleteReasonCodesInitialization, DownloadReasonCodes, GetAllReasonCodesRowsAsDataFrame, GetReasonCodesInitialization, GetReasonCodesInitializationFromJobId, GetReasonCodesMetadata, GetReasonCodesMetadataFromJobId, GetReasonCodesRows, ListReasonCodesMetadata, RequestReasonCodes, and RequestReasonCodesInitialization have all been removed and replaced with appropriately renamed functions and a new workflow. See GetPredictionExplanations for more.
  • SetupProjectFromMySQL, SetupProjectFromOracle, SetupProjectFromPostgreSQL, SetupProjectFromHDFS are now deprecated. They will be removed in v2.17. Use SetupProjectFromDataSource instead.
  • RequestPredictionsForDataset has been renamed to RequestPredictions. The original RequestPredictionsForDataset has been removed.
  • GetDatetimeModelObject has been renamed to GetDatetimeModel. The original GetDatetimeModelObject has been removed.
  • The defaultToAPriori parameter in CreateDatetimePartitionSpecification has been renamed to defaultToKnownInAdvance. defaultToAPriori is now removed.
  • The aPriori flag in the featureSettings parameter in CreateDatetimePartitionSpecification as been renamed to knownInAdvance. aPriori is now removed.
  • GetTransferrableModel, ListTransferrableModels, UpdateTransferrableModel, DeleteTransferrableModel, DownloadTransferrableModel, and UploadTransferrableModel have all been deprecated and replaced with their correctly spelled counterparts (GetTransferableModel, ListTransferableModels, UpdateTransferableModel, DeleteTransferableModel, DownloadTransferableModel, and UploadTransferableModel). The misspelled versions will be removed in v2.17.
  • Support for numeric modes in StartAutopilot has now been fully removed.

Dependency Changes:

  • To support new features, curl at version 2.7 or higher is now required.

Documentation Changes:

  • Fixed more spelling mistakes in the documentation.

datarobot v2.14.2

Enhancements:

  • Training predictions for multiseries projects will now return the SeriesID, forecastPoint, and forecastDistance.
  • GetDatetimePartition now returns isCrossSeries to indicate whether the datetime partition uses cross-series features.
  • ScoreBacktests now accepts a parameter wait = TRUE to wait for job completion.
  • Predict and GetPredictions no longer return positiveProbability for non-binary problems.
  • Predict and GetPredictions no longer return seriesId for non-multiseries problems.

Documentation Changes:

  • Fixed a typo in how knownInAdvance was defined in the featureSettings in the "time series" vignette.

datarobot v2.14.1

Bugfixes:

  • Requesting a multiseries project now will work even if the Series ID cannot be automatically inferred by DataRobot.

datarobot v2.14.0

New Features:

  • DownloadComplianceDocumentation can be used to download compliance documentation. Compliance documentation also can be created with default or custom templates - use GetComplianceDocTemplate to get particular templates and UploadComplianceDocTemplate to use your own. See the vignette on "Compliance Documentation" for more information.
  • Data sources and data stores can now be shared with other users. Use Share to share a data source or data store. Use ListSharingAccess to see current access rights. Use UpdateAccess for more complex access right modification operations.
  • Multiseries projects can now include derived cross series features. Use useCrossSeries = TRUE in CreateDatetimePartitionSpecification to enable.
  • You can now get a feature histogram (a histogram of feature counts and target distribution over bins of values for a particular feature) using GetFeatureHistogram.
  • Get supported capabilities for a model using GetModelCapabilities.

Enhancements:

  • Data sources and data stores can be passed into functions directly in addition to being passed as IDs.
  • Binary classification for time series is now supported as a project type.
  • Calls to StartProject and UpdateProject that up the worker count can now set the worker count to "max", which uses the maximum available number of workers.
  • fallbackToParentInsights is now available as a parameter on all insights functions (GetRocCurve, ListRocCurves, GetLiftChart, ListLiftCharts, GetConfusionChart, ListConfusionCharts). When TRUE, a frozen model with missing insights will attempt to retrieve the missing insight data from its parent model.
  • Time series partitions can now have the forecast window and feature derivation windows defined in a number of rows by using windowBasisUnit and setting it to "ROW".
  • Time series partitions can now be defined in millisecond intervals.
  • Training predictions for datetime partitioned projects now support the new data subset DataSubset$AllBacktests for requesting the predictions for all backtest validation folds.
  • Training predictions for datetime partitioned projects now return the relevant timestamp associated with the prediction.

Bugfixes:

  • In cases where a request would return hundreds of responses, sometimes not all those responses would be returned due to improper pagination. This has now been fixed.
  • Variables can now be correctly used as tuning parameters for StartTuningSession.
  • If you use StartProject without defining a project name, it now correctly uses the name of the variable passed (like SetupProject) rather than erroneously just calling it "dataSource".

Deprecated and Defunct:

  • GetAllLiftCharts has now been removed (use ListLiftCharts instead).
  • GetModelJobs has now been removed (use ListModelJobs instead).
  • GetProjectList has now been removed (use ListProjects instead).
  • GetAllRocCurves has now been removed (use ListRocCurves instead).
  • RecommendedModelType$Recommended type for GetModelRecommendation and GetRecommendedModel has been deprecated and replaced with RecommendedModelType$RecommendedForDeployment. It will be removed in v2.16.
  • PeriodicityTimeUnits has been renamed to TimeUnits. PeriodicityTimeUnits still exists for backwards compatibility.

Documentation Changes:

  • Added more documentation for various enums.
  • The use of Predict in the "Introduction to DataRobot" vignette was previously inaccurate. It has been fixed.
  • RecommendedModelType and GetModelRecommendation now have more documentation about the model recommendation process.
  • Vignettes have been updated to use StartProject throughout.
  • The intro vignette has been cleaned up and now has an example of using feature impact.
  • Data used in vignettes has now been added to package data. Broken references to data have been fixed.

datarobot v2.13.0

New Features:

  • An API for advanced tuning is now available, which allows you to manually set model parameters and override the DataRobot default selections. You can get information on available model hyperparameters via GetTuningParameters and start a tuning session with StartTuningSession. You can interactively iterate through all the parameters for a model using RunInteractiveTuning. These advanced tuning features are currently generally available for Eureqa models. To use this feature with other model types, contact your CFDS for more information.
  • Predict can now be used to create predictions directly from a model and a test dataset, bypassing the need to UploadPredictionDataset, RequestPredictions, and GetPredictions.
  • ListPredictions can be used to summarize all the different predictions available for a particular project, model, and/or prediction dataset.
  • GetPredictionExplanations now supports a single workflow to get prediction explanations (previously called reason codes, see "Deprecated and Defunct") for a model and a dataset without needing all the various intermediary steps.
  • GetPredictionDataset can be used to get metadata on a particular prediction dataset.
  • The prediction threshold for binary classification models can now be changed via SetPredictionThreshold.
  • GetFeatureImpact works like GetFeatureImpactForModel, but will also request the feature impact if it has not already been requested.
  • GetTrainingPredictionsForModel retrieves training predictions for a given model object, requesting them in the process.
  • Models can now be starred, which highlights them. StarModel will star a model. UnstarModel will unstar it. ToggleStarForModel will toggle the star status. ListStarredModels will list all the starred models for a particular project. Model objects will also have an isStarred parameter returned to tell whether they are starred or not. (All models are unstarred by default.)
  • DeleteFeaturelist and DeleteModelingFeaturelist can now be used to delete featurelists and modeling featurelists respectively.
  • UpdateFeaturelist can be used to change the name and description of a featurelist. UpdateModelingFeaturelist works for modeling featurelists as well.

Enhancements:

  • Using type = "raw" in GetPredictions (or Predict) will return the raw dataframe of predictions metadata.
  • ListModels now can take an orderBy parameter to sort the output list by metric or samplePct.
  • ListModels now can take a filter parameter to filter output by samplePct, name, and/or isStarred.
  • StartProject and SetupProject can now work without a projectName
  • StartProject can now take a workerCount parameter to set the worker count for the project.
  • StartProject can now take wait = TRUE to automatically wait for the autopilot to complete (thus making an explicit call to WaitForAutopilot unnecessary).
  • ProjectStage can now be used to get a list of the available project stages.
  • It is now no longer necessary to call RequestMultiSeriesDetection manually for a multiseries project.
  • Feature impact now returns not only the impact score for the features but also whether they were detected to be redundant with other high-impact features.
  • Jobs now report a parameter called isBlocked that specifies whether a job is blocked from execution because one or more dependencies have not yet been met.
  • ListModelJobs now returns the trainingRowCount key.
  • GetPredictions now can get predictions using a projectId and a predictionId (see ListPredictions) in addition to its prior ability to retrieve predictions using a predictionJobId.
  • Featurelists (see GetFeaturelist and GetModelingFeaturelist) now return a created value with the timestamp, a isUserCreated value explaining whether or not the feature was created by a user (as opposed to DataRobot automation), numModels showing how many models use the featurelist, and description which gives a text description of the featurelist.
  • Prediction datasets are now dataRobotPredictionDataset class in addition to being list class.
  • GetPredictions now can get predictions using a projectId and a predictionId (see ListPredictions) in addition to its prior ability to retrieve predictions using a predictionJobId.
  • Featurelists (see GetFeaturelist and GetModelingFeaturelist) now return a created value with the timestamp, a isUserCreated value explaining whether or not the feature was created by a user (as opposed to DataRobot automation), numModels showing how many models use the featurelist, and description which gives a text description of the featurelist.
  • GetDatetimePartition now reports information on the number of "known in advance" features.
  • GetDatetimePartition reports if the partition was drawn from a time series project and/or a multiseries project.

Bugfixes:

  • In rare instances, requests would fail due to not being able to properly create authentication headers. This has now been resolved.

API Changes:

  • RequestMultiSeriesDetection (which is now no longer necessary to invoke directly) now blocks until the multiseries request is complete and returns details about which multiseries columns were detected.

Deprecated and Defunct:

  • Reason codes have been renamed to Prediction Explanations. DeleteReasonCodes, DeleteReasonCodesInitialization, DownloadReasonCodes, GetAllReasonCodesRowsAsDataFrame, GetReasonCodesInitialization, GetReasonCodesInitializationFromJobId, GetReasonCodesMetadata, GetReasonCodesMetadataFromJobId, GetReasonCodesRows, ListReasonCodesMetadata, RequestReasonCodes, and RequestReasonCodesInitialization have all been deprecated (and will be removed in v2.15). These functions have been replaced with appropriately renamed functions and a new workflow. See GetPredictionExplanations for more.
  • RequestPredictionsForDataset is replaced by RequestPredictions and deprecated (and will be removed in v2.15).
  • GetDatetimeModelObject has been renamed to GetDatetimeModel. The original GetDatetimeModelObject has been deprecated and will be removed in v2.15.

Documentation Changes:

  • Added a vignette explaining the advanced tuning interfaces.
  • Vignettes have been updated to use the new Predict workflow.
  • The vignette on time series and multiseries has been expanded to include more useful information.
  • Vignettes now use GetRecommendedModel instead of calculating the best mode instead of calculating the best model.
  • Corrected various typos in docstrings.

datarobot v2.12.1

Enhancements:

  • plot on ListOfModels returns an error if the desired percent passed to pct is not found.

Bugfixes:

  • Fix as.data.frame to handle missing featurelist IDs when simple = FALSE.
  • Fix as.data.frame to handle a list of prediction datasets.
  • Fix as.data.frame to handle a list of models when samplePct is not set.
  • Fix plot on ListOfModels to work with missing featurelist IDs.
  • Fix summary methods to work on zero-length lists.

Deprecated and Defunct:

  • Passing a non-logical value to simple parameter in as.data.frame now produces an error instead of a warning.

datarobot v2.12.0

New Features:

  • A new shorthand, StartProject, combines both SetupProject and SetTarget into one function.
  • A report on how a model handles missing values for all features in the project can now be retrieved using GetMissingValuesReport.
  • You can now use UploadPredictionDatasetFromDataSource to create a prediction dataset from a DataRobot data source (introduced in v2.11).

Enhancements:

  • If you have enabled monotonic constraints for your project, you can now disable these constraints for training a new model. You do this by using RequestNewModel and passing "" (an empty string) as the value for each monotonic constraint featurelist you wish to override.

Bugfixes:

  • In v2.10 and v2.11, the AutopilotMode$Quick "mode" for SetTarget had been broken and no longer triggered the quick mode (instead it ran the full autopilot). This has been fixed.

Deprecated and Defunct:

  • GetProjectList is replaced by ListProjects and deprecated (and will be removed in v2.14).
  • GetAllRocCurves is replaced by ListRocCurves and deprecated (and will be removed in v2.14).
  • GetAllLiftCharts is replaced by ListLiftCharts and deprecated (and will be removed in v2.14).
  • GetModelJobs is replaced by ListModelJobs and deprecated (and will be removed in v2.14).

Dependency Changes:

  • To support new features, httr at version 1.2.0 or higher is now required.

datarobot v2.11.0

New Features:

  • DataRobot now recommends particular models. ListModelRecommendations has been added to get all the model recommendations, GetModelRecommendation can return a particular recommendation, and GetRecommendedModel returns the particular model object corresponding with a particular recommendation.
  • DataRobot now supports "Database Connectivity", allowing databases to be used as the source of data for projects and prediction datasets. The feature works on top of the JDBC standard, so a variety of databases conforming to that standard are available; a list of databases with tested support for DataRobot is available in the user guide in the web application. See ListDrivers and GetDriver to get available drivers, CreateDataStore to create a data store from a driver, and CreateDataSource to create a data source from a data store.
  • You can also create a project from a specified data source using SetupProjectFromDataSource.
  • Time series projects support multiseries as well as single series data. See the vignette on time series for details.
  • GetTimeSeriesFeatureDerivationLog can now be used to retrieve a lot of information on details for derived features for time series projects. DownloadTimeSeriesFeatureDerivationLog can download the log to a text document.

Enhancements:

  • GetFeatureInfo and ListFeatureInfo now report targetLeakage, specifying whether a feature is considered to have target leakage or not.
  • Added a helper method to easily cross validate a model. Just call CrossValidateModel on your model object.
  • ConnectToDataRobot now works with environment variables. Set DATAROBOT_API_ENDPOINT and DATAROBOT_API_TOKEN to connect to DataRobot. Note that previously the R client unofficially used DataRobot_URL and DataRobot_Token as environment variables to facilitate connecting to DataRobot, but these variables are now no longer supported.

Bugfixes:

  • Fix as.data.frame to handle missing featurelist IDs.

API Changes:

  • New parameters predictionsStartDate and predictionsEndDate added to UploadPredictionDataset to support bulk predictions upload for time series projects.

datarobot v2.10.0

Bugfixes:

  • The Model Deployment interface which was previously visible in the client has been removed to allow the interface to mature.
  • Fix as.data.frame to handle multiple featurelists.
  • Clarified the time series workflow in the time series vignette.
  • Fix partitionKeyCols parameter in CreateGroupPartition to more clearly error if more than one partition key is passed.
  • Formatting within vignettes has been cleaned and standardized.

Deprecated and Defunct:

  • The following were deprecated and have now been removed: the quickrun parameter on SetTarget, the ability to use GetFeatureInfo with feature IDs, the GetRecommendedBlueprints function, GetModelObject, GetAllModels, GetBlueprintDocuments, and the RequestPredictions function.
  • The defaultToAPriori parameter in CreateDatetimePartitionSpecification is being deprecated and has been renamed to defaultToKnownInAdvance. defaultToAPriori will be fully removed in v2.15.
  • The aPriori flag in the featureSettings parameter in CreateDatetimePartitionSpecification is being deprecated and has been renamed to knownInAdvance. The aPriori will be fully removed in v2.15.

datarobot v2.9.0

New features:

  • Models can now be deployed to dedicated prediction servers using the new model monitoring system via the API. Create a deployment via RequestModelDeployment, get information on a specific deployment using GetModelDeployment, and list information on all deployments across all projects via ListModelDeployments. You can also get more information on the service health of a particular deployment using GetModelDeploymentServiceStatistics or get the action log for a deployed model using GetModelDeploymentActionLog.

Enhancements:

  • DataRobot API now supports creating 3 new blender types - Random Forest, TensorFlow, LightGBM.
  • Multiclass projects now support blenders creation for 3 new blender types as well as Average and ENET blenders.
  • New attributes maxTrainRows, scaleoutMaxTrainPct, and scaleoutMaxTrainRows have been added to projects retrieved by GetProject. maxTrainRows specified the equivalent value to the existing maxTrainPct as a row count. The scaleout fields can be used to see how far scaleout models can be trained on projects, which for projects taking advantage of scalable ingest may exceed the limits on the data available to non-scaleout blueprints.
  • Models can be trained by requesting a particular row count using the new trainingRowCount argument, specifying a desired amount of rows instead of a desired percentage of the dataset (via the current samplePct parameter). Specifying model size by row count is recommended when the float precision of sample_pct could be problematic, e.g. when training on a small percentage of the dataset or when training up to partition boundaries. This new approach is available for RequestNewModel, RequestFrozenModel, and RequestSampleSizeUpdate. RequestFrozenDatetimeModel already had this feature.
  • GetPredictions now returns a more informative error message when the async service times out.
  • Individual features can now be marked as a priori or not a priori using the new featureSettings attribute when setting the target or specifying datetime partitioning settings on time series projects. Any features not specified in the featureSettings parameter will be assigned according to the defaultToAPriori value.
  • Three new options have been made available in the DatetimePartitioningSpecification to fine-tune how time-series projects derive modeling features. treatAsExponential can control whether data is analyzed as an exponential trend and transformations like log-transform are applied. differencingMethod can control which differencing method to use for stationary data. periodicities can be used to specify periodicities occurring within the data. All are optional and defaults will be chosen automatically if they are unspecified.
  • An error is now raised if you do not pass a valid partition to partitioning in SetTarget.

Bugfixes:

  • Fixed latency issues in UploadPredictionDataset and GetPredictions. These functions have now been fully tested to handle data up to 1GB, and likely can handle more than that. If you run into issues, try incrementing the maxWait parameter.
  • You can now set ssl_verify: FALSE in drconfig.yaml to not verify SSL when connecting with DataRobot.
  • Fixed a typo in the training predictions vignette. It previously read DownloadRatingTable when it meant to read DownloadTrainingPredictions.
  • Fixed a typo in the reason codes docstring examples. It previously read reasonCodeId <- GetReasonCodesMetadataFromJobId(projectId, jobId) when it should read reasonCodeId <- GetReasonCodesMetadataFromJobId(projectId, jobId)$id.

API Changes:

  • Now trainingRowCount is available on non-datetime models as well as "rowCount" based datetime models. It reports the number of rows used to train the model (equivalent to samplePct).
  • Added support for retrieving details about the Pareto Front for a Eureqa model. Use GetParetoFront to get the Pareto Front details and AddEureqaSolution to add a new solution to the leaderboard.

datarobot v2.8.0

New features:

  • A new premium feature, time series, is now available. New projects can be created as time series projects which automatically derive features from past data and forecast the future. See the time series documentation in the web app for more information.
  • The DataRobot API supports the creation, training, and predicting of multiclass classification projects. DataRobot, by default, handles a dataset with a numeric target column as regression. If your data has a numeric cardinality of up to 10 classes, you can override this behavior to instead create a multiclass classification project from the data. To do so, use the SetTarget function, setting targetType = TargetType$Multiclass. If DataRobot recognizes your data as categorical, and it has up to 10 classes, using multiclass will create a project that classifies which label the data belongs to.
  • With the introduction of Multiclass Classification projects, DataRobot needed a better way to explain the performance of a multiclass model so we created a new Confusion Chart. The API now supports retrieving and interacting with confusion charts.
  • GetFeatureInfo and ListFeatureInfo now return the EDA summary statistics (i.e., mean, median, minum, maximum, and standard deviation) for features where this is available (e.g., numeric, date, time, currency, and length features). These summary statistics will be formatted in the same format as the data it summarizes.
  • The DataRobot API now includes Rating Tables. A rating table is an exportable CSV representation of a model. Users can influence predictions by modifying them and creating a new model with the modified table.
  • You can now set scaleoutModelingMode when setting a project target. It can be used to control whether scaleout models appear in the autopilot and/or available blueprints. Scaleout models are only supported in the Hadoop enviroment with the corresponding user permission set.
  • You can now set accuracyOptimizedBlueprints when setting a project target. Accuracy optimized blueprints are longer running model blueprints that provide increased accuracy over the normal blueprints that run during autopilot.
  • DataRobot now supports retrieving model blueprint charts via GetModelBlueprintChart and model blueprint documentation via GetModelBlueprintDocumentation. These are like regular blueprint charts and blueprint documentation, except for model blueprints, which are a reduced representation of the blueprint run by the model to only include the relevant branches actually executed by the model.
  • The Datarobot API now supports generating and retrieving training predictions, which are predictions made by the model on out-of-fold training data. Users can start a job which will make training predictions and retrieve them. See the training predictions documentation in the web app for more information on how to use training predictions.

Enhancements:

  • CreateDatetimePartitionSpecification now includes the optional disableHoldout flag that can be used to disable the holdout fold when creating a project with datetime partitioning.
  • The advanced options available when setting the target have been extended to include the new parameters offset and exposure to allow specifying offset and exposure columns to apply to predictions generated by models within the project. See the user guide documentation in the web app for more information on offset and exposure columns.
  • The advanced options available when setting the target have been extended to include the new parameter eventsCount to allow specifying the events count column. See the user guide documentation in the webapp for more information on events count.
  • File URIs can now be used as sourcedata when creating a project or uploading a prediction dataset. The file URI must refer to an allowed location on the server, which is configured as described in the user guide documentation.
  • If this package is used in RStudio v1.1 or higher, it is possible to use the RStudio Connections UI to open a DataRobot connection.
  • When retrieving reason codes on a project using an exposure column, predictions that are adjusted for exposure can be retrieved.
  • ConnectToDataRobot now supports an option sslVerify that turns off SSL verification if set to FALSE.

Bugfixes:

  • Fixes a bug that prevented GetReasonCodesMetadataFromJobId from being called with a project directly (instead of a project id).
  • Fixes a bug that prevented RequestNewModel from being called when options(stringsAsFactors = TRUE) is set.
  • Fixes a bug that prevented more than one blueprint document from being returned by GetBlueprintDocuments (now named GetBlueprintDocumentation).

Deprecated and Defunct:

  • The quickrun parameter on SetTarget, the ability to use GetFeatureInfo with feature IDs, the GetRecommendedBlueprints function, and the RequestPredictions function were all originally planned to be deprecated in version 3.0. These features and functions will now be deprecated in v2.10 instead.
  • GetModelObject is replaced by GetModel and deprecated (and will be removed in v2.10).
  • GetAllModels is replaced by ListModels and deprecated (and will be removed in v2.10).
  • GetBlueprintDocuments is replaced by GetBlueprintDocumentation and deprecated (and will be removed in v2.10).

datarobot v2.7.1

Documentation Changes:

  • The modelwordcloud package is now available on CRAN, so the documentation has been updated to reflect CRAN installation instructions.

datarobot v2.7.0

New features:

  • Word cloud data for text processing models can be retrieved using GetWordCloud function.
  • Scoring code JAR file can be downloaded for models supporting code generation using 'DownloadScoringCode` function.
  • Lift Chart data can be retrieved using GetLiftCharts and GetAllLiftCharts function.
  • Roc Curve data for binary classification projects can be retrieved using GetRocCurve and GetAllRocCurves
  • Status and information about individual jobs can be retrieved using GetPredictJob',GetModelJob,GetJobfunctions. Any job can be retrieve viaGetJobwhich is less specific. Only prediction jobs can be retrieved withGetPredictJoband only modeling jobs can be retrieved withGetModelJob`.

Enhancements:

  • GetModelParameters now includes an additional key showing the coefficients for individual stages of multistage models (e.g. Frequency-Severity models).
  • When training a DatetimeModel on a window of data, a timeWindowSamplePct can be specified to take a uniform random sample of the training data instead of using all data within the window.

Bugfixes:

  • Fixed a bug where depending on what version of the R curl library was installed, the client could hang after requesting certain DataRobot jobs.
  • DownloadTransferrableModel now correctly handles HTTP errors.

Dependency Changes:

  • To support new features, jsonlite at version 1.0 or higher and curl at version 1.1 or higher are now required.

Deprecated and Defunct:

  • Semi-automatic autopilot mode is removed. Quick or manual mode can be used instead to get a sparser autopilot.

datarobot v2.6.0

New features:

  • Function CreateDerivedFeatureIntAsCategorical has been added. It creates new categorical feature based on parent numerical feature while truncating numerical values to integer. (All of the data in the column should be considered categorical in its string form when cast to an int by truncation. For example the value 3 will be cast as the string 3 and the value 3.14 will also be cast as the string 3. Further, the value -3.6 will become the string -3. Missing values will still be recognized as missing.)
  • Reason Codes, a new feature in DataRobot, is fully supported in the package through several new functions.
  • Functions which allow to access blueprint chart and documentation have been added.
  • Model parameters can now be retrieved using GetModelParameters function.
  • A new partitioning method (datetime partitioning) has been added. The recommended workflow is to preview the partitioning by creating a DatetimePartitioningSpecification using CreateDatetimePartition and CreateBacktestSpecification function and passing it into GenerateDatetimePartition, inspect the results and adjust as needed for the specific project dataset by adjusting the DatetimePartitioningSpecification and re-generating, and then set the target by passing the final DatetimePartitioningSpecification object to the partitioning_method parameter of SetTarget.

Enhancements:

  • The default value of the maxWait parameter used to control how long asynchronous routes are polled has been changed from 1 minute to 10 minutes.

API Changes:

  • projectId has been added to Feature schema
  • The UnpauseQueue function will not longer set the autopilot mode of a project to full autopilot. This means that projects using the (deprecated) SemiAuto autopilot mode will require the autopilot to be advanced via the webapp.

datarobot v2.5.0

New features:

  • Functions RequestFrozenModel, GetFrozenModel, GetFrozenModelFromJobId have been added. They allow user to create model with the same tuning parameters as parent model but with different data sample size and get information about frozen models in a project.
  • Functions RequestBlender, GetBlenderModelFromJobId, GetBlenderModel have been added. They allow user to create blender models and get information about blender models in a project.
  • Projects created via the API can now use smart downsampling when setting the target by passing smartDownsampled and majorityDownsamplingRate into the SetTarget function.

Enhancements:

  • Meaningful error messages have been added when the DataRobot endpoint is incorrectly specified in a way that causes redirects (e.g. specifying http for an https endpoint).
  • Previously it was not possible to use user partition columns with cross-validation without specifying a holdout level using the API. This can now be be done by either omitting the cvHoldoutLevel parameter or providing it as NA.

Bugfixes:

API Changes:

Deprecated and Defunct:

  • Support for recommender models has been removed from the DataRobot API. The package has been updated to remove functionality that formerly used this feature.

Documentation Changes:

datarobot v2.4.0

New features:

  • The premium feature DataRobot Prime has been added. You can now approximate a model on the leaderboard and download executable code for it. Talk to your account representative if the feature is not available on your account. The new related functions are GetPrimeEligibility, RequestApproximation, ListPrimeModels, GetPrimeModel, GetRulesets, RequestPrimeModel, GetPrimeModelFromJobId, CreatePrimeCode, GetPrimeFileFromJobid, ListPrimeFiles, GetPrimeFile, DownloadPrimeCode
  • A utility function, WaitForJobToComplete, has been added. It will block until the specified job finishes, or raise an error if it does not finish within a specified timeout.
  • Functions SetupProjectFromMySQL, SetupProjectFromOracle, SetupProjectFromPostgreSQL and SetupProjectFromHDFS have been added. They allow user to create DataRobot projects from MySQL, Oracle, PostgreSQL and HDFS data sources.
  • Functions RequestTransferrrableModel, DownloadTransferrableModel, UploadTransferrableModel, GetTransferrrableModel, ListTransferrrableModels, UpdateTransferrrableModel, DeleteTransferrrableModel have been added. They allow user to download models from modeling server and transfer them to special dedicated prediction server (those functions are only useful to users with on-premise environment)

Enhancements:

  • An optional maxWait parameter has been added to GetModelFromJobId and GetFeatureImpactForJobId, to allow users to specify an amount of time to wait for the job to complete other than the default 60 seconds.
  • Projects can now be run in quickrun mode (which skips some autopilot stages and longer-running models) by passing "quick" as the mode parameter, in the same way "auto" and "manual" modes can be specified.
  • The client will now check the API version offered by the server specified in configuration, and a warning if the client version is newer than the server version. The DataRobot server is always backwards compatible with old clients, but new clients may have functionality that is not implemented on older server versions. This issue mainly affects users with on-premise deployments of DataRobot.
  • SetupProject and UploadPredictionDataset accept url as dataSource parameter now

Bugfixes:

  • If a model job errors, GetModelFromJobId will now immediately raise an exception, rather than waiting for the timeout.
  • The maxWait parameter on UploadPredictionDataset will now be correctly applied.

API Changes:

Deprecated and Defunct:

  • The quickrun parameter on SetTarget is deprecated (and will be removed in 3.0). Pass "quick" as the mode parameter instead.

Documentation Changes:

datarobot v2.3.0

Enhancements:

  • When project creation using SetupProject times out, the error message now includes a URL to use with the new ProjectFromAsyncUrl function to resume waiting for the project creation.
  • GetFeatureInfo now supports retrieving features by feature name. (For backwards compatibility, feature IDs are still supported until 3.0.)
  • The package no longer relies on a particular version of the methods package. (This dependency was too strict and required some users to unnecessarily upgrade R.)
  • The projectName argument of SetupProject no longer defaults to the string 'None'. (The new default is not to send a name, which results in the name 'Untitled Project'.)
  • The maxWait argument for SetupProject now controls the timeout for the initial POST request and has a larger default value. The reason for this is that for large project creation file uploads, the server may take a longer-than-normal amount of time to respond, and waiting longer than the default timeout may be necessary.

Deprecated and Defunct:

  • The ability to use GetFeatureInfo with feature IDs is deprecated (and will be removed in 3.0). Use feature names instead.
  • GetRecommendedBlueprints is replaced by ListBlueprints and deprecated (and will be removed in 3.0).
  • RequestPredictions is deprecated and replaced by RequestPredictionsForDataset. RequestPredictionsForDataset will be renamed to RequestPredictions in 3.0.
  • DeletePendingJobs is removed; use DeleteModelJob instead
  • GetFeatures is removed; use ListModelFeatures instead
  • GetPendingJobs is removed; use GetModelJobs instead
  • StartAutopilot is removed; use SetTarget instead
  • parameter url is removed from ConnectToDataRobot
  • parameter jobStatus is removed from GetModelJobs
  • parameters saveFile and csvExtension are removed from RequestPredictions
  • parameters saveFile and csvExtension are removed from SetupProject
  • "semi" mode option (functions SetTarget, StartNewAutopilot) is deprecated (and will be removed in 3.0).

New features:

  • The API now supports the new Feature Impact feature. Use RequestFeatureImpact to start a job to compute FeatureImpact, and GetFeatureImpactForModel or GetFeatureImpactForJobId to retrieve the completed Feature Impact results.
  • The new functions CreateDerivedFeatureAsCategorical, CreateDerivedFeatureAsText, CreateDerivedFeatureAsNumeric can be used to create derived features as type transforms of existing features.
  • The API now supports uploading (UploadPredictionDataset), listing (ListPredictionDatasets), and deleting (DeletePredictionDataset) datasets for prediction as well as requesting predictions (RequestPredictionsForDataset) against such datasets.

Bugfixes

  • as.data.frame fixed for empty listOfBlueprints, listOfFeaturelists, listOfModels
  • The documentation for SetTarget incorrectly referred to the 'semiauto' (rather than 'semi') autopilot setting. This is fixed.
  • GetPredictions previously used a maxWait of 60, regardless of what maxWait the user specified. This is fixed.

datarobot v2.2.33

Bugfixes

  • GetModelJobFromId was broken by v2.2.32 and is now fixed.
  • CreateFeaturelist was broken by v2.2.32 and is now fixed.

datarobot v2.2.32

API Changes

  • Package renamed to datarobot.

New features:

  • ListJobs and DeleteJob functions added. ListJobs lists the jobs in the project queue (of any type). DeleteJob can be used to cancel one of these jobs.
  • ListFeatureInfo (for all features) and GetFeatureInfo (for one feature) have been added for retrieving feature details.

Enhancements:

  • In line with new functionality in version 2.2 of the DataRobot API, CreateUserPartition now allows holdoutLevel to be NULL (which results in not sending the holdout level, in line with backend API changes to allow user partitions to be created without a holdout level).
  • Slices using [ from objects of type listOfBlueprints, listOfFeaturelists, and listOfModels will now retain the appropriate type.
  • Several functions (e.g. ConnectToDataRobot, DeleteModel, PauseQueue, etc.) used to return TRUE as their only possible return value. Now they return nothing instead.
  • GetValidMetrics no longer has special-casing for the situation when the project is not yet ready to give you the valid metrics for a potential metric. In this case, an error will now be returned from the server.
  • Error messages from the server now include additional detail.
  • To improve error messages, in several places error messages no longer reference the top-level function the user called.
  • The SetTarget function will now properly block execution until the server indicates the project has finished initializing and is ready to build models

Deprecated and Defunct:

  • GetFeatures has been deprecated and renamed to ListModelFeatures (for more more clarity/consistency in naming and to avoid confusion with the now GetFeatureInfo and ListFeatureInfo)
  • Support for authenticating via username/password has been removed. Use an API Token instead
  • Removed broken UpdateDefaultPartition. To use one of the default partition methods with updated settings, please use CreateRandomPartition or CreateStratifiedPartition.

datarobot v2.1.31

Enhancements

  • Use of the WaitForAutopilot function will no longer trigger deprecation warnings

datarobot v2.1.30

Bugfixes

  • Due to a dependency on the methods package (which is loaded by default interactively but not running Rscript), RequestPredictions did not work when invoked with Rscript. This is fixed. The methods package is now in 'depends' instead of 'imports' to prevent this problem from ever occurring again.

datarobot v2.1.29

Deprecated & Defunct

  • Removed broken UpdateDefaultPartition. Please use the other partition-creating functions.

datarobot v2.1.28

Bugfixes

  • Due to a dependency on the methods package (which is loaded by default interactively but not running Rscript), some functions did not work when invoked with Rscript. This is fixed.
  • SetupProject and GetPredictions now check for and displays errors in project creation (previously they would keep waiting and time out if there are errors)
  • Previously errors would sometimes appear missing a space between two words. This is fixed.

datarobot v2.1.27

Bugfixes

  • Fixed a problem that caused an error when getting predictions if the installed version of the httr package was 1.0 and older.

datarobot v2.1.26

Enhancements:

  • HTTP requests now include User-Agent headers for logging purposes, e.g. "DataRobotRClient/2.0.25 (Darwin 14.5.0 x86_64)".
  • We now provide a more informative error message after receiving HTML from the server when we expected JSON.
  • We avoid httr encoding warning messages by specifying UTF-8.
  • It is now possible to not specify the desired jobStatus in GetPendingJobs (by passing NULL for the jobStatus argument, which is now the default).
  • GetPredictions now checks whether a prediction job has errored or been canceled and will error right away in that case (instead of waiting until the timeout)
  • When specifying the data source as a dataframe (in RequestPredictions or SetupProject), the class may now be a subclass of dataframe (it need not be equal to dataframe).
  • Previously GetModelJobs returned a dataframe when there are jobs but an empty list when there are none. Now it consistently returns a dataframe (with zero rows if there are no jobs) either way.

New features:

  • ConnectToDataRobot can now read from a YAML config file.
  • On package startup, we look for a config file in the default location, so the user does not need to call ConnectToDataRobot explicitly
  • WaitForAutopilot function added. This function periodically checks whether Autopilot is finished and returns only after it is.
  • SetupProject and RequestPredictions now default to using a tempfile instead of placing the file to be uploaded into the current working directory.
  • New function StartNewAutopilot can be used to restart autopilot on a specific featurelist if it was previously running on a different one.
  • New function SetTarget provides the functionality that StartAutopilot used to be responsible for. StartAutopilot is now deprecated, and SetTarget should be used instead. This function can now take a featurelistId argument, specifying which featurelist to use.

Bugfixes:

  • GetPendingJobs (now deprecated in favor of GetModelJobs) was broken and is now fixed.
  • GetValidMetrics was broken and is now fixed.
  • GetProjectList no longer errors when there are no projects. It now returns an object whose structure matches the returned object when there are projects.

Deprecated and Defunct:

  • The arguments controlling where the tempfile goes (in SetupProject and RequestPredictions) are now deprecated
  • DeletePendingJob is deprecated (use DeleteModelJob instead)
  • GetPendingJob is deprecated (use GetModelJob instead)
  • jobStatus argument to GetModelJob/GetPendingJob is deprecated (use status instead)
  • StartAutopilot is deprecated (use SetTarget instead).

API Changes Summary:

  • Support for the experimental date partitioning has been removed in DataRobot API, so it is being removed from the client immediately - the CreateDatePartition function has been removed.

datarobot v2.0.25

Enhancements:

  • Codebase cleaned of many lint violations.

New Features:

  • DeletePredictJob, GetPredictJobs, GetPredictions, RequestPredictions all added to control the prediction functionality created in v2.0 featureset of the API.
  • "quickrun" parameter added to StartAutopilot. This boolean enables use of the quickrun autopilot feature of DataRobot.

Bugfixes: None

Deprecated and Defunct: None

API Changes: None

datarobot v0.2.24

  • fixes the maxWait parameter that was unsuccessfully introduced in 0.2.23

datarobot v0.2.23

  • maxWait parameter added to SetupProject to allow for datasets that take very long to initialize on the DataRobot server

datarobot v0.2.22

  • Documentation structure changed to use Roxygen2