Skip to content
For Managed AI Platform users, see the cloud platform documentation
DataRobot API resources > API reference documentation > API changelogs > Python client changelog > v3.6 changelog

v3.6 changelog

Reference the changes introduced to v3.6 of DataRobot's Python API client.

Python client v3.6

New features

  • Added a new class, OCRJobResource, for running OCR jobs.
  • Added new Jina V2 embedding model in VectorDatabaseEmbeddingModel.
  • Added new small multi-lingual embedding model in VectorDatabaseEmbeddingModel.
  • Added the method Deployment.get_segment_attributes to retrieve segment attributes.
  • Added the method Deployment.get_segment_values to retrieve segment values.
  • Added the method AutomatedDocument.list_all_available_document_types to return a list of document types.
  • Added the method Model.request_per_class_fairness_insights to return per-class bias and fairness insights.
  • Added the class MLOpsEvent <datarobot.mlops.events.MLOpsEvent> to report MLOps events. This class only supports moderation MLOps events.
  • Added the method Deployment.get_moderation_events to retrieve moderation events for that deployment.
  • Extended the advanced options available when setting a target to include new parameters: number_of_incremental_learning_iterations_before_best_model_selection (part of the AdvancedOptions object). This parameter allows you to specify how long the top five models will run prior to the selection of the best model.
  • Added support for connector_type in the method Connector.create.
  • Added the class DataQualityExport and the method Deployment.list_data_quality_exports, used to to retrieve a list of data quality records.
  • Added secure config support for Azure Service Principal credentials.
  • Added support for categorical custom metrics in CustomMetric.
  • Added the class NemoConfiguration <datarobot.models.genai.nemo_configuration.NemoConfiguration> to manage Nemo configurations.
  • Added the method NemoConfiguration.create <datarobot.models.genai.nemo_configuration.NemoConfiguration.upsert> to create or update a Nemo configuration.
  • Added the method NemoConfiguration.get <datarobot.models.genai.nemo_configuration.NemoConfiguration.get> to retrieve a Nemo configuration.
  • Added a new class ShapDistributions <datarobot.insights.ShapDistributions> to interact with SHAP distribution insights.
  • Added the MODEL_COMPLIANCE_GEN_AI value to the attribute document_type from the class DocumentOption <datarobot.models.automated_documentation.DocumentOption> to generate compliance documentation for LLMs in the Registry.
  • Added new attribute prompts_count to the class Chat.
  • Added the class Recipe modules for Data Wrangling.
  • Added the class RecipeOperation and a set of subclasses to represent a single Recipe.operations operation.
  • Added a new attribute similarity_score to the class Citation.
  • Added a new attributes retriever and add_neighbor_chunks to the class VectorDatabaseSettings.
  • Added a new attribute metadata to the class Citation.
  • Added a new attribute metadata_filter to the class ChatPrompt.
  • Added a new attribute metadata_filter to the class ComparisonPrompt.
  • Added a new attribute custom_chunking to the class ChunkingParameters.
  • Added a new attribute custom_chunking to the class VectorDatabase.
  • Added a new class LLMTestConfiguration for LLM test configurations.
  • Use the method LLMTestConfiguration.get <datarobot.models.genai.llm_test_configuration.LLMTestConfiguration.get> to retrieve a hosted LLM test configuration.
  • Use the method LLMTestConfiguration.list <datarobot.models.genai.llm_test_configuration.LLMTestConfiguration.list> to list hosted LLM test configurations.
  • Use the method LLMTestConfiguration.create <datarobot.models.genai.llm_test_configuration.LLMTestConfiguration.create> to create an LLM test configuration.
  • Use the method LLMTestConfiguration.update <datarobot.models.genai.llm_test_configuration.LLMTestConfiguration.update> to update an LLM test configuration.
  • Use the method LLMTestConfiguration.delete <datarobot.models.genai.llm_test_configuration.LLMTestConfiguration.delete> to delete an LLM test configuration.
  • Added a new classLLMTestConfigurationSupportedInsights for LLM test configuration supported insights.
  • Use the method LLMTestConfigurationSupportedInsights.list to list hosted LLM test configuration supported insights.
  • Added a new class LLMTestResult for LLM test results.
  • Use the method LLMTestResult.get to retrieve a hosted LLM test result.
  • Use the method LLMTestResult.list to list hosted LLM test results.
  • Use the method LLMTestResult.create to create an LLM test result.
  • Use the method LLMTestResult.delete to delete an LLM test result.
  • Added a new attribute dataset_name to the class OOTBDatasetDict.
  • Added a new attribute rows_count to the class OOTBDatasetDict.
  • Added a new attribute max_num_prompts to the class DatasetEvaluationDict.
  • Added a new attribute prompt_sampling_strategy to the class DatasetEvaluationDict.
  • Added a new class DatasetEvaluationRequestDict for dataset evaluations in create and edit requests.
  • Added a new attribute evaluation_dataset_name to the class InsightEvaluationResult <datarobot.models.genai.llm_test_result.InsightEvaluationResult>.
  • Added a new attribute chat_name to the class InsightEvaluationResult.
  • Added a new attribute llm_test_configuration_name to the class LLMTestResult.
  • Added a new attribute creation_user_name to the class LLMTestResult.
  • Added a new attribute pass_percentage to the class LLMTestResult.
  • Added a new attribute evaluation_dataset_name to the class DatasetEvaluation.
  • Added a new attribute datasets_compatibility to the class LLMTestConfigurationSupportedInsights.
  • Added a new class NonOOTBDataset for non out-of-the-box (OOTB) dataset entities.
  • Use the method NonOOTBDataset.list to retrieve non OOTB datasets for compliance testing.
  • Added a new class {class}OOTBDataset for out-of-the-box dataset entities.
  • Use the method OOTBDataset.list to retrieve out-of-the-box datasets for compliance testing.
  • Added a new class TraceMetadata to retrieve trace metadata.
  • Add new attributes to the class VectorDatabase: parent_id, family_id, metadata_columns, added_dataset_ids, added_dataset_names, and version.
  • Use the method VectorDatabase.get_supported_retrieval_settings to retrieve supported retrieval settings.
  • Use the method VectorDatabase.submit_export_dataset_job to submit the vector database as a dataset to the AI catalog.
  • Updated the method VectorDatabase.create to create a new vector database version.
  • Added a new class SupportedRetrievalSettings for supported vector database retrieval settings.
  • Added a new class SupportedRetrievalSetting for supported vector database retrieval setting.
  • Added a new class VectorDatabaseDatasetExportJob for vector database dataset export jobs.
  • Added new attribute playground_id to the class CostMetricConfiguration.
  • Added new attribute name to the class CostMetricConfiguration.
  • Added a new class SupportedInsights to support lists.
  • Use the method SupportedInsights.list to list supported insights.
  • Added a new class MetricInsights for the new metric insights routes.
  • Use the method MetricInsights.list to list metric insights.
  • Use the method MetricInsights.copy_to_playground to copy metrics to another playground.
  • Added a new class PlaygroundOOTBMetricConfiguration for OOTB metric configurations.
  • Updated the schema for the class EvaluationDatasetMetricAggregation to include the new attributes ootb_dataset_name, dataset_id, and dataset_name.
  • Updated the method EvaluationDatasetMetricAggregation.list with additional optional filter parameters.
  • Added new attribute warning to the class OOTBDataset.
  • Added new attribute warning to the class OOTBDatasetDict.
  • Added new attribute warnings to the class LLMTestConfiguration.
  • Added a new parameter playground_id to the method SidecarModelMetricValidation.create to support sidecar model metrics transition to playground.
  • Updated the schema for the class NemoConfiguration to include the new attributes prompt_pipeline_template_id and response_pipeline_template_id.
  • Added new attributes to the class EvaluationDatasetConfiguration: rows_count and playground_id.

API changes

  • Updated the exc_message for ServerError to be constructed with a request ID to help with debugging.
  • Added the method Deployment.get_capabilities to retrieve a list of Capability objects containing capability details.
  • Advanced options parameters: modelGroupId, modelRegimeId, and modelBaselines were renamed into seriesId, forecastDistance, and forecastOffsets.
  • Added the parameter use_sample_from_dataset from the method Project.create_from_dataset. This parameter, when set, uses the EDA sample of the dataset to start the project.
  • Added the parameter quick_compute to functions in the classes ShapMatrix, ShapImpact, and ShapPreview.
  • Added the parameter copy_insights to the method Playground.create to copy the insights from an existing playground to a new one.
  • Added the parameter llm_test_configuration_ids to the method LLMBlueprint.register_custom_model, used to run LLM compliance tests when a blueprint is sent to the custom model workshop.

Enhancements

  • Added standard pagination parameters (e.g. limit, offset) to the method Deployment.list <datarobot.models.Deployment.list>, allowing you to get deployment data in smaller chunks.

Bugfixes

  • Fixed a field in the class CustomTaskVersion for controlling network policies. This is changed from outgoing_network_policy to outbound_network_policy. When performing a GET action, this field was incorrect and always resolved to None. When attempting a POST or PATCH action, the incorrect field would result in a 422 error. A
  • Changed the name of datarobot.enums.CustomTaskOutgoingNetworkPolicy to datarobot.enums.CustomTaskOutboundNetworkPolicy to reflect the proper field name.
  • Fixed a schema for the class DataSliceSizeInfo so that it allows an empty list for the messages field.

Deprecation summary

  • Removed the parameter in_use from the method ImageAugmentationList.create. This parameter was deprecated in v3.1.0.
  • Deprecated the method AutomatedDocument.list_available_document_types. Please use the method AutomatedDocument.list_all_available_document_types instead.
  • Deprecated the method Model.request_fairness_insights. Please use the method Model.request_per_class_fairness_insights instead, to return StatusCheckJob instead of status_id.
  • Deprecated the method Model.get_prime_eligibility. Prime models are no longer supported.
  • The eligibleForPrime field will no longer be returned from the method Model.get_supported_capabilities and will be removed after version 3.8 is released.
  • Deprecated the property ShapImpact.row_count and it will be removed after version 3.7 is released.
  • Advanced options parameters: modelGroupId, modelRegimeId, and modelBaselines were renamed into seriesId, forecastDistance, and forecastOffsets. The former names are deprecated and they will be removed after version 3.6 is released.
  • Renamed datarobot.enums.CustomTaskOutgoingNetworkPolicy to datarobot.enums.CustomTaskOutboundNetworkPolicy to reflect bugfix changes. The original enum was unusable.
  • Removed the parameter user_agent_suffix in datarobot.Client. Please use trace_context instead.
  • Removed the deprecated method DataStore.get_access_list. Please use the method DataStore.get_shared_roles instead.
  • Removed support for SharingAccess instances in the method DataStore.update_access_list. Use SharingRole instances instead.
  • Deprecated file_path for the methods Connector.create and Connector.update.

Configuration changes

  • Removed the upper bound pin on the urllib3 package to allow versions 2.0.2 and above.
  • Upgraded the Pillow library to version 10.3.0. Users installing DataRobot with the "images" extra (pip install datarobot[images]) should note that this is a required library.

Documentation changes

  • The API Reference page has been split into multiple sections for better usability.
  • Fixed documentation for the method Project.refresh to clarify that it does not return a value.
  • Fixed code examples for the class ExternalScores.
  • Added a copy button to code examples in ReadTheDocs documentation, for convenience.
  • Removed the outdated "Examples" section from the documentation. Please refer to DataRobot's API Documentation Home for more examples.
  • Removed the duplicate "Getting started" section from the documentation.
  • Updated to Sphinx RTD Theme v3.

Experimental changes

  • Added the force_update parameter to the update method in the class ChunkDefinition.
  • Removed attribute select_columns from the class ChunkDefinition
  • Added initial experimental support for Chunking Service V2
  • DatasetDefinition
  • DatasetProps
  • DatasetInfo
  • DynamicDatasetProps
  • RowsChunkDefinition
  • FeaturesChunkDefinition
  • ChunkDefinitionStats

  • ChunkDefinition

  • Added a new method, update, to the class ChunkDefinition.

  • Added experimental support for time-aware wrangling, including a usage template:

  • datarobot._experimental.models.time_series_wrangling_template.user_flow_template: Experimental changes offer automated time series feature engineering for the data in Snowflake or Postgres.

  • Added the ability to use the Spark dialect when creating a recipe, allowing data wrangling support for files.

  • Added new attribute warning to the class Chat.

  • Moved all modules from datarobot._experimental.models.genai to datarobot.models.genai.

  • Added a new method Model.train_first_incremental_from_sample that trains the first incremental learning iteration from an existing sample model. Requires the "Project Creation from a Dataset Sample" feature flag.