RestApiCategories.autoMachineLearning.desc

auto_machine_learning

2025.08

2025.09

2025.10

2025.11

2025.07

The Machine Learning Pipeline Automation (MLPA) API enables CRUD operations on automation projects, which automates VDMML project creation, pipeline building and training, and the production of champion models.

Machine Learning Pipeline Automation

2024.09

2025.03

2024.03

Viya_35

Score data with the automation project's champion model.

Score Data Input

projectId

Score Data Output

Example of an error response when required fields are missing

Error

application/vnd.sas.error

No required resources could be found. For example, automation project, analytics project, or champion model. This error condition could also occur if the champion model was not published to the supported destination.

Score data with the champion model of the automation project

This object contains a list of analytics project attributes related to Model Studio project settings.


Analytics Project Attributes

An analytics report consisting of rows of data and associated metadata (parameters).

Analytics Report

The list of links to top-level resources and operations available from the root of the API.

A collection of optional settings to configure an automation project after Model Studio project is created based on analyticsProjectAttributes.


Automation Project Settings

Automation Project state. One of the enums [pending, preparing, waiting, ready, modeling, constructingPipeline, runningPipeline, quiescing, quiesced, completed, canceled, failed, oversampling, retraining].

Automation Project State

This is a base schema used to define paginated collections of resources. This base schema is extended by other schemas in APIs by adding an 'items' array property. These extensions define the application/vnd.sas.collection media type (version 2)

Base Collection

This object contains a summary of champion model.

Champion Model

This object contains the response to a champion model action.

Champion Model Action Response

Data Row

The callback schema for Data Mining GenAI jobs.

genAiCallback

Input Data that will be scored.  This doesn't represent any top-level media type.

Input Data

Job Definition

Job Definition Parameter

A link to a related operation or resource.

Link

This object contains metadata and information about an automation project.

Machine Learning Pipeline Automation Project

This object contains a list of automation project objects.

Machine Learning Pipeline Automation Project Collection

This object contains information for a pipeline template that can be used in project creation.

Machine Learning Pipeline Template

This object contains a list of machine learning pipeline templates.

Machine Learning Pipeline Template Collection

Assigning Input Data column to a variable of the generated Mapped Code of Score Object. This doesn't represent any top-level media type.

Mapping

Descriptor of a Score Object whose logic is used to produce a score for Input Data. This doesn't represent any top-level media type.

Object Descriptor

Output generated by score execution or analysis. This doesn't represent any top-level media type.

Output Table

Overrides the details of the existing Score definition.  This doesn't represent any top-level media type.

Override Score Definition

Metadata describing a particular variable.

Parameter

A pair consisting of a string name and a string value. This is an element of `application/vnd.sas.properties`

Property

This object contains publish destination information.

Publish Destination

The collection of application/vnd.sas.analytics.report+json representations.

Report Collection

This object contains a data field name value pair for score data inputs and outputs.

Score Data Field Name/Value Pair

A definition containing details about Input Data, Score Object and Mapping. This represents application/vnd.sas.score.definition media type (version 1).

Score Definition

Contains details about execution results. This represents application/vnd.sas.score.execution media type (version 1).

Score Execution

Contains request details about how to generate score. This represents application/vnd.sas.score.execution.request media type (version 1).

Score Execution Request

Supervised Learning Algorithm. One of the enums [DecisionTree, Forest, GAM, GradientBoosting, NeuralNetwork, Regression, Ensemble, SVM].

Supervised Learning Algorithm

Describes the behavior of a supervised learning algorithm during pipeline generation. One of the enums [consider, exclude, forceInclude].

Supervised Learning Algorithm Behavior in Pipeline Generation

This object contains a list of supervised learning algorithms.

Supervised Learning Algorithm Collection

A collection of settings to configure a supervised learning algorithm. defaultBehavior field is read-only.

Name	Type	Required	Description
scoreType	string	true	Score type, either 'Individual' or 'Bulk'. Allowed values: IndividualBulk
destinationName	string	false	Name of the destination where the champion model is published. Required for individual scoring.
inputs	array [Score Data Field Name/Value Pair]	false	Data inputs for individual data scoring. Required for individual scoring.
inputDataTableUri	string	false	Input data table URI for bulk data scoring. Required for bulk scoring.
scoreOutputCaslibUri	string	false	Output caslib URI where the score output table is stored. Required for bulk scoring.
scoreOutputTableName	string	false	Output table name where the score outputs are stored. Required for bulk scoring.

Status	Meaning	Description
200	OK	The request succeeded.	Schema
400	Bad Request	The request cannot be processed.	Schema
404	Not Found	No required resources could be found. For example, automation project, analytics project, or champion model. This error condition could also occur if the champion model was not published to the supported destination.	Schema

Score data with the champion model of the automation project

Request Samples

Response Samples

Path Parameters

Request Body

Responses