class

OpenAI::CreateEvalLabelModelGrader

Inherits JSON::Serializable < Reference < Object

A LabelModelGrader object which uses a model to assign labels to each item in the evaluation.

Constructors

new(pull : JSON::PullParser)

new(type_value : CreateEvalLabelModelGraderType = CreateEvalLabelModelGraderType::LabelModel, name : String = "", model : String = "", input : Array(CreateEvalItem) = [] of CreateEvalItem, labels : Array(String) = [] of String, passing_labels : Array(String) = [] of String)

new(*, __pull_for_json_serializable pull : JSON::PullParser)

Instance methods

input

A list of chat messages forming the prompt or context. May include variable references to the item namespace, ie {{item.name}}.

input=(input : Array(CreateEvalItem))

A list of chat messages forming the prompt or context. May include variable references to the item namespace, ie {{item.name}}.

labels

The labels to classify to each item in the evaluation.

labels=(labels : Array(String))

The labels to classify to each item in the evaluation.

model

The model to use for the evaluation. Must support structured outputs.

model=(model : String)

The model to use for the evaluation. Must support structured outputs.

name

The name of the grader.

name=(name : String)

The name of the grader.

passing_labels

The labels that indicate a passing result. Must be a subset of labels.

passing_labels=(passing_labels : Array(String))

The labels that indicate a passing result. Must be a subset of labels.

type_value

The object type, which is always label_model.

type_value=(type_value : CreateEvalLabelModelGraderType)

The object type, which is always label_model.