class

OpenAI::Eval

Inherits JSON::Serializable < Reference < Object

An Eval object with a data source config and testing criteria. An Eval represents a task to be done for your LLM integration. Like:

  • Improve the quality of my chatbot
  • See how well my chatbot handles customer support
  • Check if o4-mini is better at my usecase than gpt-4o

Constructors

new(pull : JSON::PullParser)
Source
new(object : EvalObject = EvalObject::Eval, id : String = "", name : String = "", data_source_config : EvalDataSourceConfig = EvalCustomDataSourceConfig.new, testing_criteria : Array(EvalTestingCriteriaItem) = [] of EvalTestingCriteriaItem, created_at : Int64 = 0, metadata : Metadata = nil)
Source
new(*, __pull_for_json_serializable pull : JSON::PullParser)
Source

Instance methods

created_at

The Unix timestamp (in seconds) for when the eval was created.

Source
created_at=(created_at : Int64)

The Unix timestamp (in seconds) for when the eval was created.

Source
data_source_config

Configuration of data sources used in runs of the evaluation.

Source
data_source_config=(data_source_config : EvalDataSourceConfig)

Configuration of data sources used in runs of the evaluation.

Source
id

Unique identifier for the evaluation.

Source
id=(id : String)

Unique identifier for the evaluation.

Source
metadata
Source
metadata=(metadata : Metadata)
Source
name

The name of the evaluation.

Source
name=(name : String)

The name of the evaluation.

Source
object

The object type.

Source
object=(object : EvalObject)

The object type.

Source
testing_criteria

A list of testing criteria.

Source
testing_criteria=(testing_criteria : Array(EvalTestingCriteriaItem))

A list of testing criteria.

Source