Package dev.braintrust.eval

package dev.braintrust.eval
  • Class
    Description
    Classification
    A single structured classification produced by a Classifier.
    Classifier<INPUT,OUTPUT>
    A classifier categorizes and labels eval outputs, producing zero or more structured Classification items.
    Dataset<INPUT,OUTPUT>
    Datasets define the cases for evals.
    Dataset.Cursor<CASE>
     
    DatasetBrainstoreImpl<INPUT,OUTPUT>
    A dataset loaded externally from Braintrust using paginated API fetches
    DatasetCase<INPUT,OUTPUT>
    A single row in a dataset.
    Eval<INPUT,OUTPUT>
    An evaluation framework for testing AI models.
    Eval.Builder<INPUT,OUTPUT>
    Builder for creating evaluations with fluent API.
    EvalCase<INPUT,OUTPUT>
    Deprecated.
    EvalResult
    Results of all eval cases of an experiment.
    ParameterDef<T>
    Definition of a named parameter that can be configured from the Braintrust Playground UI.
    ParameterDef.Type
     
    Parameters
    Holds the merged parameter values for a single eval run, along with the parameter definitions.
    Score
    Individual metric value assigned by a scorer.
    Scorer<INPUT,OUTPUT>
    A scorer evaluates the result of a task against a dataset case, producing a score between 0 (inclusive) and 1 (inclusive).
    ScorerBrainstoreImpl<INPUT,OUTPUT>
    A scorer that invokes a remote Braintrust function to compute scores.
    Task<INPUT,OUTPUT>
    A task function that runs against a single dataset case and produces an output.
    TaskResult<INPUT,OUTPUT>
    Result from a single task run.
    TracedClassifier<INPUT,OUTPUT>
    A classifier that receives access to the full distributed trace of the task that was evaluated.
    TracedScorer<INPUT,OUTPUT>
    A scorer that receives access to the full distributed trace of the task that was evaluated.