LangSmith · Method
Client.evaluate
Client.evaluate() runs an application over a LangSmith dataset and scores each output with evaluators. It creates an experiment whose results can be viewed in the UI or processed through the SDK.
Last updated
- Since
- langsmith >= 0.3.13
Syntax
Client.evaluate(target, *, data, evaluators=None, experiment_prefix=None, description=None, max_concurrency=None, metadata=None, blocking=True)In-depth API docs are only for subscribers: parameters, examples, common mistakes and interview practice.
SubscribeRelated
aevaluate()ExperimentResultsEvaluationResultRun an evaluation asynchronouslyHow to read experiment results locally