interface EvaluateOptionsBaseEvaluateOptionsThe LangSmith client to use.
The dataset to evaluate on. Can be a dataset name, a list of examples, or a generator of examples.
A free-form description of the experiment.
Whether to skip tracing evaluator invocations to the evaluators project.
When true, evaluators run without creating evaluator traces. Feedback is
still created and attached to the experiment runs, but it won't link back
to an evaluator run, which also means it can't be corrected from the UI.
Tracing of the target itself is unaffected.
The maximum number of concurrent evaluators to run.
If not provided, defaults to maxConcurrency when set.
A list of evaluators to run on each example.
A prefix to provide for your experiment name.
Whether to use attachments for the experiment.
The maximum concurrency to use for predictions/evaluations when a more specific concurrency option is not provided.
Metadata to attach to the experiment.
The number of repetitions to perform. Each example will be run this many times.
A list of summary evaluators to run on the entire dataset.
The maximum number of concurrent predictions to run.
If not provided, defaults to maxConcurrency when set.