prepare() — langchain Function Reference
Architecture documentation for the prepare() function in runner_utils.py from the langchain codebase.
Entity Profile
Dependency Diagram
graph TD 00d82cfb_ba59_4f67_e504_1faad0617f06["prepare()"] 3aaa6e94_b6a8_1c13_86d0_1709a1d93909["_DatasetRunContainer"] 00d82cfb_ba59_4f67_e504_1faad0617f06 -->|defined in| 3aaa6e94_b6a8_1c13_86d0_1709a1d93909 1ac4f6b0_183e_bd75_7f40_81c2246416b6["arun_on_dataset()"] 1ac4f6b0_183e_bd75_7f40_81c2246416b6 -->|calls| 00d82cfb_ba59_4f67_e504_1faad0617f06 0522e7ee_f1e6_b6f7_6738_1dad72cfffba["run_on_dataset()"] 0522e7ee_f1e6_b6f7_6738_1dad72cfffba -->|calls| 00d82cfb_ba59_4f67_e504_1faad0617f06 9a9f493e_7864_c75d_ebe0_af192df494f6["_prepare_eval_run()"] 00d82cfb_ba59_4f67_e504_1faad0617f06 -->|calls| 9a9f493e_7864_c75d_ebe0_af192df494f6 c2ae8ee6_ba74_2f11_df16_cafb61b88f1e["_wrap_in_chain_factory()"] 00d82cfb_ba59_4f67_e504_1faad0617f06 -->|calls| c2ae8ee6_ba74_2f11_df16_cafb61b88f1e ae9076b6_e76d_5271_0240_412b70e62fda["_setup_evaluation()"] 00d82cfb_ba59_4f67_e504_1faad0617f06 -->|calls| ae9076b6_e76d_5271_0240_412b70e62fda 42662eb3_17b6_0aaa_f09d_12aabcf769e7["_validate_example_inputs()"] 00d82cfb_ba59_4f67_e504_1faad0617f06 -->|calls| 42662eb3_17b6_0aaa_f09d_12aabcf769e7 style 00d82cfb_ba59_4f67_e504_1faad0617f06 fill:#6366f1,stroke:#818cf8,color:#fff
Relationship Graph
Source Code
libs/langchain/langchain_classic/smith/evaluation/runner_utils.py lines 1221–1293
def prepare(
cls,
client: Client,
dataset_name: str,
llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY,
project_name: str | None,
evaluation: smith_eval.RunEvalConfig | None = None,
tags: list[str] | None = None,
input_mapper: Callable[[dict], Any] | None = None,
concurrency_level: int = 5,
project_metadata: dict[str, Any] | None = None,
revision_id: str | None = None,
dataset_version: datetime | str | None = None,
) -> _DatasetRunContainer:
project_name = project_name or name_generation.random_name()
if revision_id:
if not project_metadata:
project_metadata = {}
project_metadata.update({"revision_id": revision_id})
wrapped_model, project, dataset, examples = _prepare_eval_run(
client,
dataset_name,
llm_or_chain_factory,
project_name,
project_metadata=project_metadata,
tags=tags,
dataset_version=dataset_version,
)
tags = tags or []
for k, v in (project.metadata.get("git") or {}).items():
tags.append(f"git:{k}={v}")
run_metadata = {"dataset_version": project.metadata["dataset_version"]}
if revision_id:
run_metadata["revision_id"] = revision_id
wrapped_model = _wrap_in_chain_factory(llm_or_chain_factory)
run_evaluators = _setup_evaluation(
wrapped_model,
examples,
evaluation,
dataset.data_type or DataType.kv,
)
_validate_example_inputs(examples[0], wrapped_model, input_mapper)
progress_bar = progress.ProgressBarCallback(len(examples))
configs = [
RunnableConfig(
callbacks=[
LangChainTracer(
project_name=project.name,
client=client,
example_id=example.id,
),
EvaluatorCallbackHandler(
evaluators=run_evaluators or [],
client=client,
example_id=example.id,
max_concurrency=0,
),
progress_bar,
],
tags=tags,
max_concurrency=concurrency_level,
metadata=run_metadata,
)
for example in examples
]
return cls(
client=client,
project=project,
wrapped_model=wrapped_model,
examples=examples,
configs=configs,
batch_evaluators=evaluation.batch_evaluators if evaluation else None,
)Domain
Subdomains
Called By
Source
Frequently Asked Questions
What does prepare() do?
prepare() is a function in the langchain codebase, defined in libs/langchain/langchain_classic/smith/evaluation/runner_utils.py.
Where is prepare() defined?
prepare() is defined in libs/langchain/langchain_classic/smith/evaluation/runner_utils.py at line 1221.
What does prepare() call?
prepare() calls 4 function(s): _prepare_eval_run, _setup_evaluation, _validate_example_inputs, _wrap_in_chain_factory.
What calls prepare()?
prepare() is called by 2 function(s): arun_on_dataset, run_on_dataset.
Analyze Your Own Codebase
Get architecture documentation, dependency graphs, and domain analysis for your codebase in minutes.
Try Supermodel Free