evals with generation

This commit is contained in:
Xi Yan 2024-10-24 11:30:13 -07:00
parent 071dba8871
commit 737fcb795f
15 changed files with 385 additions and 15 deletions

View file

@ -61,20 +61,26 @@ def data_url_from_file(file_path: str) -> str:
return data_url
async def register_dataset(datasets_impl: Datasets):
async def register_dataset(
datasets_impl: Datasets, include_generated_answer=True, dataset_id="test_dataset"
):
test_file = Path(os.path.abspath(__file__)).parent / "test_dataset.csv"
test_url = data_url_from_file(str(test_file))
dataset_schema = {
"expected_answer": StringType(),
"input_query": StringType(),
}
if include_generated_answer:
dataset_schema["generated_answer"] = StringType()
dataset = DatasetDefWithProvider(
identifier="test_dataset",
identifier=dataset_id,
provider_id=os.environ["PROVIDER_ID"],
url=URL(
uri=test_url,
),
dataset_schema={
"generated_answer": StringType(),
"expected_answer": StringType(),
"input_query": StringType(),
},
dataset_schema=dataset_schema,
)
await datasets_impl.register_dataset(dataset)