phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-31 11:10:02 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

include benchmarks

Browse source
This commit is contained in:
Xi Yan 2025-03-12 00:43:24 -07:00
parent e68e8c96ae
commit b4d868a1e5
3 changed files with 70 additions and 35 deletions
Download patch file Download diff file Expand all files Collapse all files

28
llama_stack/apis/benchmarks/benchmarks.py
View file

@ -8,15 +8,22 @@ from typing import Any, Dict, List, Literal, Optional, Protocol, runtime_checkab
from pydantic import BaseModel, Field
from llama_stack.apis.resource import Resource, ResourceType
from llama_stack.apis.scoring_functions import ScoringFnParams
from llama_stack.schema_utils import json_schema_type, webmethod
class CommonBenchmarkFields(BaseModel):
"""
:param dataset_id: The ID of the dataset to used to run the benchmark.
:param scoring_functions: The scoring functions with parameters to use for this benchmark.
:param metadata: Metadata for this benchmark for additional descriptions.
"""
dataset_id: str
scoring_functions: List[str]
scoring_functions: List[ScoringFnParams]
metadata: Dict[str, Any] = Field(
default_factory=dict,
description="Metadata for this evaluation task",
description="Metadata for this benchmark",
)
@ -57,10 +64,17 @@ class Benchmarks(Protocol):
@webmethod(route="/eval/benchmarks", method="POST")
async def register_benchmark(
self,
benchmark_id: str,
dataset_id: str,
scoring_functions: List[str],
provider_benchmark_id: Optional[str] = None,
provider_id: Optional[str] = None,
scoring_functions: List[ScoringFnParams],
benchmark_id: Optional[str] = None,
metadata: Optional[Dict[str, Any]] = None,
) -> None: ...
) -> Benchmark:
"""
Register a new benchmark.
:param dataset_id: The ID of the dataset to used to run the benchmark.
:param scoring_functions: The scoring functions with parameters to use for this benchmark.
:param benchmark_id: (Optional) The ID of the benchmark to register. If not provided, a random ID will be generated.
:param metadata: (Optional) Metadata for this benchmark for additional descriptions.
"""
...