phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-08-07 19:12:09 +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

45
docs/_static/llama-stack-spec.html vendored
View file

@ -2426,7 +2426,14 @@
"post": { "post": {
"responses": { "responses": {
"200": { "200": {
"description": "OK" "description": "OK",
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/Benchmark"
}
}
}
}, },
"400": { "400": {
"$ref": "#/components/responses/BadRequest400" "$ref": "#/components/responses/BadRequest400"
@ -2444,7 +2451,7 @@
"tags": [ "tags": [
"Benchmarks" "Benchmarks"
], ],
"description": "", "description": "Register a new benchmark.",
"parameters": [], "parameters": [],
"requestBody": { "requestBody": {
"content": { "content": {
@ -7098,13 +7105,15 @@
"default": "benchmark" "default": "benchmark"
}, },
"dataset_id": { "dataset_id": {
"type": "string" "type": "string",
"description": "The ID of the dataset to used to run the benchmark."
}, },
"scoring_functions": { "scoring_functions": {
"type": "array", "type": "array",
"items": { "items": {
"type": "string" "$ref": "#/components/schemas/ScoringFnParams"
} },
"description": "The scoring functions with parameters to use for this benchmark."
}, },
"metadata": { "metadata": {
"type": "object", "type": "object",
@ -7129,7 +7138,8 @@
"type": "object" "type": "object"
} }
] ]
} },
"description": "Metadata for this benchmark for additional descriptions."
} }
}, },
"additionalProperties": false, "additionalProperties": false,
@ -9448,23 +9458,20 @@
"RegisterBenchmarkRequest": { "RegisterBenchmarkRequest": {
"type": "object", "type": "object",
"properties": { "properties": {
"benchmark_id": {
"type": "string"
},
"dataset_id": { "dataset_id": {
"type": "string" "type": "string",
"description": "The ID of the dataset to used to run the benchmark."
}, },
"scoring_functions": { "scoring_functions": {
"type": "array", "type": "array",
"items": { "items": {
"type": "string" "$ref": "#/components/schemas/ScoringFnParams"
} },
"description": "The scoring functions with parameters to use for this benchmark."
}, },
"provider_benchmark_id": { "benchmark_id": {
"type": "string" "type": "string",
}, "description": "(Optional) The ID of the benchmark to register. If not provided, a random ID will be generated."
"provider_id": {
"type": "string"
}, },
"metadata": { "metadata": {
"type": "object", "type": "object",
@ -9489,12 +9496,12 @@
"type": "object" "type": "object"
} }
] ]
} },
"description": "(Optional) Metadata for this benchmark for additional descriptions."
} }
}, },
"additionalProperties": false, "additionalProperties": false,
"required": [ "required": [
"benchmark_id",
"dataset_id", "dataset_id",
"scoring_functions" "scoring_functions"
], ],

32
docs/_static/llama-stack-spec.yaml vendored
View file

@ -1635,6 +1635,10 @@ paths:
responses: responses:
'200': '200':
description: OK description: OK
content:
application/json:
schema:
$ref: '#/components/schemas/Benchmark'
'400': '400':
$ref: '#/components/responses/BadRequest400' $ref: '#/components/responses/BadRequest400'
'429': '429':
@ -1647,7 +1651,7 @@ paths:
$ref: '#/components/responses/DefaultError' $ref: '#/components/responses/DefaultError'
tags: tags:
- Benchmarks - Benchmarks
description: '' description: Register a new benchmark.
parameters: [] parameters: []
requestBody: requestBody:
content: content:
@ -4950,10 +4954,14 @@ components:
default: benchmark default: benchmark
dataset_id: dataset_id:
type: string type: string
description: >-
The ID of the dataset to used to run the benchmark.
scoring_functions: scoring_functions:
type: array type: array
items: items:
type: string $ref: '#/components/schemas/ScoringFnParams'
description: >-
The scoring functions with parameters to use for this benchmark.
metadata: metadata:
type: object type: object
additionalProperties: additionalProperties:
@ -4964,6 +4972,8 @@ components:
- type: string - type: string
- type: array - type: array
- type: object - type: object
description: >-
Metadata for this benchmark for additional descriptions.
additionalProperties: false additionalProperties: false
required: required:
- identifier - identifier
@ -6438,18 +6448,21 @@ components:
RegisterBenchmarkRequest: RegisterBenchmarkRequest:
type: object type: object
properties: properties:
benchmark_id:
type: string
dataset_id: dataset_id:
type: string type: string
description: >-
The ID of the dataset to used to run the benchmark.
scoring_functions: scoring_functions:
type: array type: array
items: items:
type: string $ref: '#/components/schemas/ScoringFnParams'
provider_benchmark_id: description: >-
type: string The scoring functions with parameters to use for this benchmark.
provider_id: benchmark_id:
type: string type: string
description: >-
(Optional) The ID of the benchmark to register. If not provided, a random
ID will be generated.
metadata: metadata:
type: object type: object
additionalProperties: additionalProperties:
@ -6460,9 +6473,10 @@ components:
- type: string - type: string
- type: array - type: array
- type: object - type: object
description: >-
(Optional) Metadata for this benchmark for additional descriptions.
additionalProperties: false additionalProperties: false
required: required:
- benchmark_id
- dataset_id - dataset_id
- scoring_functions - scoring_functions
title: RegisterBenchmarkRequest title: RegisterBenchmarkRequest

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 pydantic import BaseModel, Field
from llama_stack.apis.resource import Resource, ResourceType 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 from llama_stack.schema_utils import json_schema_type, webmethod
class CommonBenchmarkFields(BaseModel): 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 dataset_id: str
scoring_functions: List[str] scoring_functions: List[ScoringFnParams]
metadata: Dict[str, Any] = Field( metadata: Dict[str, Any] = Field(
default_factory=dict, 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") @webmethod(route="/eval/benchmarks", method="POST")
async def register_benchmark( async def register_benchmark(
self, self,
benchmark_id: str,
dataset_id: str, dataset_id: str,
scoring_functions: List[str], scoring_functions: List[ScoringFnParams],
provider_benchmark_id: Optional[str] = None, benchmark_id: Optional[str] = None,
provider_id: Optional[str] = None,
metadata: Optional[Dict[str, Any]] = 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.
"""
...