forked from phoenix-oss/llama-stack-mirror
feat(eval api): (2.3/n) remove scoring / eval impls + benchmarks (#1766)
# What does this PR do? - Remove `/eval` and `/scoring` impls - Clean up benchmarks. The benchmarks exists in the `llama-stack-evals` repo. - Rest of grading functions will be added in follow up PR. [//]: # (If resolving an issue, uncomment and update the line below) [//]: # (Closes #[issue-number]) ## Test Plan - CI [//]: # (## Documentation)
This commit is contained in:
Xi Yan
GitHub
parent
97e7717c9b
commit
7f12ea290f
57 changed files with 0 additions and 7691 deletions
|
|
@ -1,7 +0,0 @@
|
|||
# Copyright (c) Meta Platforms, Inc. and affiliates.
|
||||
# All rights reserved.
|
||||
#
|
||||
# This source code is licensed under the terms described in the LICENSE file in
|
||||
# the root directory of this source tree.
|
||||
|
||||
from .open_benchmark import get_distribution_template # noqa: F401
|
||||
|
|
@ -1,30 +0,0 @@
|
|||
version: '2'
|
||||
distribution_spec:
|
||||
description: Distribution for running open benchmarks
|
||||
providers:
|
||||
inference:
|
||||
- remote::openai
|
||||
- remote::anthropic
|
||||
- remote::gemini
|
||||
- remote::groq
|
||||
- remote::together
|
||||
vector_io:
|
||||
- inline::sqlite-vec
|
||||
- remote::chromadb
|
||||
- remote::pgvector
|
||||
safety:
|
||||
- inline::llama-guard
|
||||
agents:
|
||||
- inline::meta-reference
|
||||
telemetry:
|
||||
- inline::meta-reference
|
||||
datasetio:
|
||||
- remote::huggingface
|
||||
- inline::localfs
|
||||
tool_runtime:
|
||||
- remote::brave-search
|
||||
- remote::tavily-search
|
||||
- inline::code-interpreter
|
||||
- inline::rag-runtime
|
||||
- remote::model-context-protocol
|
||||
image_type: conda
|
||||
|
|
@ -1,304 +0,0 @@
|
|||
# Copyright (c) Meta Platforms, Inc. and affiliates.
|
||||
# All rights reserved.
|
||||
#
|
||||
# This source code is licensed under the terms described in the LICENSE file in
|
||||
# the root directory of this source tree.
|
||||
|
||||
from typing import Dict, List, Tuple
|
||||
|
||||
from llama_stack.apis.datasets import DatasetPurpose, URIDataSource
|
||||
from llama_stack.apis.models.models import ModelType
|
||||
from llama_stack.distribution.datatypes import (
|
||||
DatasetInput,
|
||||
ModelInput,
|
||||
Provider,
|
||||
ShieldInput,
|
||||
ToolGroupInput,
|
||||
)
|
||||
from llama_stack.providers.inline.vector_io.sqlite_vec.config import (
|
||||
SQLiteVectorIOConfig,
|
||||
)
|
||||
from llama_stack.providers.remote.inference.anthropic.config import AnthropicConfig
|
||||
from llama_stack.providers.remote.inference.gemini.config import GeminiConfig
|
||||
from llama_stack.providers.remote.inference.groq.config import GroqConfig
|
||||
from llama_stack.providers.remote.inference.openai.config import OpenAIConfig
|
||||
from llama_stack.providers.remote.inference.together.config import TogetherImplConfig
|
||||
from llama_stack.providers.remote.vector_io.chroma.config import ChromaVectorIOConfig
|
||||
from llama_stack.providers.remote.vector_io.pgvector.config import (
|
||||
PGVectorVectorIOConfig,
|
||||
)
|
||||
from llama_stack.providers.utils.inference.model_registry import ProviderModelEntry
|
||||
from llama_stack.templates.template import (
|
||||
DistributionTemplate,
|
||||
RunConfigSettings,
|
||||
get_model_registry,
|
||||
)
|
||||
|
||||
|
||||
def get_inference_providers() -> Tuple[List[Provider], Dict[str, List[ProviderModelEntry]]]:
|
||||
# in this template, we allow each API key to be optional
|
||||
providers = [
|
||||
(
|
||||
"openai",
|
||||
[
|
||||
ProviderModelEntry(
|
||||
provider_model_id="openai/gpt-4o",
|
||||
model_type=ModelType.llm,
|
||||
)
|
||||
],
|
||||
OpenAIConfig.sample_run_config(api_key="${env.OPENAI_API_KEY:}"),
|
||||
),
|
||||
(
|
||||
"anthropic",
|
||||
[
|
||||
ProviderModelEntry(
|
||||
provider_model_id="anthropic/claude-3-5-sonnet-latest",
|
||||
model_type=ModelType.llm,
|
||||
)
|
||||
],
|
||||
AnthropicConfig.sample_run_config(api_key="${env.ANTHROPIC_API_KEY:}"),
|
||||
),
|
||||
(
|
||||
"gemini",
|
||||
[
|
||||
ProviderModelEntry(
|
||||
provider_model_id="gemini/gemini-1.5-flash",
|
||||
model_type=ModelType.llm,
|
||||
)
|
||||
],
|
||||
GeminiConfig.sample_run_config(api_key="${env.GEMINI_API_KEY:}"),
|
||||
),
|
||||
(
|
||||
"groq",
|
||||
[],
|
||||
GroqConfig.sample_run_config(api_key="${env.GROQ_API_KEY:}"),
|
||||
),
|
||||
(
|
||||
"together",
|
||||
[],
|
||||
TogetherImplConfig.sample_run_config(api_key="${env.TOGETHER_API_KEY:}"),
|
||||
),
|
||||
]
|
||||
inference_providers = []
|
||||
available_models = {}
|
||||
for provider_id, model_entries, config in providers:
|
||||
inference_providers.append(
|
||||
Provider(
|
||||
provider_id=provider_id,
|
||||
provider_type=f"remote::{provider_id}",
|
||||
config=config,
|
||||
)
|
||||
)
|
||||
available_models[provider_id] = model_entries
|
||||
return inference_providers, available_models
|
||||
|
||||
|
||||
def get_distribution_template() -> DistributionTemplate:
|
||||
inference_providers, available_models = get_inference_providers()
|
||||
providers = {
|
||||
"inference": [p.provider_type for p in inference_providers],
|
||||
"vector_io": ["inline::sqlite-vec", "remote::chromadb", "remote::pgvector"],
|
||||
"safety": ["inline::llama-guard"],
|
||||
"agents": ["inline::meta-reference"],
|
||||
"telemetry": ["inline::meta-reference"],
|
||||
"datasetio": ["remote::huggingface", "inline::localfs"],
|
||||
"tool_runtime": [
|
||||
"remote::brave-search",
|
||||
"remote::tavily-search",
|
||||
"inline::code-interpreter",
|
||||
"inline::rag-runtime",
|
||||
"remote::model-context-protocol",
|
||||
],
|
||||
}
|
||||
name = "open-benchmark"
|
||||
|
||||
vector_io_providers = [
|
||||
Provider(
|
||||
provider_id="sqlite-vec",
|
||||
provider_type="inline::sqlite-vec",
|
||||
config=SQLiteVectorIOConfig.sample_run_config(f"~/.llama/distributions/{name}"),
|
||||
),
|
||||
Provider(
|
||||
provider_id="${env.ENABLE_CHROMADB+chromadb}",
|
||||
provider_type="remote::chromadb",
|
||||
config=ChromaVectorIOConfig.sample_run_config(url="${env.CHROMADB_URL:}"),
|
||||
),
|
||||
Provider(
|
||||
provider_id="${env.ENABLE_PGVECTOR+pgvector}",
|
||||
provider_type="remote::pgvector",
|
||||
config=PGVectorVectorIOConfig.sample_run_config(
|
||||
db="${env.PGVECTOR_DB:}",
|
||||
user="${env.PGVECTOR_USER:}",
|
||||
password="${env.PGVECTOR_PASSWORD:}",
|
||||
),
|
||||
),
|
||||
]
|
||||
|
||||
default_tool_groups = [
|
||||
ToolGroupInput(
|
||||
toolgroup_id="builtin::websearch",
|
||||
provider_id="tavily-search",
|
||||
),
|
||||
ToolGroupInput(
|
||||
toolgroup_id="builtin::rag",
|
||||
provider_id="rag-runtime",
|
||||
),
|
||||
ToolGroupInput(
|
||||
toolgroup_id="builtin::code_interpreter",
|
||||
provider_id="code-interpreter",
|
||||
),
|
||||
]
|
||||
|
||||
default_models = get_model_registry(available_models) + [
|
||||
ModelInput(
|
||||
model_id="meta-llama/Llama-3.3-70B-Instruct",
|
||||
provider_id="groq",
|
||||
provider_model_id="groq/llama-3.3-70b-versatile",
|
||||
model_type=ModelType.llm,
|
||||
),
|
||||
ModelInput(
|
||||
model_id="meta-llama/Llama-3.1-405B-Instruct",
|
||||
provider_id="together",
|
||||
provider_model_id="meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo",
|
||||
model_type=ModelType.llm,
|
||||
),
|
||||
]
|
||||
|
||||
default_datasets = [
|
||||
DatasetInput(
|
||||
dataset_id="simpleqa",
|
||||
purpose=DatasetPurpose.eval_messages_answer,
|
||||
source=URIDataSource(
|
||||
uri="huggingface://datasets/llamastack/simpleqa?split=train",
|
||||
),
|
||||
),
|
||||
DatasetInput(
|
||||
dataset_id="mmlu_cot",
|
||||
purpose=DatasetPurpose.eval_messages_answer,
|
||||
source=URIDataSource(
|
||||
uri="huggingface://datasets/llamastack/mmlu_cot?split=test&name=all",
|
||||
),
|
||||
),
|
||||
DatasetInput(
|
||||
dataset_id="gpqa_cot",
|
||||
purpose=DatasetPurpose.eval_messages_answer,
|
||||
source=URIDataSource(
|
||||
uri="huggingface://datasets/llamastack/gpqa_0shot_cot?split=test&name=gpqa_main",
|
||||
),
|
||||
),
|
||||
DatasetInput(
|
||||
dataset_id="math_500",
|
||||
purpose=DatasetPurpose.eval_messages_answer,
|
||||
source=URIDataSource(
|
||||
uri="huggingface://datasets/llamastack/math_500?split=test",
|
||||
),
|
||||
),
|
||||
DatasetInput(
|
||||
dataset_id="bfcl",
|
||||
purpose=DatasetPurpose.eval_messages_answer,
|
||||
source=URIDataSource(
|
||||
uri="huggingface://datasets/llamastack/bfcl_v3?split=train",
|
||||
),
|
||||
),
|
||||
DatasetInput(
|
||||
dataset_id="ifeval",
|
||||
purpose=DatasetPurpose.eval_messages_answer,
|
||||
source=URIDataSource(
|
||||
uri="huggingface://datasets/llamastack/IfEval?split=train",
|
||||
),
|
||||
),
|
||||
DatasetInput(
|
||||
dataset_id="docvqa",
|
||||
purpose=DatasetPurpose.eval_messages_answer,
|
||||
source=URIDataSource(
|
||||
uri="huggingface://datasets/llamastack/docvqa?split=val",
|
||||
),
|
||||
),
|
||||
]
|
||||
|
||||
# TODO(xiyan): fix this back as registerable resources
|
||||
# default_benchmarks = [
|
||||
# BenchmarkInput(
|
||||
# benchmark_id="meta-reference-simpleqa",
|
||||
# dataset_id="simpleqa",
|
||||
# grader_ids=["llm-as-judge::405b-simpleqa"],
|
||||
# ),
|
||||
# BenchmarkInput(
|
||||
# benchmark_id="meta-reference-mmlu-cot",
|
||||
# dataset_id="mmlu_cot",
|
||||
# grader_ids=["basic::regex_parser_multiple_choice_answer"],
|
||||
# ),
|
||||
# BenchmarkInput(
|
||||
# benchmark_id="meta-reference-gpqa-cot",
|
||||
# dataset_id="gpqa_cot",
|
||||
# grader_ids=["basic::regex_parser_multiple_choice_answer"],
|
||||
# ),
|
||||
# BenchmarkInput(
|
||||
# benchmark_id="meta-reference-math-500",
|
||||
# dataset_id="math_500",
|
||||
# grader_ids=["basic::regex_parser_math_response"],
|
||||
# ),
|
||||
# BenchmarkInput(
|
||||
# benchmark_id="meta-reference-bfcl",
|
||||
# dataset_id="bfcl",
|
||||
# grader_ids=["basic::bfcl"],
|
||||
# ),
|
||||
# BenchmarkInput(
|
||||
# benchmark_id="meta-reference-ifeval",
|
||||
# dataset_id="ifeval",
|
||||
# grader_ids=["basic::ifeval"],
|
||||
# ),
|
||||
# BenchmarkInput(
|
||||
# benchmark_id="meta-reference-docvqa",
|
||||
# dataset_id="docvqa",
|
||||
# grader_ids=["basic::docvqa"],
|
||||
# ),
|
||||
# ]
|
||||
|
||||
return DistributionTemplate(
|
||||
name=name,
|
||||
distro_type="self_hosted",
|
||||
description="Distribution for running open benchmarks",
|
||||
container_image=None,
|
||||
template_path=None,
|
||||
providers=providers,
|
||||
available_models_by_provider=available_models,
|
||||
run_configs={
|
||||
"run.yaml": RunConfigSettings(
|
||||
provider_overrides={
|
||||
"inference": inference_providers,
|
||||
"vector_io": vector_io_providers,
|
||||
},
|
||||
default_models=default_models,
|
||||
default_tool_groups=default_tool_groups,
|
||||
default_shields=[ShieldInput(shield_id="meta-llama/Llama-Guard-3-8B")],
|
||||
default_datasets=default_datasets,
|
||||
),
|
||||
},
|
||||
run_config_env_vars={
|
||||
"LLAMA_STACK_PORT": (
|
||||
"8321",
|
||||
"Port for the Llama Stack distribution server",
|
||||
),
|
||||
"TOGETHER_API_KEY": (
|
||||
"",
|
||||
"Together API Key",
|
||||
),
|
||||
"OPENAI_API_KEY": (
|
||||
"",
|
||||
"OpenAI API Key",
|
||||
),
|
||||
"GEMINI_API_KEY": (
|
||||
"",
|
||||
"Gemini API Key",
|
||||
),
|
||||
"ANTHROPIC_API_KEY": (
|
||||
"",
|
||||
"Anthropic API Key",
|
||||
),
|
||||
"GROQ_API_KEY": (
|
||||
"",
|
||||
"Groq API Key",
|
||||
),
|
||||
},
|
||||
)
|
||||
|
|
@ -1,190 +0,0 @@
|
|||
version: '2'
|
||||
image_name: open-benchmark
|
||||
apis:
|
||||
- agents
|
||||
- datasetio
|
||||
- inference
|
||||
- safety
|
||||
- telemetry
|
||||
- tool_runtime
|
||||
- vector_io
|
||||
providers:
|
||||
inference:
|
||||
- provider_id: openai
|
||||
provider_type: remote::openai
|
||||
config:
|
||||
api_key: ${env.OPENAI_API_KEY:}
|
||||
- provider_id: anthropic
|
||||
provider_type: remote::anthropic
|
||||
config:
|
||||
api_key: ${env.ANTHROPIC_API_KEY:}
|
||||
- provider_id: gemini
|
||||
provider_type: remote::gemini
|
||||
config:
|
||||
api_key: ${env.GEMINI_API_KEY:}
|
||||
- provider_id: groq
|
||||
provider_type: remote::groq
|
||||
config:
|
||||
url: https://api.groq.com
|
||||
api_key: ${env.GROQ_API_KEY:}
|
||||
- provider_id: together
|
||||
provider_type: remote::together
|
||||
config:
|
||||
url: https://api.together.xyz/v1
|
||||
api_key: ${env.TOGETHER_API_KEY:}
|
||||
vector_io:
|
||||
- provider_id: sqlite-vec
|
||||
provider_type: inline::sqlite-vec
|
||||
config:
|
||||
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/open-benchmark}/sqlite_vec.db
|
||||
- provider_id: ${env.ENABLE_CHROMADB+chromadb}
|
||||
provider_type: remote::chromadb
|
||||
config:
|
||||
url: ${env.CHROMADB_URL:}
|
||||
- provider_id: ${env.ENABLE_PGVECTOR+pgvector}
|
||||
provider_type: remote::pgvector
|
||||
config:
|
||||
host: ${env.PGVECTOR_HOST:localhost}
|
||||
port: ${env.PGVECTOR_PORT:5432}
|
||||
db: ${env.PGVECTOR_DB:}
|
||||
user: ${env.PGVECTOR_USER:}
|
||||
password: ${env.PGVECTOR_PASSWORD:}
|
||||
safety:
|
||||
- provider_id: llama-guard
|
||||
provider_type: inline::llama-guard
|
||||
config:
|
||||
excluded_categories: []
|
||||
agents:
|
||||
- provider_id: meta-reference
|
||||
provider_type: inline::meta-reference
|
||||
config:
|
||||
persistence_store:
|
||||
type: sqlite
|
||||
namespace: null
|
||||
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/open-benchmark}/agents_store.db
|
||||
telemetry:
|
||||
- provider_id: meta-reference
|
||||
provider_type: inline::meta-reference
|
||||
config:
|
||||
sinks: ${env.TELEMETRY_SINKS:console,sqlite}
|
||||
sqlite_db_path: ${env.SQLITE_DB_PATH:~/.llama/distributions/open-benchmark/trace_store.db}
|
||||
datasetio:
|
||||
- provider_id: huggingface
|
||||
provider_type: remote::huggingface
|
||||
config:
|
||||
kvstore:
|
||||
type: sqlite
|
||||
namespace: null
|
||||
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/open-benchmark}/huggingface_datasetio.db
|
||||
- provider_id: localfs
|
||||
provider_type: inline::localfs
|
||||
config:
|
||||
kvstore:
|
||||
type: sqlite
|
||||
namespace: null
|
||||
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/open-benchmark}/localfs_datasetio.db
|
||||
tool_runtime:
|
||||
- provider_id: brave-search
|
||||
provider_type: remote::brave-search
|
||||
config:
|
||||
api_key: ${env.BRAVE_SEARCH_API_KEY:}
|
||||
max_results: 3
|
||||
- provider_id: tavily-search
|
||||
provider_type: remote::tavily-search
|
||||
config:
|
||||
api_key: ${env.TAVILY_SEARCH_API_KEY:}
|
||||
max_results: 3
|
||||
- provider_id: code-interpreter
|
||||
provider_type: inline::code-interpreter
|
||||
config: {}
|
||||
- provider_id: rag-runtime
|
||||
provider_type: inline::rag-runtime
|
||||
config: {}
|
||||
- provider_id: model-context-protocol
|
||||
provider_type: remote::model-context-protocol
|
||||
config: {}
|
||||
metadata_store:
|
||||
type: sqlite
|
||||
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/open-benchmark}/registry.db
|
||||
models:
|
||||
- metadata: {}
|
||||
model_id: openai/gpt-4o
|
||||
provider_id: openai
|
||||
provider_model_id: openai/gpt-4o
|
||||
model_type: llm
|
||||
- metadata: {}
|
||||
model_id: anthropic/claude-3-5-sonnet-latest
|
||||
provider_id: anthropic
|
||||
provider_model_id: anthropic/claude-3-5-sonnet-latest
|
||||
model_type: llm
|
||||
- metadata: {}
|
||||
model_id: gemini/gemini-1.5-flash
|
||||
provider_id: gemini
|
||||
provider_model_id: gemini/gemini-1.5-flash
|
||||
model_type: llm
|
||||
- metadata: {}
|
||||
model_id: meta-llama/Llama-3.3-70B-Instruct
|
||||
provider_id: groq
|
||||
provider_model_id: groq/llama-3.3-70b-versatile
|
||||
model_type: llm
|
||||
- metadata: {}
|
||||
model_id: meta-llama/Llama-3.1-405B-Instruct
|
||||
provider_id: together
|
||||
provider_model_id: meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo
|
||||
model_type: llm
|
||||
shields:
|
||||
- shield_id: meta-llama/Llama-Guard-3-8B
|
||||
vector_dbs: []
|
||||
datasets:
|
||||
- purpose: eval/messages-answer
|
||||
source:
|
||||
type: uri
|
||||
uri: huggingface://datasets/llamastack/simpleqa?split=train
|
||||
metadata: {}
|
||||
dataset_id: simpleqa
|
||||
- purpose: eval/messages-answer
|
||||
source:
|
||||
type: uri
|
||||
uri: huggingface://datasets/llamastack/mmlu_cot?split=test&name=all
|
||||
metadata: {}
|
||||
dataset_id: mmlu_cot
|
||||
- purpose: eval/messages-answer
|
||||
source:
|
||||
type: uri
|
||||
uri: huggingface://datasets/llamastack/gpqa_0shot_cot?split=test&name=gpqa_main
|
||||
metadata: {}
|
||||
dataset_id: gpqa_cot
|
||||
- purpose: eval/messages-answer
|
||||
source:
|
||||
type: uri
|
||||
uri: huggingface://datasets/llamastack/math_500?split=test
|
||||
metadata: {}
|
||||
dataset_id: math_500
|
||||
- purpose: eval/messages-answer
|
||||
source:
|
||||
type: uri
|
||||
uri: huggingface://datasets/llamastack/bfcl_v3?split=train
|
||||
metadata: {}
|
||||
dataset_id: bfcl
|
||||
- purpose: eval/messages-answer
|
||||
source:
|
||||
type: uri
|
||||
uri: huggingface://datasets/llamastack/IfEval?split=train
|
||||
metadata: {}
|
||||
dataset_id: ifeval
|
||||
- purpose: eval/messages-answer
|
||||
source:
|
||||
type: uri
|
||||
uri: huggingface://datasets/llamastack/docvqa?split=val
|
||||
metadata: {}
|
||||
dataset_id: docvqa
|
||||
benchmarks: []
|
||||
tool_groups:
|
||||
- toolgroup_id: builtin::websearch
|
||||
provider_id: tavily-search
|
||||
- toolgroup_id: builtin::rag
|
||||
provider_id: rag-runtime
|
||||
- toolgroup_id: builtin::code_interpreter
|
||||
provider_id: code-interpreter
|
||||
server:
|
||||
port: 8321
|
||||
Loading…
Add table
Add a link
Reference in a new issue