llama-stack-api 0.4.3__py3-none-any.whl → 0.5.0rc1__py3-none-any.whl

llama_stack_api/datasets/__init__.py

@@ -0,0 +1,61 @@
+# 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.
+
+"""Datasets API protocol and models.
+
+This module contains the Datasets protocol definition.
+Pydantic models are defined in llama_stack_api.datasets.models.
+The FastAPI router is defined in llama_stack_api.datasets.fastapi_routes.
+"""
+
+# Import fastapi_routes for router factory access
+from . import fastapi_routes
+
+# Import new protocol for FastAPI router
+from .api import Datasets
+
+# Import models for re-export
+from .models import (
+    CommonDatasetFields,
+    Dataset,
+    DatasetPurpose,
+    DatasetType,
+    DataSource,
+    GetDatasetRequest,
+    ListDatasetsResponse,
+    RegisterDatasetRequest,
+    RowsDataSource,
+    UnregisterDatasetRequest,
+    URIDataSource,
+)
+
+
+# Define DatasetInput for backward compatibility
+class DatasetInput(CommonDatasetFields):
+    """Input parameters for dataset operations.
+
+    :param dataset_id: Unique identifier for the dataset
+    """
+
+    dataset_id: str
+
+
+__all__ = [
+    "Datasets",
+    "Dataset",
+    "CommonDatasetFields",
+    "DatasetPurpose",
+    "DataSource",
+    "DatasetInput",
+    "DatasetType",
+    "RowsDataSource",
+    "URIDataSource",
+    "ListDatasetsResponse",
+    "RegisterDatasetRequest",
+    "GetDatasetRequest",
+    "UnregisterDatasetRequest",
+    "fastapi_routes",
+]

llama_stack_api/datasets/api.py

@@ -0,0 +1,35 @@
+# 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.
+
+"""Datasets API protocol definition.
+
+This module contains the Datasets protocol definition.
+Pydantic models are defined in llama_stack_api.datasets.models.
+The FastAPI router is defined in llama_stack_api.datasets.fastapi_routes.
+"""
+
+from typing import Protocol, runtime_checkable
+
+from .models import (
+    Dataset,
+    GetDatasetRequest,
+    ListDatasetsResponse,
+    RegisterDatasetRequest,
+    UnregisterDatasetRequest,
+)
+
+
+@runtime_checkable
+class Datasets(Protocol):
+    """Protocol for dataset management operations."""
+
+    async def register_dataset(self, request: RegisterDatasetRequest) -> Dataset: ...
+
+    async def get_dataset(self, request: GetDatasetRequest) -> Dataset: ...
+
+    async def list_datasets(self) -> ListDatasetsResponse: ...
+
+    async def unregister_dataset(self, request: UnregisterDatasetRequest) -> None: ...

llama_stack_api/datasets/fastapi_routes.py

@@ -0,0 +1,104 @@
+# 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.
+
+"""FastAPI router for the Datasets API.
+
+This module defines the FastAPI router for the Datasets API using standard
+FastAPI route decorators.
+"""
+
+from typing import Annotated
+
+from fastapi import APIRouter, Body, Depends
+
+from llama_stack_api.router_utils import create_path_dependency, standard_responses
+from llama_stack_api.version import LLAMA_STACK_API_V1BETA
+
+from .api import Datasets
+from .models import (
+    Dataset,
+    GetDatasetRequest,
+    ListDatasetsResponse,
+    RegisterDatasetRequest,
+    UnregisterDatasetRequest,
+)
+
+# Path parameter dependencies for single-field models
+get_dataset_request = create_path_dependency(GetDatasetRequest)
+unregister_dataset_request = create_path_dependency(UnregisterDatasetRequest)
+
+
+def create_router(impl: Datasets) -> APIRouter:
+    """Create a FastAPI router for the Datasets API.
+
+    Args:
+        impl: The Datasets implementation instance
+
+    Returns:
+        APIRouter configured for the Datasets API
+    """
+    router = APIRouter(
+        prefix=f"/{LLAMA_STACK_API_V1BETA}",
+        tags=["Datasets"],
+        responses=standard_responses,
+    )
+
+    @router.post(
+        "/datasets",
+        response_model=Dataset,
+        summary="Register a new dataset.",
+        description="Register a new dataset.",
+        responses={
+            200: {"description": "The registered dataset object."},
+        },
+        deprecated=True,
+    )
+    async def register_dataset(
+        request: Annotated[RegisterDatasetRequest, Body(...)],
+    ) -> Dataset:
+        return await impl.register_dataset(request)
+
+    @router.get(
+        "/datasets/{dataset_id:path}",
+        response_model=Dataset,
+        summary="Get a dataset by its ID.",
+        description="Get a dataset by its ID.",
+        responses={
+            200: {"description": "The dataset object."},
+        },
+    )
+    async def get_dataset(
+        request: Annotated[GetDatasetRequest, Depends(get_dataset_request)],
+    ) -> Dataset:
+        return await impl.get_dataset(request)
+
+    @router.get(
+        "/datasets",
+        response_model=ListDatasetsResponse,
+        summary="List all datasets.",
+        description="List all datasets.",
+        responses={
+            200: {"description": "A list of dataset objects."},
+        },
+    )
+    async def list_datasets() -> ListDatasetsResponse:
+        return await impl.list_datasets()
+
+    @router.delete(
+        "/datasets/{dataset_id:path}",
+        summary="Unregister a dataset by its ID.",
+        description="Unregister a dataset by its ID.",
+        responses={
+            200: {"description": "The dataset was successfully unregistered."},
+        },
+        deprecated=True,
+    )
+    async def unregister_dataset(
+        request: Annotated[UnregisterDatasetRequest, Depends(unregister_dataset_request)],
+    ) -> None:
+        return await impl.unregister_dataset(request)
+
+    return router

llama_stack_api/datasets/models.py

@@ -0,0 +1,152 @@
+# 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.
+
+"""Pydantic models for Datasets API requests and responses.
+
+This module defines the request and response models for the Datasets API
+using Pydantic with Field descriptions for OpenAPI schema generation.
+"""
+
+from enum import Enum, StrEnum
+from typing import Annotated, Any, Literal
+
+from pydantic import BaseModel, Field
+
+from llama_stack_api.resource import Resource, ResourceType
+from llama_stack_api.schema_utils import json_schema_type, register_schema
+
+
+class DatasetPurpose(StrEnum):
+    """Purpose of the dataset. Each purpose has a required input data schema."""
+
+    post_training_messages = "post-training/messages"
+    """The dataset contains messages used for post-training."""
+    eval_question_answer = "eval/question-answer"
+    """The dataset contains a question column and an answer column."""
+    eval_messages_answer = "eval/messages-answer"
+    """The dataset contains a messages column with list of messages and an answer column."""
+
+
+class DatasetType(Enum):
+    """Type of the dataset source."""
+
+    uri = "uri"
+    """The dataset can be obtained from a URI."""
+    rows = "rows"
+    """The dataset is stored in rows."""
+
+
+@json_schema_type
+class URIDataSource(BaseModel):
+    """A dataset that can be obtained from a URI."""
+
+    type: Literal["uri"] = Field(default="uri", description="The type of data source.")
+    uri: str = Field(
+        ...,
+        description='The dataset can be obtained from a URI. E.g. "https://mywebsite.com/mydata.jsonl", "lsfs://mydata.jsonl", "data:csv;base64,{base64_content}"',
+    )
+
+
+@json_schema_type
+class RowsDataSource(BaseModel):
+    """A dataset stored in rows."""
+
+    type: Literal["rows"] = Field(default="rows", description="The type of data source.")
+    rows: list[dict[str, Any]] = Field(
+        ...,
+        description='The dataset is stored in rows. E.g. [{"messages": [{"role": "user", "content": "Hello, world!"}, {"role": "assistant", "content": "Hello, world!"}]}]',
+    )
+
+
+DataSource = Annotated[
+    URIDataSource | RowsDataSource,
+    Field(discriminator="type"),
+]
+register_schema(DataSource, name="DataSource")
+
+
+class CommonDatasetFields(BaseModel):
+    """Common fields for a dataset."""
+
+    purpose: DatasetPurpose = Field(..., description="Purpose of the dataset indicating its intended use")
+    source: DataSource = Field(..., description="Data source configuration for the dataset")
+    metadata: dict[str, Any] = Field(
+        default_factory=dict,
+        description="Any additional metadata for this dataset",
+    )
+
+
+@json_schema_type
+class Dataset(CommonDatasetFields, Resource):
+    """Dataset resource for storing and accessing training or evaluation data."""
+
+    type: Literal[ResourceType.dataset] = Field(
+        default=ResourceType.dataset,
+        description="Type of resource, always 'dataset' for datasets",
+    )
+
+    @property
+    def dataset_id(self) -> str:
+        return self.identifier
+
+    @property
+    def provider_dataset_id(self) -> str | None:
+        return self.provider_resource_id
+
+
+@json_schema_type
+class ListDatasetsResponse(BaseModel):
+    """Response from listing datasets."""
+
+    data: list[Dataset] = Field(..., description="List of datasets")
+
+
+# Request models for each endpoint
+
+
+@json_schema_type
+class RegisterDatasetRequest(BaseModel):
+    """Request model for registering a dataset."""
+
+    purpose: DatasetPurpose = Field(..., description="The purpose of the dataset.")
+    source: DataSource = Field(..., description="The data source of the dataset.")
+    metadata: dict[str, Any] | None = Field(
+        default=None,
+        description="The metadata for the dataset.",
+    )
+    dataset_id: str | None = Field(
+        default=None,
+        description="The ID of the dataset. If not provided, an ID will be generated.",
+    )
+
+
+@json_schema_type
+class GetDatasetRequest(BaseModel):
+    """Request model for getting a dataset by ID."""
+
+    dataset_id: str = Field(..., description="The ID of the dataset to get.")
+
+
+@json_schema_type
+class UnregisterDatasetRequest(BaseModel):
+    """Request model for unregistering a dataset."""
+
+    dataset_id: str = Field(..., description="The ID of the dataset to unregister.")
+
+
+__all__ = [
+    "CommonDatasetFields",
+    "Dataset",
+    "DatasetPurpose",
+    "DatasetType",
+    "DataSource",
+    "RowsDataSource",
+    "URIDataSource",
+    "ListDatasetsResponse",
+    "RegisterDatasetRequest",
+    "GetDatasetRequest",
+    "UnregisterDatasetRequest",
+]