datasets api

llama_stack/apis/datasets/__init__.py

@@ -4,4 +4,4 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from .dataset import *  # noqa: F401 F403
+from .datasets import *  # noqa: F401 F403

llama_stack/apis/datasets/client.py

@@ -0,0 +1,92 @@
+# 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.
+
+import asyncio
+import json
+
+import fire
+import httpx
+
+from .datasets import *  # noqa: F403
+
+
+class DatasetClient(Datasets):
+    def __init__(self, base_url: str):
+        self.base_url = base_url
+
+    async def initialize(self) -> None:
+        pass
+
+    async def shutdown(self) -> None:
+        pass
+
+    async def create_dataset(
+        self,
+        dataset_def: DatasetDef,
+    ) -> None:
+        async with httpx.AsyncClient() as client:
+            response = await client.post(
+                f"{self.base_url}/datasets/create",
+                json={
+                    "dataset_def": json.loads(dataset_def.json()),
+                },
+                headers={"Content-Type": "application/json"},
+                timeout=60,
+            )
+            response.raise_for_status()
+            return None
+
+    async def get_dataset(
+        self,
+        dataset_identifier: str,
+    ) -> DatasetDef:
+        async with httpx.AsyncClient() as client:
+            response = await client.post(
+                f"{self.base_url}/datasets/create",
+                json={
+                    "dataset_identifier": dataset_identifier,
+                },
+                headers={"Content-Type": "application/json"},
+                timeout=60,
+            )
+            response.raise_for_status()
+            return DatasetDef(**response.json())
+
+    async def delete_dataset(
+        self,
+        dataset_identifier: str,
+    ) -> DatasetDef:
+        async with httpx.AsyncClient() as client:
+            response = await client.post(
+                f"{self.base_url}/datasets/delete",
+                json={
+                    "dataset_identifier": dataset_identifier,
+                },
+                headers={"Content-Type": "application/json"},
+                timeout=60,
+            )
+            response.raise_for_status()
+            return None
+
+
+async def run_main(host: str, port: int):
+    client = DatasetClient(f"http://{host}:{port}")
+
+    # Custom Eval Task
+    response = await client.create_dataset(
+        dataset_def=CustomDatasetDef(
+            identifier="test-dataset",
+            url="https://openaipublic.blob.core.windows.net/simple-evals/mmlu.csv",
+        ),
+    )
+
+
+def main(host: str, port: int):
+    asyncio.run(run_main(host, port))
+
+
+if __name__ == "__main__":
+    fire.Fire(main)

llama_stack/apis/datasets/datasets.py

@@ -143,19 +143,19 @@ class BaseDataset(ABC, Generic[TDatasetSample]):
 
 class Datasets(Protocol):
     @webmethod(route="/datasets/create")
-    def create_dataset(
+    async def create_dataset(
         self,
-        dataset: DatasetDef,
+        dataset_def: DatasetDef,
     ) -> None: ...
 
     @webmethod(route="/datasets/get")
-    def get_dataset(
+    async def get_dataset(
         self,
         dataset_identifier: str,
     ) -> DatasetDef: ...
 
     @webmethod(route="/datasets/delete")
-    def delete_dataset(
+    async def delete_dataset(
         self,
-        dataset_uuid: str,
+        dataset_identifier: str,
     ) -> None: ...

llama_stack/apis/evals/evals.py

@@ -11,7 +11,7 @@ from llama_models.schema_utils import webmethod
 from pydantic import BaseModel
 
 from llama_models.llama3.api.datatypes import *  # noqa: F403
-from llama_stack.apis.dataset import *  # noqa: F403
+from llama_stack.apis.datasets import *  # noqa: F403
 
 
 class EvaluationJob(BaseModel):