llama-stack/llama_stack/providers/remote/datasetio/nvidia/datasetio.py
Rashmi Pawar e6bbf8d20b
feat: Add NVIDIA NeMo datastore (#1852)
# What does this PR do?
Implemetation of NeMO Datastore register, unregister API.

Open Issues: 
- provider_id gets set to `localfs` in client.datasets.register() as it
is specified in routing_tables.py: DatasetsRoutingTable
see: #1860

Currently I have passed `"provider_id":"nvidia"` in metadata and have
parsed that in `DatasetsRoutingTable`
(Not the best approach, but just a quick workaround to make it work for
now.)

## Test Plan
- Unit test cases: `pytest
tests/unit/providers/nvidia/test_datastore.py`
```bash
========================================================== test session starts ===========================================================
platform linux -- Python 3.10.0, pytest-8.3.5, pluggy-1.5.0
rootdir: /home/ubuntu/llama-stack
configfile: pyproject.toml
plugins: anyio-4.9.0, asyncio-0.26.0, nbval-0.11.0, metadata-3.1.1, html-4.1.1, cov-6.1.0
asyncio: mode=strict, asyncio_default_fixture_loop_scope=None, asyncio_default_test_loop_scope=function
collected 2 items                                                                                                                        

tests/unit/providers/nvidia/test_datastore.py ..                                                                                   [100%]

============================================================ warnings summary ============================================================

====================================================== 2 passed, 1 warning in 0.84s ======================================================
```

cc: @dglogo, @mattf, @yanxi0830
2025-04-28 09:41:59 -07:00

112 lines
3.7 KiB
Python

# 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 Any, Dict, List, Optional
import aiohttp
from llama_stack.apis.common.content_types import URL
from llama_stack.apis.common.responses import PaginatedResponse
from llama_stack.apis.common.type_system import ParamType
from llama_stack.apis.datasets import Dataset
from .config import NvidiaDatasetIOConfig
class NvidiaDatasetIOAdapter:
"""Nvidia NeMo DatasetIO API."""
def __init__(self, config: NvidiaDatasetIOConfig):
self.config = config
self.headers = {}
async def _make_request(
self,
method: str,
path: str,
headers: Optional[Dict[str, Any]] = None,
params: Optional[Dict[str, Any]] = None,
json: Optional[Dict[str, Any]] = None,
**kwargs,
) -> Dict[str, Any]:
"""Helper method to make HTTP requests to the Customizer API."""
url = f"{self.config.datasets_url}{path}"
request_headers = self.headers.copy()
if headers:
request_headers.update(headers)
async with aiohttp.ClientSession(headers=request_headers) as session:
async with session.request(method, url, params=params, json=json, **kwargs) as response:
if response.status != 200:
error_data = await response.json()
raise Exception(f"API request failed: {error_data}")
return await response.json()
async def register_dataset(
self,
dataset_def: Dataset,
) -> Dataset:
"""Register a new dataset.
Args:
dataset_def [Dataset]: The dataset definition.
dataset_id [str]: The ID of the dataset.
source [DataSource]: The source of the dataset.
metadata [Dict[str, Any]]: The metadata of the dataset.
format [str]: The format of the dataset.
description [str]: The description of the dataset.
Returns:
Dataset
"""
## add warnings for unsupported params
request_body = {
"name": dataset_def.identifier,
"namespace": self.config.dataset_namespace,
"files_url": dataset_def.source.uri,
"project": self.config.project_id,
}
if dataset_def.metadata:
request_body["format"] = dataset_def.metadata.get("format")
request_body["description"] = dataset_def.metadata.get("description")
await self._make_request(
"POST",
"/v1/datasets",
json=request_body,
)
return dataset_def
async def update_dataset(
self,
dataset_id: str,
dataset_schema: Dict[str, ParamType],
url: URL,
provider_dataset_id: Optional[str] = None,
provider_id: Optional[str] = None,
metadata: Optional[Dict[str, Any]] = None,
) -> None:
raise NotImplementedError("Not implemented")
async def unregister_dataset(
self,
dataset_id: str,
) -> None:
await self._make_request(
"DELETE",
f"/v1/datasets/{self.config.dataset_namespace}/{dataset_id}",
headers={"Accept": "application/json", "Content-Type": "application/json"},
)
async def iterrows(
self,
dataset_id: str,
start_index: Optional[int] = None,
limit: Optional[int] = None,
) -> PaginatedResponse:
raise NotImplementedError("Not implemented")
async def append_rows(self, dataset_id: str, rows: List[Dict[str, Any]]) -> None:
raise NotImplementedError("Not implemented")