forked from phoenix-oss/llama-stack-mirror
# What does this PR do?
1) Implement `unregister_dataset(dataset_id)` API in both llama stack
routing table and providers: It removes {dataset_id -> Dataset} mapping
from routing table and removes the dataset_id references in provider as
well (ex. for huggingface, we use a KV store to store the dataset id =>
dataset. we delete it during unregistering as well)
2) expose the datasets/unregister_dataset api endpoint
## Test Plan
**Unit test:**
`
pytest llama_stack/providers/tests/datasetio/test_datasetio.py -m
"huggingface" -v -s --tb=short --disable-warnings
`
**Test on endpoint:**
tested llama stack using an ollama distribution template:
1) start an ollama server
2) Start a llama stack server with the default ollama distribution
config + dataset/datasetsio APIs + datasetio provider
```
---- .../ollama-run.yaml
...
apis:
- agents
- inference
- memory
- safety
- telemetry
- datasetio
- datasets
providers:
datasetio:
- provider_id: localfs
provider_type: inline::localfs
config: {}
...
```
saw that the new API showed up in startup script
```
Serving API datasets
GET /alpha/datasets/get
GET /alpha/datasets/list
POST /alpha/datasets/register
POST /alpha/datasets/unregister
```
3) query `/alpha/datasets/unregister` through curl (since we have not implemented unregister api in llama stack client)
```
(base) sxyi@sxyi-mbp llama-stack % llama-stack-client datasets register
--dataset-id sixian --url
https://raw.githubusercontent.com/pytorch/torchtune/main/docs/source/tutorials/chat.rst
--schema {}
(base) sxyi@sxyi-mbp llama-stack % llama-stack-client datasets list
┏━━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━┓
┃ identifier ┃ provider_id ┃ metadata ┃ type ┃
┡━━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━┩
│ sixian │ localfs │ {} │ dataset │
└────────────┴─────────────┴──────────┴─────────┘
(base) sxyi@sxyi-mbp llama-stack % llama-stack-client datasets register
--dataset-id sixian2 --url
https://raw.githubusercontent.com/pytorch/torchtune/main/docs/source/tutorials/chat.rst
--schema {}
(base) sxyi@sxyi-mbp llama-stack % llama-stack-client datasets list
┏━━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━┓
┃ identifier ┃ provider_id ┃ metadata ┃ type ┃
┡━━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━┩
│ sixian │ localfs │ {} │ dataset │
│ sixian2 │ localfs │ {} │ dataset │
└────────────┴─────────────┴──────────┴─────────┘
(base) sxyi@sxyi-mbp llama-stack % curl
http://localhost:5001/alpha/datasets/unregister \
-H "Content-Type: application/json" \
-d '{"dataset_id": "sixian"}'
null%
(base) sxyi@sxyi-mbp llama-stack % llama-stack-client datasets list
┏━━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━┓
┃ identifier ┃ provider_id ┃ metadata ┃ type ┃
┡━━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━┩
│ sixian2 │ localfs │ {} │ dataset │
└────────────┴─────────────┴──────────┴─────────┘
(base) sxyi@sxyi-mbp llama-stack % curl
http://localhost:5001/alpha/datasets/unregister \
-H "Content-Type: application/json" \
-d '{"dataset_id": "sixian2"}'
null%
(base) sxyi@sxyi-mbp llama-stack % llama-stack-client datasets list
```
## Sources
## Before submitting
- [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
- [ ] Ran pre-commit to handle lint / formatting issues.
- [ ] Read the [contributor guideline](https://github.com/meta-llama/llama-stack/blob/main/CONTRIBUTING.md),
Pull Request section?
- [ ] Updated relevant documentation.
- [ ] Wrote necessary unit or integration tests.
120 lines
4.1 KiB
Python
120 lines
4.1 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.
|
|
|
|
import os
|
|
|
|
import pytest
|
|
from llama_stack.apis.common.type_system import * # noqa: F403
|
|
from llama_stack.apis.datasetio import * # noqa: F403
|
|
from llama_stack.distribution.datatypes import * # noqa: F403
|
|
import base64
|
|
import mimetypes
|
|
from pathlib import Path
|
|
|
|
# How to run this test:
|
|
#
|
|
# pytest llama_stack/providers/tests/datasetio/test_datasetio.py
|
|
# -m "meta_reference"
|
|
# -v -s --tb=short --disable-warnings
|
|
|
|
|
|
def data_url_from_file(file_path: str) -> str:
|
|
if not os.path.exists(file_path):
|
|
raise FileNotFoundError(f"File not found: {file_path}")
|
|
|
|
with open(file_path, "rb") as file:
|
|
file_content = file.read()
|
|
|
|
base64_content = base64.b64encode(file_content).decode("utf-8")
|
|
mime_type, _ = mimetypes.guess_type(file_path)
|
|
|
|
data_url = f"data:{mime_type};base64,{base64_content}"
|
|
|
|
return data_url
|
|
|
|
|
|
async def register_dataset(
|
|
datasets_impl: Datasets, for_generation=False, dataset_id="test_dataset"
|
|
):
|
|
test_file = Path(os.path.abspath(__file__)).parent / "test_dataset.csv"
|
|
test_url = data_url_from_file(str(test_file))
|
|
|
|
if for_generation:
|
|
dataset_schema = {
|
|
"expected_answer": StringType(),
|
|
"input_query": StringType(),
|
|
"chat_completion_input": ChatCompletionInputType(),
|
|
}
|
|
else:
|
|
dataset_schema = {
|
|
"expected_answer": StringType(),
|
|
"input_query": StringType(),
|
|
"generated_answer": StringType(),
|
|
}
|
|
|
|
await datasets_impl.register_dataset(
|
|
dataset_id=dataset_id,
|
|
dataset_schema=dataset_schema,
|
|
url=URL(uri=test_url),
|
|
)
|
|
|
|
|
|
class TestDatasetIO:
|
|
@pytest.mark.asyncio
|
|
async def test_datasets_list(self, datasetio_stack):
|
|
# NOTE: this needs you to ensure that you are starting from a clean state
|
|
# but so far we don't have an unregister API unfortunately, so be careful
|
|
_, datasets_impl = datasetio_stack
|
|
response = await datasets_impl.list_datasets()
|
|
assert isinstance(response, list)
|
|
assert len(response) == 0
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_register_dataset(self, datasetio_stack):
|
|
_, datasets_impl = datasetio_stack
|
|
await register_dataset(datasets_impl)
|
|
response = await datasets_impl.list_datasets()
|
|
assert isinstance(response, list)
|
|
assert len(response) == 1
|
|
assert response[0].identifier == "test_dataset"
|
|
|
|
with pytest.raises(Exception) as exc_info:
|
|
# unregister a dataset that does not exist
|
|
await datasets_impl.unregister_dataset("test_dataset2")
|
|
|
|
await datasets_impl.unregister_dataset("test_dataset")
|
|
response = await datasets_impl.list_datasets()
|
|
assert isinstance(response, list)
|
|
assert len(response) == 0
|
|
|
|
with pytest.raises(Exception) as exc_info:
|
|
await datasets_impl.unregister_dataset("test_dataset")
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_get_rows_paginated(self, datasetio_stack):
|
|
datasetio_impl, datasets_impl = datasetio_stack
|
|
await register_dataset(datasets_impl)
|
|
response = await datasetio_impl.get_rows_paginated(
|
|
dataset_id="test_dataset",
|
|
rows_in_page=3,
|
|
)
|
|
assert isinstance(response.rows, list)
|
|
assert len(response.rows) == 3
|
|
assert response.next_page_token == "3"
|
|
|
|
provider = datasetio_impl.routing_table.get_provider_impl("test_dataset")
|
|
if provider.__provider_spec__.provider_type == "remote":
|
|
pytest.skip("remote provider doesn't support get_rows_paginated")
|
|
|
|
# iterate over all rows
|
|
response = await datasetio_impl.get_rows_paginated(
|
|
dataset_id="test_dataset",
|
|
rows_in_page=2,
|
|
page_token=response.next_page_token,
|
|
)
|
|
assert isinstance(response.rows, list)
|
|
assert len(response.rows) == 2
|
|
assert response.next_page_token == "5"
|