feat: create HTTP DELETE API endpoints to unregister ScoringFn and Benchmark resources in Llama Stack (#3371)

# What does this PR do?
<!-- Provide a short summary of what this PR does and why. Link to
relevant issues if applicable. -->
This PR provides functionality for users to unregister ScoringFn and
Benchmark resources for `scoring` and `eval` APIs.

<!-- If resolving an issue, uncomment and update the line below -->
<!-- Closes #[issue-number] -->
Closes #3051 

## Test Plan
<!-- Describe the tests you ran to verify your changes with result
summaries. *Provide clear instructions so the plan can be easily
re-executed.* -->
Updated integration and unit tests via CI workflow
This commit is contained in:
IAN MILLER 2025-09-15 20:43:38 +01:00 committed by GitHub
parent 01bdcce4d2
commit ab321739f2
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
13 changed files with 241 additions and 3 deletions

View file

@ -52,14 +52,19 @@ class TestNVIDIAEvalImpl(unittest.TestCase):
self.evaluator_post_patcher = patch(
"llama_stack.providers.remote.eval.nvidia.eval.NVIDIAEvalImpl._evaluator_post"
)
self.evaluator_delete_patcher = patch(
"llama_stack.providers.remote.eval.nvidia.eval.NVIDIAEvalImpl._evaluator_delete"
)
self.mock_evaluator_get = self.evaluator_get_patcher.start()
self.mock_evaluator_post = self.evaluator_post_patcher.start()
self.mock_evaluator_delete = self.evaluator_delete_patcher.start()
def tearDown(self):
"""Clean up after each test."""
self.evaluator_get_patcher.stop()
self.evaluator_post_patcher.stop()
self.evaluator_delete_patcher.stop()
def _assert_request_body(self, expected_json):
"""Helper method to verify request body in Evaluator POST request is correct"""
@ -115,6 +120,13 @@ class TestNVIDIAEvalImpl(unittest.TestCase):
self.mock_evaluator_post.assert_called_once()
self._assert_request_body({"namespace": benchmark.provider_id, "name": benchmark.identifier, **eval_config})
def test_unregister_benchmark(self):
# Unregister the benchmark
self.run_async(self.eval_impl.unregister_benchmark(benchmark_id=MOCK_BENCHMARK_ID))
# Verify the Evaluator API was called correctly
self.mock_evaluator_delete.assert_called_once_with(f"/v1/evaluation/configs/nvidia/{MOCK_BENCHMARK_ID}")
def test_run_eval(self):
benchmark_config = BenchmarkConfig(
eval_candidate=ModelCandidate(