5a558eb09e
Backend Tests / discover-tests (push) Waiting to run
Backend Tests / ${{ matrix.name }} (push) Blocked by required conditions
Build and Publish SDK / build-and-publish (push) Waiting to run
Build Opik Docker Images / set-version (push) Waiting to run
Build Opik Docker Images / build-backend (push) Blocked by required conditions
Build Opik Docker Images / build-sandbox-executor-python (push) Blocked by required conditions
Build Opik Docker Images / build-python-backend (push) Blocked by required conditions
Build Opik Docker Images / build-frontend (push) Blocked by required conditions
Build Opik Docker Images / create-git-tag (push) Blocked by required conditions
ClickHouse Migration Cluster Check / validate-clickhouse-migrations (push) Waiting to run
Docs - Publish / run (push) Waiting to run
E2E Tests - Post Merge (v2) / 🧪 E2E v2 Tests (${{ github.event.inputs.tier || 't1' }}) (push) Waiting to run
E2E Tests - Post Merge (v2) / 📢 Slack Notification (push) Blocked by required conditions
Frontend Unit Tests / Test on Node 20 (push) Waiting to run
Guardrails E2E Tests / Select Python version matrix (push) Waiting to run
Guardrails E2E Tests / Guardrails E2E Tests ${{matrix.python_version}} (push) Blocked by required conditions
Guardrails E2E Tests / 📢 Slack Notification (push) Blocked by required conditions
Guardrails Backend Unit Tests / Guardrails Backend Unit Tests (push) Waiting to run
Guardrails Backend Unit Tests / 📢 Slack Notification (push) Blocked by required conditions
Lint Opik Helm Chart / lint-helm-chart (Helm v3.21.0) (push) Waiting to run
Lint Opik Helm Chart / lint-helm-chart (Helm v4.2.0) (push) Waiting to run
Lint Opik Helm Chart / unittest-helm-chart (push) Waiting to run
Lint Opik Helm Chart / render-equality (push) Waiting to run
Opik Optimizer - Unit Tests / Opik Optimizer Unit Tests Python ${{matrix.python_version}} (push) Waiting to run
Python BE E2E Tests / Python BE E2E (push) Waiting to run
Python Backend Tests / run-python-backend-tests (push) Waiting to run
Python SDK Unit Tests / Python SDK Unit Tests ${{matrix.python_version}} (push) Waiting to run
Release Drafter / update_release_draft (push) Waiting to run
SDK E2E Libraries Integration Tests / Check Secrets (push) Waiting to run
SDK E2E Libraries Integration Tests / Missed OpenAI API Key Warning (push) Blocked by required conditions
SDK E2E Libraries Integration Tests / build-opik (push) Blocked by required conditions
SDK E2E Libraries Integration Tests / E2E Lib Integration Python ${{matrix.python_version}} (push) Blocked by required conditions
TypeScript SDK Integration Build & Publish / build-and-publish (opik-gemini) (push) Waiting to run
TypeScript SDK Integration Build & Publish / build-and-publish (opik-langchain) (push) Waiting to run
TypeScript SDK Integration Build & Publish / build-and-publish (opik-openai) (push) Waiting to run
TypeScript SDK Integration Build & Publish / build-and-publish (opik-otel) (push) Waiting to run
TypeScript SDK Integration Build & Publish / build-and-publish (opik-vercel) (push) Waiting to run
TypeScript SDK Build & Publish / build-and-publish (push) Waiting to run
TypeScript SDK Unit Tests / Test on Node ${{ matrix.node-version }} (push) Waiting to run
TypeScript SDK Compatibility V1.x E2E Tests / Select Node version matrix (push) Has been cancelled
TypeScript SDK Compatibility V1.x E2E Tests / TypeScript SDK Compatibility V1.x E2E Tests Node ${{matrix.node_version}} (push) Has been cancelled
TypeScript SDK E2E Tests / TypeScript SDK E2E Tests Node ${{matrix.node_version}} (push) Has been cancelled
Opik Optimizer - E2E Tests / build-opik (push) Has been cancelled
TypeScript SDK Compatibility V1.x E2E Tests / build-opik (push) Has been cancelled
Python SDK E2E Tests / Select Python version matrix (push) Has been cancelled
Python SDK E2E Tests / Python SDK E2E Tests ${{matrix.python_version}} (push) Has been cancelled
Python SDK E2E Tests / build-opik (push) Has been cancelled
Python SDK Compatibility V1.x E2E Tests / Select Python version matrix (push) Has been cancelled
Python SDK Compatibility V1.x E2E Tests / Python SDK Compatibility V1.x E2E Tests ${{matrix.python_version}} (push) Has been cancelled
Python SDK Compatibility V1.x E2E Tests / build-opik (push) Has been cancelled
TypeScript SDK E2E Tests / Select Node version matrix (push) Has been cancelled
TypeScript SDK E2E Tests / build-opik (push) Has been cancelled
Opik Optimizer - E2E Tests / Opik Optimizer E2E Tests Python ${{matrix.python_version}} (push) Has been cancelled
Opik Optimizer - E2E Tests / Opik Optimizer Integration Smoke Tests (push) Has been cancelled
🐙 Code Quality / detect (push) Has been cancelled
🐙 Code Quality / lint (${{ matrix.leg.name }}) (push) Has been cancelled
🐙 Code Quality / summary (push) Has been cancelled
TypeScript SDK Library Integration Tests / Check Secrets (push) Has been cancelled
TypeScript SDK Library Integration Tests / opik-vercel (Vercel AI SDK / eve) (push) Has been cancelled
SDK Library Integration Tests Runner / Check Secrets (push) Has been cancelled
SDK Library Integration Tests Runner / Missed OpenAI API Key Warning (push) Has been cancelled
SDK Library Integration Tests Runner / Build (push) Has been cancelled
SDK Library Integration Tests Runner / openai_tests (push) Has been cancelled
SDK Library Integration Tests Runner / langchain_tests (push) Has been cancelled
SDK Library Integration Tests Runner / langchain_legacy_tests (push) Has been cancelled
SDK Library Integration Tests Runner / llama_index_tests (push) Has been cancelled
SDK Library Integration Tests Runner / anthropic_tests (push) Has been cancelled
SDK Library Integration Tests Runner / mistral_tests (push) Has been cancelled
SDK Library Integration Tests Runner / groq_tests (push) Has been cancelled
SDK Library Integration Tests Runner / aisuite_tests (push) Has been cancelled
SDK Library Integration Tests Runner / haystack_tests (push) Has been cancelled
SDK Library Integration Tests Runner / dspy_tests (push) Has been cancelled
SDK Library Integration Tests Runner / crewai_v0_tests (push) Has been cancelled
SDK Library Integration Tests Runner / crewai_v1_tests (push) Has been cancelled
SDK Library Integration Tests Runner / genai_tests (push) Has been cancelled
SDK Library Integration Tests Runner / adk_tests (push) Has been cancelled
SDK Library Integration Tests Runner / adk_legacy_1_3_0_tests (push) Has been cancelled
SDK Library Integration Tests Runner / evaluation_metrics_tests (push) Has been cancelled
SDK Library Integration Tests Runner / bedrock_tests (push) Has been cancelled
SDK Library Integration Tests Runner / litellm_tests (push) Has been cancelled
SDK Library Integration Tests Runner / harbor_tests (push) Has been cancelled
SDK Library Integration Tests Runner / Slack Notification (push) Has been cancelled
148 lines
4.3 KiB
Python
148 lines
4.3 KiB
Python
from unittest.mock import MagicMock
|
|
|
|
from opik.evaluation.engine.metrics_evaluator import build_metrics_evaluator
|
|
from opik.evaluation.metrics import base_metric
|
|
from opik.evaluation.suite_evaluators import llm_judge
|
|
|
|
|
|
def _make_judge(assertions, **kwargs):
|
|
defaults = {"track": False, "name": "llm_judge"}
|
|
defaults.update(kwargs)
|
|
return llm_judge.LLMJudge(assertions=assertions, **defaults)
|
|
|
|
|
|
def _make_non_judge_metric(name="other_metric"):
|
|
metric = MagicMock(spec=base_metric.BaseMetric)
|
|
metric.name = name
|
|
return metric
|
|
|
|
|
|
def _build(regular_metrics):
|
|
"""Build a MetricsEvaluator with no item and extract its regular metrics."""
|
|
evaluator = build_metrics_evaluator(
|
|
item=None,
|
|
regular_metrics=regular_metrics,
|
|
scoring_key_mapping={},
|
|
evaluator_model=None,
|
|
)
|
|
return evaluator.regular_metrics
|
|
|
|
|
|
class TestLLMJudgeMerged:
|
|
def test_two_judges__combines_assertions(self):
|
|
j1 = _make_judge(["A", "B"])
|
|
j2 = _make_judge(["C"])
|
|
|
|
merged = llm_judge.LLMJudge.merged([j1, j2])
|
|
|
|
assert merged is not None
|
|
assert merged.assertions == ["A", "B", "C"]
|
|
|
|
def test_duplicate_assertions__deduplicated(self):
|
|
j1 = _make_judge(["A", "B"])
|
|
j2 = _make_judge(["B", "C"])
|
|
|
|
merged = llm_judge.LLMJudge.merged([j1, j2])
|
|
|
|
assert merged is not None
|
|
assert merged.assertions == ["A", "B", "C"]
|
|
|
|
def test_mismatched_settings__returns_none(self):
|
|
j1 = _make_judge(["A"], temperature=0.5, seed=42)
|
|
j2 = _make_judge(["B"], temperature=0.9, seed=99)
|
|
|
|
assert llm_judge.LLMJudge.merged([j1, j2]) is None
|
|
|
|
def test_mismatched_model__returns_none(self):
|
|
j1 = _make_judge(["A"], model="gpt-4o")
|
|
j2 = _make_judge(["B"], model="gpt-4o-mini")
|
|
|
|
assert llm_judge.LLMJudge.merged([j1, j2]) is None
|
|
|
|
def test_empty_list__returns_none(self):
|
|
assert llm_judge.LLMJudge.merged([]) is None
|
|
|
|
def test_matching_settings__merges(self):
|
|
j1 = _make_judge(["A"], temperature=0.5, seed=42, name="suite_judge")
|
|
j2 = _make_judge(["B"], temperature=0.5, seed=42, name="item_judge")
|
|
|
|
merged = llm_judge.LLMJudge.merged([j1, j2])
|
|
|
|
assert merged is not None
|
|
assert merged.name == "suite_judge"
|
|
assert merged._temperature == 0.5
|
|
assert merged._seed == 42
|
|
assert merged.assertions == ["A", "B"]
|
|
|
|
def test_three_judges__all_merged(self):
|
|
j1 = _make_judge(["A"])
|
|
j2 = _make_judge(["B"])
|
|
j3 = _make_judge(["C"])
|
|
|
|
merged = llm_judge.LLMJudge.merged([j1, j2, j3])
|
|
|
|
assert merged is not None
|
|
assert merged.assertions == ["A", "B", "C"]
|
|
|
|
def test_single_judge__returns_none(self):
|
|
j1 = _make_judge(["A", "B"])
|
|
|
|
assert llm_judge.LLMJudge.merged([j1]) is None
|
|
|
|
|
|
class TestBuildMetricsEvaluatorMerging:
|
|
def test_no_judges__returns_unchanged(self):
|
|
m1 = _make_non_judge_metric("m1")
|
|
m2 = _make_non_judge_metric("m2")
|
|
|
|
result = _build([m1, m2])
|
|
|
|
assert result == [m1, m2]
|
|
|
|
def test_single_judge__not_merged(self):
|
|
judge = _make_judge(["assertion A"])
|
|
|
|
result = _build([judge])
|
|
|
|
assert len(result) == 1
|
|
assert result[0] is judge
|
|
|
|
def test_two_judges__merges_into_one(self):
|
|
j1 = _make_judge(["A", "B"])
|
|
j2 = _make_judge(["C"])
|
|
|
|
result = _build([j1, j2])
|
|
|
|
assert len(result) == 1
|
|
assert isinstance(result[0], llm_judge.LLMJudge)
|
|
assert result[0].assertions == ["A", "B", "C"]
|
|
|
|
def test_mixed_metrics__merged_judge_first(self):
|
|
m1 = _make_non_judge_metric("m1")
|
|
j1 = _make_judge(["A"])
|
|
m2 = _make_non_judge_metric("m2")
|
|
j2 = _make_judge(["B"])
|
|
|
|
result = _build([m1, j1, m2, j2])
|
|
|
|
assert len(result) == 3
|
|
assert isinstance(result[0], llm_judge.LLMJudge)
|
|
assert result[0].assertions == ["A", "B"]
|
|
assert result[1] is m1
|
|
assert result[2] is m2
|
|
|
|
def test_mismatched_settings__skips_merge(self):
|
|
j1 = _make_judge(["A"], temperature=0.5)
|
|
j2 = _make_judge(["B"], temperature=0.9)
|
|
|
|
result = _build([j1, j2])
|
|
|
|
assert len(result) == 2
|
|
assert result[0] is j1
|
|
assert result[1] is j2
|
|
|
|
def test_empty_list__returns_empty(self):
|
|
result = _build([])
|
|
|
|
assert result == []
|