Context coverage score

cognee/eval_framework/evaluation/deep_eval_adapter.py

@@ -4,7 +4,7 @@
 from cognee.eval_framework.evaluation.base_eval_adapter import BaseEvalAdapter
 from cognee.eval_framework.evaluation.metrics.exact_match import ExactMatchMetric
 from cognee.eval_framework.evaluation.metrics.f1 import F1ScoreMetric
-from cognee.eval_framework.evaluation.metrics.context_match import ContextMatchMetric
+from cognee.eval_framework.evaluation.metrics.context_coverage import ContextCoverageMetric
 from typing import Any, Dict, List
 from deepeval.metrics import ContextualRelevancyMetric
 
@@ -16,7 +16,7 @@ def __init__(self):
             "EM": ExactMatchMetric(),
             "f1": F1ScoreMetric(),
             "contextual_relevancy": ContextualRelevancyMetric(),
-            "context_match": ContextMatchMetric(),
+            "context_coverage": ContextCoverageMetric(),
         }
 
     async def evaluate_answers(

cognee/eval_framework/evaluation/evaluation_executor.py

@@ -23,6 +23,6 @@ def __init__(
     async def execute(self, answers: List[Dict[str, str]], evaluator_metrics: Any) -> Any:
         if self.evaluate_contexts:
             evaluator_metrics.append("contextual_relevancy")
-            evaluator_metrics.append("context_match")
+            evaluator_metrics.append("context_coverage")
         metrics = await self.eval_adapter.evaluate_answers(answers, evaluator_metrics)
         return metrics

cognee/eval_framework/evaluation/metrics/context_match.py → cognee/eval_framework/evaluation/metrics/context_coverage.py

@@ -3,10 +3,12 @@
 from deepeval.metrics.summarization.schema import ScoreType
 from deepeval.metrics.indicator import metric_progress_indicator
 from deepeval.utils import get_or_create_event_loop
-import asyncio
+from deepeval.metrics.summarization.template import SummarizationTemplate
+from deepeval.metrics.summarization.schema import Reason
+from deepeval.metrics.utils import trimAndLoadJson
 
 
-class ContextMatchMetric(SummarizationMetric):
+class ContextCoverageMetric(SummarizationMetric):
     def measure(
         self,
         test_case,
@@ -16,7 +18,7 @@ def measure(
             input=test_case.context[0],
             actual_output=test_case.retrieval_context[0],
         )
-
+        self.assessment_questions = None
         self.evaluation_cost = 0 if self.using_native_model else None
         with metric_progress_indicator(self, _show_indicator=_show_indicator):
             if self.async_mode:
@@ -26,7 +28,9 @@ def measure(
                 )
             else:
                 self.coverage_verdicts = self._generate_coverage_verdicts(mapped_test_case)
+                self.alignment_verdicts = []
                 self.score = self._calculate_score(ScoreType.COVERAGE)
+                self.reason = self._generate_reason()
                 self.success = self.score >= self.threshold
                 return self.score
 
@@ -42,7 +46,9 @@ async def a_measure(
             _show_indicator=_show_indicator,
         ):
             self.coverage_verdicts = await self._a_generate_coverage_verdicts(test_case)
+            self.alignment_verdicts = []
             self.score = self._calculate_score(ScoreType.COVERAGE)
+            self.reason = self._generate_reason()
             self.success = self.score >= self.threshold
             return self.score
 

cognee/eval_framework/metrics_dashboard.py

@@ -5,7 +5,7 @@
 
 metrics_fields = {
     "contextual_relevancy": ["question", "retrieval_context"],
-    "context_match": ["question", "retrieval_context", "golden_context"],
+    "context_coverage": ["question", "retrieval_context", "golden_context"],
 }
 default_metrics_fields = ["question", "answer", "golden_answer"]
 

cognee/tests/unit/eval_framework/deepeval_adapter_test.py

@@ -9,7 +9,7 @@
         "deepeval": MagicMock(),
         "deepeval.metrics": MagicMock(),
         "deepeval.test_case": MagicMock(),
-        "cognee.eval_framework.evaluation.metrics.context_match": MagicMock(),
+        "cognee.eval_framework.evaluation.metrics.context_coverage": MagicMock(),
     },
 ):
     from cognee.eval_framework.evaluation.deep_eval_adapter import DeepEvalAdapter