fix: EvaluationRunResult.score_report() is missing the metrics co…

…lumn (#7817)

* fixing the DataFrame with the aggregated scores

* fixing tests

haystack/evaluation/eval_run_result.py

@@ -97,7 +97,9 @@ def __init__(self, run_name: str, inputs: Dict[str, List[Any]], results: Dict[st
 
     def score_report(self) -> DataFrame:  # noqa: D102
         results = {k: v["score"] for k, v in self.results.items()}
-        return DataFrame.from_dict(results, orient="index", columns=["score"])
+        df = DataFrame.from_dict(results, orient="index", columns=["score"]).reset_index()
+        df.columns = ["metrics", "score"]
+        return df
 
     def to_pandas(self) -> DataFrame:  # noqa: D102
         inputs_columns = list(self.inputs.keys())

test/evaluation/test_eval_run_result.py

@@ -88,10 +88,11 @@ def test_score_report():
 
     result = EvaluationRunResult("testing_pipeline_1", inputs=data["inputs"], results=data["metrics"])
     report = result.score_report().to_json()
+
     assert report == (
-        '{"score":{"reciprocal_rank":0.476932,"single_hit":0.75,"multi_hit":0.46428375,'
-        '"context_relevance":0.58177975,"faithfulness":0.40585375,'
-        '"semantic_answer_similarity":0.53757075}}'
+        '{"metrics":{"0":"reciprocal_rank","1":"single_hit","2":"multi_hit","3":"context_relevance",'
+        '"4":"faithfulness","5":"semantic_answer_similarity"},'
+        '"score":{"0":0.476932,"1":0.75,"2":0.46428375,"3":0.58177975,"4":0.40585375,"5":0.53757075}}'
     )