fix: do not stringify benchmark execution statistics

maxhammeralephalpha · maxhammeralephalpha · commit 32e874e9aeef · 2025-09-26T15:40:57.000+02:00
diff --git a/pharia_studio_sdk/evaluation/benchmark/studio_benchmark.py b/pharia_studio_sdk/evaluation/benchmark/studio_benchmark.py
@@ -171,7 +171,7 @@ def average_or_zero(list: list) -> float:
             eval_failed_count=evaluation_overview.failed_evaluation_count,
             aggregation_start=aggregation_overview.start,
             aggregation_end=aggregation_overview.end,
-            statistics=aggregation_overview.statistics.model_dump_json(),
+            statistics=aggregation_overview.statistics.model_dump(),
         )
 
         benchmark_execution_id = self.client.submit_benchmark_execution(

Original file line number	Diff line number	Diff line change
`@@ -171,7 +171,7 @@ def average_or_zero(list: list) -> float:`
`171`	`171`	`eval_failed_count=evaluation_overview.failed_evaluation_count,`
`172`	`172`	`aggregation_start=aggregation_overview.start,`
`173`	`173`	`aggregation_end=aggregation_overview.end,`
`174`		`- statistics=aggregation_overview.statistics.model_dump_json(),`
	`174`	`+ statistics=aggregation_overview.statistics.model_dump(),`
`175`	`175`	`)`
`176`	`176`
`177`	`177`	`benchmark_execution_id = self.client.submit_benchmark_execution(`