change name

Author: xzrderek

eval_protocol/adapters/base.py

@@ -16,6 +16,6 @@ def get_evaluation_rows(self, *args, **kwargs) -> List[EvaluationRow]:
         """Get evaluation rows from the data source."""
         pass
 
-    def push_scores(self, rows: List[EvaluationRow], model_name: str, mean_score: float) -> None:
-        """Push evaluation scores back to the data source for tracking and analysis."""
+    def upload_scores(self, rows: List[EvaluationRow], model_name: str, mean_score: float) -> None:
+        """Upload evaluation scores back to the data source for tracking and analysis."""
         pass

eval_protocol/adapters/braintrust.py

@@ -224,8 +224,8 @@ def get_evaluation_rows(
         logger.info("Successfully processed %d BTQL results into %d evaluation rows", len(all_traces), len(eval_rows))
         return eval_rows
 
-    def push_scores(self, rows: List[EvaluationRow], model_name: str, mean_score: float) -> None:
-        """Push evaluation scores back to Braintrust traces for tracking and analysis.
+    def upload_scores(self, rows: List[EvaluationRow], model_name: str, mean_score: float) -> None:
+        """Upload evaluation scores back to Braintrust traces for tracking and analysis.
 
         Creates score entries in Braintrust for each unique trace_id found in the evaluation
         rows' session data. This allows you to see evaluation results directly in the

eval_protocol/adapters/langfuse.py

@@ -434,8 +434,8 @@ def get_evaluation_rows_by_ids(
                 continue
         return eval_rows
 
-    def push_scores(self, rows: List[EvaluationRow], model_name: str, mean_score: float) -> None:
-        """Push evaluation scores back to Langfuse traces for tracking and analysis.
+    def upload_scores(self, rows: List[EvaluationRow], model_name: str, mean_score: float) -> None:
+        """Upload evaluation scores back to Langfuse traces for tracking and analysis.
 
         Creates a score entry in Langfuse for each unique trace_id found in the evaluation
         rows' session data. This allows you to see evaluation results directly in the

eval_protocol/quickstart/llm_judge.py

@@ -95,6 +95,6 @@ async def run_judgment(row):
 
     # Push scores back to adapter if provided. Note that one score per model will be pushed back onto same trace.
     if adapter:
-        adapter.push_scores(rows, model_name, mean_score)
+        adapter.upload_scores(rows, model_name, mean_score)
 
     return rows