fix results and update evaluation

agenta-backend/agenta_backend/models/api/evaluation_model.py

@@ -89,7 +89,7 @@ class SimpleEvaluationOutput(BaseModel):
     evaluation_type: EvaluationType
 
 
-class EvaluationUpdate(BaseModel):
+class HumanEvaluationUpdate(BaseModel):
     status: Optional[EvaluationStatusEnum]
     evaluation_type_settings: Optional[EvaluationTypeSettings]
 

agenta-backend/agenta_backend/routers/evaluation_router.py

@@ -14,18 +14,12 @@
     EvaluationScenario,
     CustomEvaluationOutput,
     CustomEvaluationDetail,
-    EvaluationScenarioScoreUpdate,
-    EvaluationScenarioUpdate,
     ExecuteCustomEvaluationCode,
-    HumanEvaluation,
-    HumanEvaluationScenarioUpdate,
     NewEvaluation,
     DeleteEvaluation,
     EvaluationType,
     CreateCustomEvaluation,
-    EvaluationUpdate,
     EvaluationWebhook,
-    NewHumanEvaluation,
     SimpleEvaluationOutput,
 )
 from agenta_backend.services.evaluation_service import (
@@ -35,11 +29,9 @@
     fetch_custom_evaluation_detail,
     get_evaluation_scenario_score,
     update_evaluation_scenario_score,
-    update_evaluation,
     create_custom_code_evaluation,
     update_custom_code_evaluation,
     execute_custom_code_evaluation,
-    update_human_evaluation_scenario,
 )
 from agenta_backend.services import evaluation_service
 from agenta_backend.utils.common import check_access_to_app
@@ -163,33 +155,6 @@ async def fetch_evaluation_results(evaluation_id: str, request: Request):
         raise HTTPException(status_code=500, detail=str(exc))
 
 
-@router.put("/{evaluation_id}/")
-async def update_evaluation_router(
-    request: Request,
-    evaluation_id: str,
-    update_data: EvaluationUpdate = Body(...),
-):
-    """Updates an evaluation's status.
-
-    Raises:
-        HTTPException: If the columns in the test set do not match with the inputs in the variant.
-
-    Returns:
-        None: A 204 No Content status code, indicating that the update was successful.
-    """
-    try:
-        # Get user and organization id
-        user_org_data: dict = await get_user_and_org_id(request.state.user_id)
-        await update_evaluation(evaluation_id, update_data, **user_org_data)
-        return Response(status_code=status.HTTP_204_NO_CONTENT)
-
-    except KeyError:
-        raise HTTPException(
-            status_code=400,
-            detail="columns in the test set should match the names of the inputs in the variant",
-        )
-
-
 @router.get(
     "/{evaluation_id}/evaluation_scenarios/",
     response_model=List[EvaluationScenario],

agenta-backend/agenta_backend/routers/human_evaluation_router.py

@@ -13,6 +13,7 @@
     HumanEvaluationScenario,
     HumanEvaluationScenarioUpdate,
     EvaluationType,
+    HumanEvaluationUpdate,
     NewHumanEvaluation,
     SimpleEvaluationOutput,
 )
@@ -28,6 +29,7 @@
     get_evaluation_scenario_score,
     update_evaluation_scenario_score,
     update_human_evaluation_scenario,
+    update_human_evaluation_service,
 )
 
 
@@ -153,6 +155,35 @@ async def fetch_evaluation_scenarios(
     return eval_scenarios
 
 
+@router.put("/{evaluation_id}/", operation_id="update_evaluation")
+async def update_evaluation(
+    request: Request,
+    evaluation_id: str,
+    update_data: HumanEvaluationUpdate = Body(...),
+):
+    """Updates an evaluation's status.
+
+    Raises:
+        HTTPException: If the columns in the test set do not match with the inputs in the variant.
+
+    Returns:
+        None: A 204 No Content status code, indicating that the update was successful.
+    """
+    try:
+        # Get user and organization id
+        user_org_data: dict = await get_user_and_org_id(request.state.user_id)
+        await update_human_evaluation_service(
+            evaluation_id, update_data, **user_org_data
+        )
+        return Response(status_code=status.HTTP_204_NO_CONTENT)
+
+    except KeyError:
+        raise HTTPException(
+            status_code=400,
+            detail="columns in the test set should match the names of the inputs in the variant",
+        )
+
+
 @router.put(
     "/{evaluation_id}/evaluation_scenario/{evaluation_scenario_id}/{evaluation_type}/"
 )
@@ -244,10 +275,8 @@ async def fetch_results(
     # Get user and organization id
     print("are we here")
     user_org_data: dict = await get_user_and_org_id(request.state.user_id)
-    evaluation = (
-        await evaluation_service._fetch_human_evaluation_scenario_and_check_access(
-            evaluation_id, **user_org_data
-        )
+    evaluation = await evaluation_service._fetch_human_evaluation_and_check_access(
+        evaluation_id, **user_org_data
     )
     print("really???")
     if evaluation.evaluation_type == EvaluationType.human_a_b_testing:

agenta-backend/agenta_backend/services/evaluation_service.py

@@ -14,12 +14,13 @@
     CustomEvaluationDetail,
     EvaluationScenarioInput,
     EvaluationType,
+    EvaluationTypeSettings,
     HumanEvaluation,
     HumanEvaluationScenario,
+    HumanEvaluationUpdate,
     NewEvaluation,
     EvaluationScenarioUpdate,
     CreateCustomEvaluation,
-    EvaluationUpdate,
     EvaluationStatusEnum,
     NewHumanEvaluation,
 )
@@ -37,8 +38,6 @@
     HumanEvaluationScenarioOutput,
     UserDB,
     AppDB,
-    EvaluationScenarioInputDB,
-    EvaluationScenarioOutputDB,
     CustomEvaluationDB,
 )
 
@@ -115,6 +114,9 @@ async def _fetch_human_evaluation_scenario_and_check_access(
     evaluation_scenario = await db_manager.fetch_human_evaluation_scenario_by_id(
         evaluation_scenario_id=evaluation_scenario_id
     )
+
+    print("evaluation_scenario")
+    print(evaluation_scenario)
     if evaluation_scenario is None:
         raise HTTPException(
             status_code=404,
@@ -251,8 +253,8 @@ async def create_evaluation_scenario(
     await engine.save(new_eval_scenario)
 
 
-async def update_evaluation(
-    evaluation_id: str, update_payload: EvaluationUpdate, **user_org_data: dict
+async def update_human_evaluation_service(
+    evaluation_id: str, update_payload: HumanEvaluationUpdate, **user_org_data: dict
 ) -> None:
     """
     Update an existing evaluation based on the provided payload.
@@ -265,7 +267,7 @@ async def update_evaluation(
         HTTPException: If the evaluation is not found or access is denied.
     """
     # Fetch the evaluation by ID
-    evaluation = await _fetch_evaluation_and_check_access(
+    evaluation = await _fetch_human_evaluation_and_check_access(
         evaluation_id=evaluation_id,
         **user_org_data,
     )