deepeval 3.6.5__py3-none-any.whl → 3.6.7__py3-none-any.whl

deepeval/metrics/g_eval/g_eval.py

@@ -1,5 +1,7 @@
 """LLM evaluated metric based on the GEval framework: https://arxiv.org/pdf/2303.16634.pdf"""
 
+import asyncio
+
 from typing import Optional, List, Tuple, Union, Type
 from deepeval.metrics import BaseMetric
 from deepeval.test_case import (
@@ -16,7 +18,7 @@ from deepeval.metrics.utils import (
 )
 from deepeval.models import DeepEvalBaseLLM
 from deepeval.metrics.indicator import metric_progress_indicator
-from deepeval.metrics.g_eval.schema import *
+from deepeval.metrics.g_eval import schema as gschema
 from deepeval.metrics.g_eval.utils import (
     Rubric,
     construct_g_eval_params_string,
@@ -29,6 +31,8 @@ from deepeval.metrics.g_eval.utils import (
     number_evaluation_steps,
     get_score_range,
 )
+from deepeval.metrics.api import metric_data_manager
+from deepeval.config.settings import get_settings
 
 
 class GEval(BaseMetric):
@@ -71,6 +75,7 @@ class GEval(BaseMetric):
         test_case: LLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
         _additional_context: Optional[str] = None,
     ) -> float:
         check_llm_test_case_params(test_case, self.evaluation_params, self)
@@ -81,12 +86,16 @@ class GEval(BaseMetric):
         ):
             if self.async_mode:
                 loop = get_or_create_event_loop()
+                coro = self.a_measure(
+                    test_case,
+                    _show_indicator=False,
+                    _in_component=_in_component,
+                    _additional_context=_additional_context,
+                )
                 loop.run_until_complete(
-                    self.a_measure(
-                        test_case,
-                        _show_indicator=False,
-                        _in_component=_in_component,
-                        _additional_context=_additional_context,
+                    asyncio.wait_for(
+                        coro,
+                        timeout=get_settings().DEEPEVAL_PER_TASK_TIMEOUT_SECONDS,
                     )
                 )
             else:
@@ -115,6 +124,10 @@ class GEval(BaseMetric):
                         f"Reason: {self.reason}",
                     ],
                 )
+                if _log_metric_to_confident:
+                    metric_data_manager.post_metric_if_enabled(
+                        self, test_case=test_case
+                    )
 
             return self.score
 
@@ -123,6 +136,7 @@ class GEval(BaseMetric):
         test_case: LLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
         _additional_context: Optional[str] = None,
     ) -> float:
         check_llm_test_case_params(test_case, self.evaluation_params, self)
@@ -158,6 +172,10 @@ class GEval(BaseMetric):
                     f"Reason: {self.reason}",
                 ],
             )
+            if _log_metric_to_confident:
+                metric_data_manager.post_metric_if_enabled(
+                    self, test_case=test_case
+                )
             return self.score
 
     async def _a_generate_evaluation_steps(self) -> List[str]:
@@ -177,7 +195,9 @@ class GEval(BaseMetric):
             return data["steps"]
         else:
             try:
-                res: Steps = await self.model.a_generate(prompt, schema=Steps)
+                res: gschema.Steps = await self.model.a_generate(
+                    prompt, schema=gschema.Steps
+                )
                 return res.steps
             except TypeError:
                 res = await self.model.a_generate(prompt)
@@ -201,7 +221,9 @@ class GEval(BaseMetric):
             return data["steps"]
         else:
             try:
-                res: Steps = self.model.generate(prompt, schema=Steps)
+                res: gschema.Steps = self.model.generate(
+                    prompt, schema=gschema.Steps
+                )
                 return res.steps
             except TypeError:
                 res = self.model.generate(prompt)
@@ -264,7 +286,7 @@ class GEval(BaseMetric):
                     score, res
                 )
                 return weighted_summed_score, reason
-            except:
+            except (KeyError, AttributeError, TypeError, ValueError):
                 return score, reason
         except (
             AttributeError
@@ -276,8 +298,8 @@ class GEval(BaseMetric):
                 return data["score"], data["reason"]
             else:
                 try:
-                    res: ReasonScore = await self.model.a_generate(
-                        prompt, schema=ReasonScore
+                    res: gschema.ReasonScore = await self.model.a_generate(
+                        prompt, schema=gschema.ReasonScore
                     )
                     return res.score, res.reason
                 except TypeError:
@@ -338,7 +360,7 @@ class GEval(BaseMetric):
                     score, res
                 )
                 return weighted_summed_score, reason
-            except:
+            except (KeyError, AttributeError, TypeError, ValueError):
                 return score, reason
         except AttributeError:
             # This catches the case where a_generate_raw_response doesn't exist.
@@ -349,8 +371,8 @@ class GEval(BaseMetric):
                 return data["score"], data["reason"]
             else:
                 try:
-                    res: ReasonScore = self.model.generate(
-                        prompt, schema=ReasonScore
+                    res: gschema.ReasonScore = self.model.generate(
+                        prompt, schema=gschema.ReasonScore
                     )
                     return res.score, res.reason
                 except TypeError:
@@ -364,7 +386,7 @@ class GEval(BaseMetric):
         else:
             try:
                 self.success = self.score >= self.threshold
-            except:
+            except TypeError:
                 self.success = False
         return self.success
 

deepeval/metrics/hallucination/hallucination.py

@@ -16,6 +16,7 @@ from deepeval.metrics.hallucination.template import HallucinationTemplate
 from deepeval.models import DeepEvalBaseLLM
 from deepeval.metrics.indicator import metric_progress_indicator
 from deepeval.metrics.hallucination.schema import *
+from deepeval.metrics.api import metric_data_manager
 
 required_params: List[LLMTestCaseParams] = [
     LLMTestCaseParams.INPUT,
@@ -51,6 +52,7 @@ class HallucinationMetric(BaseMetric):
         test_case: LLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
 
         check_llm_test_case_params(test_case, required_params, self)
@@ -66,6 +68,7 @@ class HallucinationMetric(BaseMetric):
                         test_case,
                         _show_indicator=False,
                         _in_component=_in_component,
+                        _log_metric_to_confident=_log_metric_to_confident,
                     )
                 )
             else:
@@ -84,6 +87,10 @@ class HallucinationMetric(BaseMetric):
                         f"Score: {self.score}\nReason: {self.reason}",
                     ],
                 )
+                if _log_metric_to_confident:
+                    metric_data_manager.post_metric_if_enabled(
+                        self, test_case=test_case
+                    )
 
             return self.score
 
@@ -92,6 +99,7 @@ class HallucinationMetric(BaseMetric):
         test_case: LLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
 
         check_llm_test_case_params(test_case, required_params, self)
@@ -118,7 +126,10 @@ class HallucinationMetric(BaseMetric):
                     f"Score: {self.score}\nReason: {self.reason}",
                 ],
             )
-
+            if _log_metric_to_confident:
+                metric_data_manager.post_metric_if_enabled(
+                    self, test_case=test_case
+                )
             return self.score
 
     async def _a_generate_reason(self):

deepeval/metrics/indicator.py

@@ -100,6 +100,7 @@ async def measure_metric_task(
                     test_case,
                     _show_indicator=False,
                     _in_component=_in_component,
+                    _log_metric_to_confident=False,
                 )
                 finish_text = "Done"
             except MissingTestCaseParamsError as e:
@@ -116,7 +117,9 @@ async def measure_metric_task(
             except TypeError:
                 try:
                     await metric.a_measure(
-                        test_case, _in_component=_in_component
+                        test_case,
+                        _in_component=_in_component,
+                        _log_metric_to_confident=False,
                     )
                     finish_text = "Done"
                 except MissingTestCaseParamsError as e:
@@ -241,7 +244,10 @@ async def safe_a_measure(
 ):
     try:
         await metric.a_measure(
-            tc, _show_indicator=False, _in_component=_in_component
+            tc,
+            _show_indicator=False,
+            _in_component=_in_component,
+            _log_metric_to_confident=False,
         )
         update_pbar(progress, pbar_eval_id)
     except MissingTestCaseParamsError as e:

deepeval/metrics/json_correctness/json_correctness.py

@@ -18,6 +18,7 @@ from deepeval.metrics.indicator import metric_progress_indicator
 from deepeval.metrics.json_correctness.template import JsonCorrectnessTemplate
 from deepeval.metrics.json_correctness.schema import JsonCorrectnessScoreReason
 from deepeval.utils import get_or_create_event_loop
+from deepeval.metrics.api import metric_data_manager
 
 DEFAULT_CORRECT_REASON = "The generated Json matches and is syntactically correct to the expected schema."
 
@@ -51,6 +52,7 @@ class JsonCorrectnessMetric(BaseMetric):
         test_case: LLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
 
         check_llm_test_case_params(test_case, self._required_params, self)
@@ -66,6 +68,7 @@ class JsonCorrectnessMetric(BaseMetric):
                         test_case,
                         _show_indicator=False,
                         _in_component=_in_component,
+                        _log_metric_to_confident=_log_metric_to_confident,
                     )
                 )
             else:
@@ -88,6 +91,10 @@ class JsonCorrectnessMetric(BaseMetric):
                         f"Score: {self.score}\nReason: {self.reason}",
                     ],
                 )
+                if _log_metric_to_confident:
+                    metric_data_manager.post_metric_if_enabled(
+                        self, test_case=test_case
+                    )
 
             return self.score
 
@@ -96,6 +103,7 @@ class JsonCorrectnessMetric(BaseMetric):
         test_case: LLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
 
         check_llm_test_case_params(test_case, self._required_params, self)
@@ -126,7 +134,10 @@ class JsonCorrectnessMetric(BaseMetric):
                     f"Score: {self.score}\nReason: {self.reason}",
                 ],
             )
-
+            if _log_metric_to_confident:
+                metric_data_manager.post_metric_if_enabled(
+                    self, test_case=test_case
+                )
             return self.score
 
     async def a_generate_reason(self, actual_output: str) -> str:

deepeval/metrics/knowledge_retention/knowledge_retention.py

@@ -20,6 +20,7 @@ from deepeval.metrics.knowledge_retention.schema import (
     KnowledgeRetentionScoreReason,
 )
 from deepeval.utils import get_or_create_event_loop, prettify_list
+from deepeval.metrics.api import metric_data_manager
 
 
 class KnowledgeRetentionMetric(BaseConversationalMetric):
@@ -47,6 +48,7 @@ class KnowledgeRetentionMetric(BaseConversationalMetric):
         test_case: ConversationalTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ):
         check_conversational_test_case_params(
             test_case, self._required_test_case_params, self
@@ -63,6 +65,7 @@ class KnowledgeRetentionMetric(BaseConversationalMetric):
                         test_case,
                         _show_indicator=False,
                         _in_component=_in_component,
+                        _log_metric_to_confident=_log_metric_to_confident,
                     )
                 )
             else:
@@ -84,6 +87,10 @@ class KnowledgeRetentionMetric(BaseConversationalMetric):
                         f"Score: {self.score}\nReason: {self.reason}",
                     ],
                 )
+                if _log_metric_to_confident:
+                    metric_data_manager.post_metric_if_enabled(
+                        self, test_case=test_case
+                    )
             return self.score
 
     async def a_measure(
@@ -91,6 +98,7 @@ class KnowledgeRetentionMetric(BaseConversationalMetric):
         test_case: ConversationalTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
         check_conversational_test_case_params(
             test_case, self._required_test_case_params, self
@@ -120,6 +128,10 @@ class KnowledgeRetentionMetric(BaseConversationalMetric):
                     f"Score: {self.score}\nReason: {self.reason}",
                 ],
             )
+            if _log_metric_to_confident:
+                metric_data_manager.post_metric_if_enabled(
+                    self, test_case=test_case
+                )
             return self.score
 
     async def _a_generate_reason(self) -> str:

deepeval/metrics/mcp/mcp_task_completion.py

@@ -16,6 +16,7 @@ from deepeval.utils import get_or_create_event_loop, prettify_list
 from deepeval.metrics.mcp.schema import Task, TaskScore
 from deepeval.metrics.mcp.template import MCPTaskCompletionTemplate
 from deepeval.errors import MissingTestCaseParamsError
+from deepeval.metrics.api import metric_data_manager
 
 
 class MCPTaskCompletionMetric(BaseConversationalMetric):
@@ -46,6 +47,7 @@ class MCPTaskCompletionMetric(BaseConversationalMetric):
         test_case: ConversationalTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ):
         check_conversational_test_case_params(
             test_case, self._required_test_case_params, self
@@ -62,6 +64,7 @@ class MCPTaskCompletionMetric(BaseConversationalMetric):
                         test_case,
                         _show_indicator=False,
                         _in_component=_in_component,
+                        _log_metric_to_confident=_log_metric_to_confident,
                     )
                 )
             else:
@@ -90,6 +93,10 @@ class MCPTaskCompletionMetric(BaseConversationalMetric):
                         f"Score: {self.score}",
                     ],
                 )
+                if _log_metric_to_confident:
+                    metric_data_manager.post_metric_if_enabled(
+                        self, test_case=test_case
+                    )
             return self.score
 
     async def a_measure(
@@ -97,6 +104,7 @@ class MCPTaskCompletionMetric(BaseConversationalMetric):
         test_case: ConversationalTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ):
         check_conversational_test_case_params(
             test_case, self._required_test_case_params, self
@@ -131,6 +139,11 @@ class MCPTaskCompletionMetric(BaseConversationalMetric):
                     f"Score: {self.score}",
                 ],
             )
+            if _log_metric_to_confident:
+                metric_data_manager.post_metric_if_enabled(
+                    self, test_case=test_case
+                )
+
         return self.score
 
     def _generate_reason(self, task_scores: List[TaskScore]) -> str:

deepeval/metrics/mcp/multi_turn_mcp_use_metric.py

@@ -16,6 +16,7 @@ from deepeval.utils import get_or_create_event_loop, prettify_list
 from deepeval.metrics.mcp.schema import Task, ArgsScore, ToolScore
 from deepeval.metrics.mcp.template import MCPTaskCompletionTemplate
 from deepeval.errors import MissingTestCaseParamsError
+from deepeval.metrics.api import metric_data_manager
 
 
 class MultiTurnMCPUseMetric(BaseConversationalMetric):
@@ -46,6 +47,7 @@ class MultiTurnMCPUseMetric(BaseConversationalMetric):
         test_case: ConversationalTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ):
         check_conversational_test_case_params(
             test_case, self._required_test_case_params, self
@@ -62,6 +64,7 @@ class MultiTurnMCPUseMetric(BaseConversationalMetric):
                         test_case,
                         _show_indicator=False,
                         _in_component=_in_component,
+                        _log_metric_to_confident=_log_metric_to_confident,
                     )
                 )
             else:
@@ -102,6 +105,11 @@ class MultiTurnMCPUseMetric(BaseConversationalMetric):
                         f"Score: {self.score}",
                     ],
                 )
+                if _log_metric_to_confident:
+                    metric_data_manager.post_metric_if_enabled(
+                        self, test_case=test_case
+                    )
+
             return self.score
 
     async def a_measure(
@@ -109,6 +117,7 @@ class MultiTurnMCPUseMetric(BaseConversationalMetric):
         test_case: ConversationalTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ):
         check_conversational_test_case_params(
             test_case, self._required_test_case_params, self
@@ -161,6 +170,10 @@ class MultiTurnMCPUseMetric(BaseConversationalMetric):
                     f"Score: {self.score}",
                 ],
             )
+            if _log_metric_to_confident:
+                metric_data_manager.post_metric_if_enabled(
+                    self, test_case=test_case
+                )
         return self.score
 
     def _get_tool_accuracy_score(

deepeval/metrics/mcp_use_metric/mcp_use_metric.py

@@ -20,6 +20,7 @@ from deepeval.models import DeepEvalBaseLLM
 from deepeval.metrics.indicator import metric_progress_indicator
 from .template import MCPUseMetricTemplate
 from .schema import MCPPrimitivesScore, MCPArgsScore
+from deepeval.metrics.api import metric_data_manager
 
 
 class MCPUseMetric(BaseMetric):
@@ -51,6 +52,7 @@ class MCPUseMetric(BaseMetric):
         test_case: LLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
         check_llm_test_case_params(test_case, self._required_params, self)
 
@@ -65,6 +67,7 @@ class MCPUseMetric(BaseMetric):
                         test_case,
                         _show_indicator=False,
                         _in_component=_in_component,
+                        _log_metric_to_confident=_log_metric_to_confident,
                     )
                 )
             else:
@@ -104,6 +107,10 @@ class MCPUseMetric(BaseMetric):
                     self,
                     steps=steps,
                 )
+                if _log_metric_to_confident:
+                    metric_data_manager.post_metric_if_enabled(
+                        self, test_case=test_case
+                    )
 
                 return self.score
 
@@ -112,6 +119,7 @@ class MCPUseMetric(BaseMetric):
         test_case: LLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
         check_llm_test_case_params(test_case, self._required_params, self)
 
@@ -154,7 +162,10 @@ class MCPUseMetric(BaseMetric):
                 self,
                 steps=steps,
             )
-
+            if _log_metric_to_confident:
+                metric_data_manager.post_metric_if_enabled(
+                    self, test_case=test_case
+                )
             return self.score
 
     def _get_primitives_used_score(

deepeval/metrics/misuse/misuse.py

@@ -16,6 +16,7 @@ from deepeval.metrics.utils import (
 )
 from deepeval.metrics.misuse.template import MisuseTemplate
 from deepeval.metrics.misuse.schema import *
+from deepeval.metrics.api import metric_data_manager
 
 
 class MisuseMetric(BaseMetric):
@@ -53,6 +54,7 @@ class MisuseMetric(BaseMetric):
         test_case: LLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
 
         check_llm_test_case_params(test_case, self._required_params, self)
@@ -68,6 +70,7 @@ class MisuseMetric(BaseMetric):
                         test_case,
                         _show_indicator=False,
                         _in_component=_in_component,
+                        _log_metric_to_confident=_log_metric_to_confident,
                     )
                 )
             else:
@@ -86,6 +89,10 @@ class MisuseMetric(BaseMetric):
                         f"Score: {self.score}\nReason: {self.reason}",
                     ],
                 )
+                if _log_metric_to_confident:
+                    metric_data_manager.post_metric_if_enabled(
+                        self, test_case=test_case
+                    )
 
             return self.score
 
@@ -94,6 +101,7 @@ class MisuseMetric(BaseMetric):
         test_case: LLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
 
         check_llm_test_case_params(test_case, self._required_params, self)
@@ -122,7 +130,10 @@ class MisuseMetric(BaseMetric):
                     f"Score: {self.score}\nReason: {self.reason}",
                 ],
             )
-
+            if _log_metric_to_confident:
+                metric_data_manager.post_metric_if_enabled(
+                    self, test_case=test_case
+                )
             return self.score
 
     async def _a_generate_reason(self) -> str:

deepeval/metrics/multimodal_metrics/image_coherence/image_coherence.py

@@ -48,6 +48,7 @@ class ImageCoherenceMetric(BaseMultimodalMetric):
         test_case: MLLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
         check_mllm_test_case_params(
             test_case, self._required_params, None, None, self
@@ -63,6 +64,7 @@ class ImageCoherenceMetric(BaseMultimodalMetric):
                         test_case,
                         _show_indicator=False,
                         _in_component=_in_component,
+                        _log_metric_to_confident=_log_metric_to_confident,
                     )
                 )
             else:
@@ -146,6 +148,7 @@ class ImageCoherenceMetric(BaseMultimodalMetric):
         test_case: MLLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
         check_mllm_test_case_params(
             test_case, self._required_params, None, None, self

deepeval/metrics/multimodal_metrics/image_editing/image_editing.py

@@ -47,6 +47,7 @@ class ImageEditingMetric(BaseMultimodalMetric):
         test_case: MLLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
         check_mllm_test_case_params(
             test_case, self._required_params, 1, 1, self
@@ -63,6 +64,7 @@ class ImageEditingMetric(BaseMultimodalMetric):
                         test_case,
                         _show_indicator=False,
                         _in_component=_in_component,
+                        _log_metric_to_confident=_log_metric_to_confident,
                     )
                 )
             else:
@@ -108,6 +110,7 @@ class ImageEditingMetric(BaseMultimodalMetric):
         test_case: MLLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
         check_mllm_test_case_params(
             test_case, self._required_params, 1, 1, self

deepeval/metrics/multimodal_metrics/image_helpfulness/image_helpfulness.py

@@ -49,6 +49,7 @@ class ImageHelpfulnessMetric(BaseMultimodalMetric):
         test_case: MLLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
         check_mllm_test_case_params(
             test_case, self._required_params, None, None, self
@@ -64,6 +65,7 @@ class ImageHelpfulnessMetric(BaseMultimodalMetric):
                         test_case,
                         _show_indicator=False,
                         _in_component=_in_component,
+                        _log_metric_to_confident=_log_metric_to_confident,
                     )
                 )
             else:
@@ -147,6 +149,7 @@ class ImageHelpfulnessMetric(BaseMultimodalMetric):
         test_case: MLLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
         check_mllm_test_case_params(
             test_case, self._required_params, None, None, self

deepeval/metrics/multimodal_metrics/image_reference/image_reference.py

@@ -49,6 +49,7 @@ class ImageReferenceMetric(BaseMultimodalMetric):
         test_case: MLLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
         check_mllm_test_case_params(
             test_case, self._required_params, None, None, self
@@ -64,6 +65,7 @@ class ImageReferenceMetric(BaseMultimodalMetric):
                         test_case,
                         _show_indicator=False,
                         _in_component=_in_component,
+                        _log_metric_to_confident=_log_metric_to_confident,
                     )
                 )
             else:
@@ -147,6 +149,7 @@ class ImageReferenceMetric(BaseMultimodalMetric):
         test_case: MLLMTestCase,
         _show_indicator: bool = True,
         _in_component: bool = False,
+        _log_metric_to_confident: bool = True,
     ) -> float:
         check_mllm_test_case_params(
             test_case, self._required_params, None, None, self