azure-ai-evaluation 1.1.0__py3-none-any.whl → 1.3.0__py3-none-any.whl

azure/ai/evaluation/simulator/_adversarial_simulator.py

@@ -7,6 +7,7 @@ import asyncio
 import logging
 import random
 from typing import Any, Callable, Dict, List, Optional, Union, cast
+import uuid
 
 from tqdm import tqdm
 
@@ -187,6 +188,8 @@ class AdversarialSimulator:
             )
         self._ensure_service_dependencies()
         templates = await self.adversarial_template_handler._get_content_harm_template_collections(scenario.value)
+        simulation_id = str(uuid.uuid4())
+        logger.warning("Use simulation_id to help debug the issue: %s", str(simulation_id))
         concurrent_async_task = min(concurrent_async_task, 1000)
         semaphore = asyncio.Semaphore(concurrent_async_task)
         sim_results = []
@@ -236,6 +239,7 @@ class AdversarialSimulator:
                             language=language,
                             semaphore=semaphore,
                             scenario=scenario,
+                            simulation_id=simulation_id,
                         )
                     )
                 )
@@ -298,9 +302,10 @@ class AdversarialSimulator:
         language: SupportedLanguages,
         semaphore: asyncio.Semaphore,
         scenario: Union[AdversarialScenario, AdversarialScenarioJailbreak],
+        simulation_id: str = "",
     ) -> List[Dict]:
         user_bot = self._setup_bot(
-            role=ConversationRole.USER, template=template, parameters=parameters, scenario=scenario
+            role=ConversationRole.USER, template=template, parameters=parameters, scenario=scenario, simulation_id=simulation_id
         )
         system_bot = self._setup_bot(
             target=target, role=ConversationRole.ASSISTANT, template=template, parameters=parameters, scenario=scenario
@@ -329,7 +334,7 @@ class AdversarialSimulator:
         )
 
     def _get_user_proxy_completion_model(
-        self, template_key: str, template_parameters: TemplateParameters
+        self, template_key: str, template_parameters: TemplateParameters, simulation_id: str = ""
     ) -> ProxyChatCompletionsModel:
         return ProxyChatCompletionsModel(
             name="raisvc_proxy_model",
@@ -340,6 +345,7 @@ class AdversarialSimulator:
             api_version="2023-07-01-preview",
             max_tokens=1200,
             temperature=0.0,
+            simulation_id=simulation_id,
         )
 
     def _setup_bot(
@@ -350,10 +356,11 @@ class AdversarialSimulator:
         parameters: TemplateParameters,
         target: Optional[Callable] = None,
         scenario: Union[AdversarialScenario, AdversarialScenarioJailbreak],
+        simulation_id: str = "",
     ) -> ConversationBot:
         if role is ConversationRole.USER:
             model = self._get_user_proxy_completion_model(
-                template_key=template.template_name, template_parameters=parameters
+                template_key=template.template_name, template_parameters=parameters, simulation_id=simulation_id,
             )
             return ConversationBot(
                 role=role,

azure/ai/evaluation/simulator/_conversation/__init__.py

@@ -128,19 +128,15 @@ class ConversationBot:
         self.conversation_starter: Optional[Union[str, jinja2.Template, Dict]] = None
         if role == ConversationRole.USER:
             if "conversation_starter" in self.persona_template_args:
-                print(self.persona_template_args)
                 conversation_starter_content = self.persona_template_args["conversation_starter"]
                 if isinstance(conversation_starter_content, dict):
                     self.conversation_starter = conversation_starter_content
-                    print(f"Conversation starter content: {conversation_starter_content}")
                 else:
                     try:
                         self.conversation_starter = jinja2.Template(
                             conversation_starter_content, undefined=jinja2.StrictUndefined
                         )
-                        print("Successfully created a Jinja2 template for the conversation starter.")
                     except jinja2.exceptions.TemplateSyntaxError as e:  # noqa: F841
-                        print(f"Template syntax error: {e}. Using raw content.")
                         self.conversation_starter = conversation_starter_content
             else:
                 self.logger.info(
@@ -153,6 +149,7 @@ class ConversationBot:
         conversation_history: List[ConversationTurn],
         max_history: int,
         turn_number: int = 0,
+        session_state: Optional[Dict[str, Any]] = None,
     ) -> Tuple[dict, dict, float, dict]:
         """
         Prompt the ConversationBot for a response.
@@ -262,6 +259,7 @@ class CallbackConversationBot(ConversationBot):
         conversation_history: List[Any],
         max_history: int,
         turn_number: int = 0,
+        session_state: Optional[Dict[str, Any]] = None,
     ) -> Tuple[dict, dict, float, dict]:
         chat_protocol_message = self._to_chat_protocol(
             self.user_template, conversation_history, self.user_template_parameters
@@ -269,7 +267,7 @@ class CallbackConversationBot(ConversationBot):
         msg_copy = copy.deepcopy(chat_protocol_message)
         result = {}
         start_time = time.time()
-        result = await self.callback(msg_copy)
+        result = await self.callback(msg_copy, session_state=session_state)
         end_time = time.time()
         if not result:
             result = {
@@ -348,6 +346,7 @@ class MultiModalConversationBot(ConversationBot):
         conversation_history: List[Any],
         max_history: int,
         turn_number: int = 0,
+        session_state: Optional[Dict[str, Any]] = None,
     ) -> Tuple[dict, dict, float, dict]:
         previous_prompt = conversation_history[-1]
         chat_protocol_message = await self._to_chat_protocol(conversation_history, self.user_template_parameters)

azure/ai/evaluation/simulator/_conversation/_conversation.py

@@ -101,6 +101,7 @@ async def simulate_conversation(
     :rtype: Tuple[Optional[str], List[ConversationTurn]]
     """
 
+    session_state = {}
     # Read the first prompt.
     (first_response, request, _, full_response) = await bots[0].generate_response(
         session=session,
@@ -149,7 +150,10 @@ async def simulate_conversation(
                 conversation_history=conversation_history,
                 max_history=history_limit,
                 turn_number=current_turn,
+                session_state=session_state,
             )
+            if "session_state" in full_response and full_response["session_state"] is not None:
+                session_state.update(full_response["session_state"])
 
             # check if conversation id is null, which means conversation starter was used. use id from next turn
             if conversation_id is None and "id" in response:

azure/ai/evaluation/simulator/_model_tools/_proxy_completion_model.py

@@ -89,6 +89,7 @@ class ProxyChatCompletionsModel(OpenAIChatCompletionsModel):
         self.tkey = template_key
         self.tparam = template_parameters
         self.result_url: Optional[str] = None
+        self.simulation_id: Optional[str] = kwargs.pop("simulation_id", "")
 
         super().__init__(name=name, **kwargs)
 
@@ -169,6 +170,7 @@ class ProxyChatCompletionsModel(OpenAIChatCompletionsModel):
             "Content-Type": "application/json",
             "X-CV": f"{uuid.uuid4()}",
             "X-ModelType": self.model or "",
+            "x-ms-client-request-id": self.simulation_id,
         }
         # add all additional headers
         headers.update(self.additional_headers)  # type: ignore[arg-type]

azure/ai/evaluation/simulator/_simulator.py

@@ -157,7 +157,7 @@ class Simulator:
                 f"You have specified 'num_queries' < len('tasks') ({num_queries} < {len(tasks)}). "
                 f"Only the first {num_queries} lines of the specified tasks will be simulated."
             )
-        num_queries = min(num_queries, len(tasks))
+
         max_conversation_turns *= 2  # account for both user and assistant turns
 
         prompty_model_config = self.model_config
@@ -586,7 +586,10 @@ class Simulator:
         for i, query_response_pair in enumerate(query_responses):
             query = query_response_pair["q"]
             response = query_response_pair["r"]
-            task = tasks[i]
+            try:
+                task = tasks[i]
+            except IndexError:
+                task = None
 
             conversation = await self._complete_conversation(
                 conversation_starter=query,
@@ -621,7 +624,7 @@ class Simulator:
         *,
         conversation_starter: str,
         max_conversation_turns: int,
-        task: str,
+        task: Optional[str],
         user_simulator_prompty: Optional[str],
         user_simulator_prompty_options: Dict[str, Any],
         target: Callable,
@@ -659,16 +662,21 @@ class Simulator:
                 user_simulator_prompty_options=user_simulator_prompty_options,
             )
             if len(conversation_history) == 0:
-                conversation_starter_from_simulated_user = await user_flow(
-                    task=task,
-                    conversation_history=[
-                        {
-                            "role": "assistant",
-                            "content": conversation_starter,
-                        }
-                    ],
-                    action="rewrite the assistant's message as you have to accomplish the task by asking the right questions. Make sure the original question is not lost in your rewrite.",
-                )
+                if task:
+                    conversation_starter_from_simulated_user = await user_flow(
+                        task=task,
+                        conversation_history=[
+                            {
+                                "role": "assistant",
+                                "content": conversation_starter,
+                            }
+                        ],
+                        action="rewrite the assistant's message as you have to accomplish the task by asking the right questions. Make sure the original question is not lost in your rewrite.",
+                    )
+                else:
+                    conversation_starter_from_simulated_user = {
+                        "content": conversation_starter,
+                    }
             else:
                 conversation_starter_from_simulated_user = await user_flow(
                     task=task,

{azure_ai_evaluation-1.1.0.dist-info → azure_ai_evaluation-1.3.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: azure-ai-evaluation
-Version: 1.1.0
+Version: 1.3.0
 Summary: Microsoft Azure Evaluation Library for Python
 Home-page: https://github.com/Azure/azure-sdk-for-python
 Author: Microsoft Corporation
@@ -13,17 +13,16 @@ Classifier: Development Status :: 5 - Production/Stable
 Classifier: Programming Language :: Python
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3 :: Only
-Classifier: Programming Language :: Python :: 3.8
 Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: License :: OSI Approved :: MIT License
 Classifier: Operating System :: OS Independent
-Requires-Python: >=3.8
+Requires-Python: >=3.9
 Description-Content-Type: text/markdown
 License-File: NOTICE.txt
-Requires-Dist: promptflow-devkit >=1.15.0
-Requires-Dist: promptflow-core >=1.15.0
+Requires-Dist: promptflow-devkit >=1.17.1
+Requires-Dist: promptflow-core >=1.17.1
 Requires-Dist: pyjwt >=2.8.0
 Requires-Dist: azure-identity >=1.16.0
 Requires-Dist: azure-core >=1.30.2
@@ -54,7 +53,7 @@ Azure AI SDK provides following to evaluate Generative AI Applications:
 
 ### Prerequisites
 
-- Python 3.8 or later is required to use this package.
+- Python 3.9 or later is required to use this package.
 - [Optional] You must have [Azure AI Project][ai_project] or [Azure Open AI][azure_openai] to use AI-assisted evaluators
 
 ### Install the package
@@ -378,11 +377,82 @@ This project has adopted the [Microsoft Open Source Code of Conduct][code_of_con
 
 # Release History
 
-## 1.1.0 (2024-12-12)
+## 1.3.0 (2025-02-28)
+
+### Breaking Changes
+- Multimodal specific evaluators `ContentSafetyMultimodalEvaluator`, `ViolenceMultimodalEvaluator`, `SexualMultimodalEvaluator`, `SelfHarmMultimodalEvaluator`, `HateUnfairnessMultimodalEvaluator` and `ProtectedMaterialMultimodalEvaluator` has been removed. Please use `ContentSafetyEvaluator`, `ViolenceEvaluator`, `SexualEvaluator`, `SelfHarmEvaluator`, `HateUnfairnessEvaluator` and `ProtectedMaterialEvaluator` instead.
+- Metric name in ProtectedMaterialEvaluator's output is changed from `protected_material.fictional_characters_label` to `protected_material.fictional_characters_defect_rate`. It's now consistent with other evaluator's metric names (ending with `_defect_rate`).
+
+## 1.2.0 (2025-01-27)
+
+### Features Added
+- CSV files are now supported as data file inputs with `evaluate()` API. The CSV file should have a header row with column names that match the `data` and `target` fields in the `evaluate()` method and the filename should be passed as the `data` parameter. Column name 'Conversation' in CSV file is not fully supported yet.  
+
+### Breaking Changes
+- `ViolenceMultimodalEvaluator`, `SexualMultimodalEvaluator`, `SelfHarmMultimodalEvaluator`, `HateUnfairnessMultimodalEvaluator` and `ProtectedMaterialMultimodalEvaluator` will be removed in next release. 
 
 ### Bugs Fixed
 - Removed `[remote]` extra. This is no longer needed when tracking results in Azure AI Studio.
 - Fixed `AttributeError: 'NoneType' object has no attribute 'get'` while running simulator with 1000+ results
+- Fixed the non adversarial simulator to run in task-free mode
+- Content safety evaluators (violence, self harm, sexual, hate/unfairness) return the maximum result as the
+  main score when aggregating per-turn evaluations from a conversation into an overall
+  evaluation score. Other conversation-capable evaluators still default to a mean for aggregation.
+- Fixed bug in non adversarial simulator sample where `tasks` undefined
+
+### Other Changes
+- Changed minimum required python version to use this package from 3.8 to 3.9
+- Stop dependency on the local promptflow service. No promptflow service will automatically start when running evaluation.
+- Evaluators internally allow for custom aggregation. However, this causes serialization failures if evaluated while the
+  environment variable `AI_EVALS_BATCH_USE_ASYNC` is set to false.
+
+## 1.1.0 (2024-12-12)
+
+### Features Added
+- Added image support in `ContentSafetyEvaluator`, `ViolenceEvaluator`, `SexualEvaluator`, `SelfHarmEvaluator`, `HateUnfairnessEvaluator` and `ProtectedMaterialEvaluator`. Provide image URLs or base64 encoded images in `conversation` input for image evaluation. See below for an example:
+
+```python
+evaluator = ContentSafetyEvaluator(credential=azure_cred, azure_ai_project=project_scope)
+conversation = {
+    "messages": [
+        {
+            "role": "system",
+            "content": [
+                {"type": "text", "text": "You are an AI assistant that understands images."}
+            ],
+        },
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": "Can you describe this image?"},
+                {
+                    "type": "image_url",
+                    "image_url": {
+                        "url": "https://cdn.britannica.com/68/178268-050-5B4E7FB6/Tom-Cruise-2013.jpg"
+                    },
+                },
+            ],
+        },
+        {
+            "role": "assistant",
+            "content": [
+                {
+                    "type": "text",
+                    "text": "The image shows a man with short brown hair smiling, wearing a dark-colored shirt.",
+                }
+            ],
+        },
+    ]
+}
+print("Calling Content Safety Evaluator for multi-modal")
+score = evaluator(conversation=conversation)
+```
+
+- Please switch to generic evaluators for image evaluations as mentioned above. `ContentSafetyMultimodalEvaluator`, `ContentSafetyMultimodalEvaluatorBase`, `ViolenceMultimodalEvaluator`, `SexualMultimodalEvaluator`, `SelfHarmMultimodalEvaluator`, `HateUnfairnessMultimodalEvaluator` and `ProtectedMaterialMultimodalEvaluator` will be deprecated in the next release.
+
+### Bugs Fixed
+- Removed `[remote]` extra. This is no longer needed when tracking results in Azure AI Foundry portal.
+- Fixed `AttributeError: 'NoneType' object has no attribute 'get'` while running simulator with 1000+ results
 
 ## 1.0.1 (2024-11-15)
 

{azure_ai_evaluation-1.1.0.dist-info → azure_ai_evaluation-1.3.0.dist-info}/RECORD

@@ -1,26 +1,26 @@
-azure/ai/evaluation/__init__.py,sha256=MFxJRoKfSsP_Qlfq0FwynxNf4csNAfTYPQX7jdXc9RU,2757
-azure/ai/evaluation/_constants.py,sha256=d41rQb-w2GmCMHOwiyDD1ieJB1U6JyPPl6APZSJbKzg,2036
-azure/ai/evaluation/_exceptions.py,sha256=MsTbgsPGYPzIxs7MyLKzSeiVKEoCxYkVjONzNfv2tXA,5162
+azure/ai/evaluation/__init__.py,sha256=3lV2dhUqPzUz4lBo8DoEr5o4PEF_Lh5yWQtuLDUQZCM,2192
+azure/ai/evaluation/_constants.py,sha256=Avp26R4qPqaze024EJXzAX8A457DExF5rnIKMWcE8X4,2881
+azure/ai/evaluation/_exceptions.py,sha256=o3AjCyIAIBo_rkJAH_TsAT-hEZ4SMKd9Uo0eLmD9VHY,5085
 azure/ai/evaluation/_http_utils.py,sha256=1bGce6pKAL-vmaUGRPxVX7DVO05XVQ8YPIwIQ3q7mfA,17221
 azure/ai/evaluation/_model_configurations.py,sha256=MNN6cQlz7P9vNfHmfEKsUcly3j1FEOEFsA8WV7GPuKQ,4043
 azure/ai/evaluation/_user_agent.py,sha256=O2y-QPBAcw7w7qQ6M2aRPC3Vy3TKd789u5lcs2yuFaI,290
-azure/ai/evaluation/_version.py,sha256=LzMvSuUB6pmU-LfCPzoYuCoTF0BAqE7ljPjk6r8YaMw,199
+azure/ai/evaluation/_version.py,sha256=xsol5X6WDVwo8FFtHWt28TPyzT0is0sJ3N5i9_ikPVI,229
 azure/ai/evaluation/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 azure/ai/evaluation/_azure/__init__.py,sha256=Yx1Iq2GNKQ5lYxTotvPwkPL4u0cm6YVxUe-iVbu1clI,180
-azure/ai/evaluation/_azure/_clients.py,sha256=1mFRSxt-Ld5UBn-m3DJkKc-VPP9CbXQHrqLNdLs9RF0,8201
-azure/ai/evaluation/_azure/_models.py,sha256=tKpxjb5Ou476UasjXPCiuvsxTjLTrnoVnSXy5Bfa51M,12483
+azure/ai/evaluation/_azure/_clients.py,sha256=N1V-LyQkItPuoKl0aieypFPdGSRSld9lQqH1x-n3L7U,9119
+azure/ai/evaluation/_azure/_models.py,sha256=7EHmrCAFOscYY3H90wfmdPPdy0sqnOlgGFvraq_L-2Q,12517
 azure/ai/evaluation/_azure/_token_manager.py,sha256=1NZHwgEc9BMXWPz5Ear_J5-oYjouD77crLHHqNLldEw,5193
 azure/ai/evaluation/_common/__init__.py,sha256=LHTkf6dMLLxikrGNgbUuREBVQcs4ORHR6Eryo4bm9M8,586
 azure/ai/evaluation/_common/_experimental.py,sha256=GVtSn9r1CeR_yEa578dJVNDJ3P24eqe8WYdH7llbiQY,5694
 azure/ai/evaluation/_common/constants.py,sha256=OsExttFGLnTAyZa26jnY5_PCDTb7uJNFqtE2qsRZ1mg,1957
 azure/ai/evaluation/_common/math.py,sha256=d4bwWe35_RWDIZNcbV1BTBbHNx2QHQ4-I3EofDyyNE0,2863
 azure/ai/evaluation/_common/rai_service.py,sha256=DcakzdOour9qNdMXU-8UFfvLb12oexAoiJXG8XFTRBs,26462
-azure/ai/evaluation/_common/utils.py,sha256=MQIZs95gH5je1L-S3twa_WQi071zRu0Dv54lzCI7ZgU,17642
+azure/ai/evaluation/_common/utils.py,sha256=wssBc9i0JrWwsrvtp45P5_z5ceB7kVGf-KKzpfVQQH4,17594
 azure/ai/evaluation/_evaluate/__init__.py,sha256=Yx1Iq2GNKQ5lYxTotvPwkPL4u0cm6YVxUe-iVbu1clI,180
-azure/ai/evaluation/_evaluate/_eval_run.py,sha256=het3cxjK4J-_hT19dT5a0mC2Cdnk93gM3ONQMJb9bxQ,21923
-azure/ai/evaluation/_evaluate/_evaluate.py,sha256=P5aL70eUBKZT9CVRM9RVSfD0DkuljQyc5ECte37Ycmo,36225
-azure/ai/evaluation/_evaluate/_utils.py,sha256=S4LUUDUBo9JNA41ojSezMC-PZzkWcihhhNdyZwZrpr0,13428
-azure/ai/evaluation/_evaluate/_batch_run/__init__.py,sha256=G8McpeLxAS_gFhNShX52_YWvE-arhJn-bVpAfzjWG3Q,427
+azure/ai/evaluation/_evaluate/_eval_run.py,sha256=QBtNBwUxqxsIVmKPU-_H8MDFkF4s_bW7arQYXAniRpo,21965
+azure/ai/evaluation/_evaluate/_evaluate.py,sha256=pyl0HscBAPllpq385p9lMqR2sqip3RZ5BGfcBa4X334,39068
+azure/ai/evaluation/_evaluate/_utils.py,sha256=sKj_4iN-QjrRlEkiZwA9UNiWozS4LgJcUZ6AWdHrTY4,14231
+azure/ai/evaluation/_evaluate/_batch_run/__init__.py,sha256=Z-TQdSxKTn0bjsF0YosIJMbQFQHDUv_b9zCBu1TeogQ,474
 azure/ai/evaluation/_evaluate/_batch_run/code_client.py,sha256=XQLaXfswF6ReHLpQthHLuLLa65Pts8uawGp7kRqmMDs,8260
 azure/ai/evaluation/_evaluate/_batch_run/eval_run_context.py,sha256=p3Bsg_shGs5RXvysOlvo0CQb4Te5herSvX1OP6ylFUQ,3543
 azure/ai/evaluation/_evaluate/_batch_run/proxy_client.py,sha256=T_QRHScDMBM4O6ejkkKdBmHPjH2NOF6owW48aVUYF6k,3775
@@ -28,47 +28,41 @@ azure/ai/evaluation/_evaluate/_batch_run/target_run_context.py,sha256=SMos3bVmD7
 azure/ai/evaluation/_evaluate/_telemetry/__init__.py,sha256=fhLqE41qxdjfBOGi23cpk6QgUe-s1Fw2xhAAUjNESF0,7045
 azure/ai/evaluation/_evaluators/__init__.py,sha256=Yx1Iq2GNKQ5lYxTotvPwkPL4u0cm6YVxUe-iVbu1clI,180
 azure/ai/evaluation/_evaluators/_bleu/__init__.py,sha256=quKKO0kvOSkky5hcoNBvgBuMeeVRFCE9GSv70mAdGP4,260
-azure/ai/evaluation/_evaluators/_bleu/_bleu.py,sha256=iT20SMmEtOnh7RWs55dFfAlKXNkNceXkCUbVyqv6aQ0,2776
+azure/ai/evaluation/_evaluators/_bleu/_bleu.py,sha256=Px3KxTyNIuxy-4U3SE4XJHCd4r144JeVrlIGMdbaqBk,3425
 azure/ai/evaluation/_evaluators/_coherence/__init__.py,sha256=GRqcSCQse02Spyki0UsRNWMIXiea2lLtPPXNGvkJzQ0,258
 azure/ai/evaluation/_evaluators/_coherence/_coherence.py,sha256=uG9hX2XWkMREKfMAWRoosjicoI4Lg3ptR3UcLEgKd0c,4643
 azure/ai/evaluation/_evaluators/_coherence/coherence.prompty,sha256=ANvh9mDFW7KMejrgdWqBLjj4SIqEO5WW9gg5pE0RLJk,6798
-azure/ai/evaluation/_evaluators/_common/__init__.py,sha256=_hPqTkAla_O6s4ebVtTaBrVLEW3KSdDz66WwxjK50cI,423
-azure/ai/evaluation/_evaluators/_common/_base_eval.py,sha256=n6qldJr8d8H0DnS7IwkQPH9Ep9PdZnVeVtSxQiunADc,19424
+azure/ai/evaluation/_evaluators/_common/__init__.py,sha256=xAymP_CZy4aPzWplMdXgQUQVDIUEMI-0nbgdm_umFYY,498
+azure/ai/evaluation/_evaluators/_common/_base_eval.py,sha256=LtlTsA0TUywMXEYj3mVggv43G0TfKnNkDG6ZgA_dWa4,23328
+azure/ai/evaluation/_evaluators/_common/_base_multi_eval.py,sha256=6WFmFMsobJjju3wzVFKx7EjuHqbBV9YXzlhbwu5vzio,2509
 azure/ai/evaluation/_evaluators/_common/_base_prompty_eval.py,sha256=hvJD7jR2ESePkRPN17ytoFhFiS0iTotOfeqmTwG2IMs,4531
-azure/ai/evaluation/_evaluators/_common/_base_rai_svc_eval.py,sha256=czyn1MfaxOmrvvFgdeblf6FaauKgKolgPFsP5f7K29w,7331
+azure/ai/evaluation/_evaluators/_common/_base_rai_svc_eval.py,sha256=1ZwWu2dwN2y6bVOU3Ws4VvJoMVQ80tzYNutSEfmpYmg,7830
+azure/ai/evaluation/_evaluators/_common/_conversation_aggregators.py,sha256=gjDBjRxJKwaHbshWH0j2idjlzfzNMnT9a9RL0fQiKeM,2129
 azure/ai/evaluation/_evaluators/_content_safety/__init__.py,sha256=PEYMIybfP64f7byhuTaiq4RiqsYbjqejpW1JsJIG1jA,556
-azure/ai/evaluation/_evaluators/_content_safety/_content_safety.py,sha256=CIGfBLNOTVXrlF5HIc2UpuDDG5BfzjD7ubJ23CbvobQ,6341
-azure/ai/evaluation/_evaluators/_content_safety/_hate_unfairness.py,sha256=sjw8FfwxC1f0K1J4TkeA8wkfq88aebiNbaKzS-8DWzk,5919
-azure/ai/evaluation/_evaluators/_content_safety/_self_harm.py,sha256=0zaB-JKm8FU6yoxD1nqoYvxp3gvjuZfcQjb-xhSHoQ0,5156
-azure/ai/evaluation/_evaluators/_content_safety/_sexual.py,sha256=q9bEMu6Dp1wxDlH3h2iTayrWv4ux-izLB0kGkxrgEhM,5396
-azure/ai/evaluation/_evaluators/_content_safety/_violence.py,sha256=W2QwPuWOc3nkLvvWOAhCrpLRDAAo-xG1SvlDhrshzUc,5467
+azure/ai/evaluation/_evaluators/_content_safety/_content_safety.py,sha256=XKnIlxbzkB65cRXXcOGsv0W37QKxo_jsHbR3gijMZ78,4654
+azure/ai/evaluation/_evaluators/_content_safety/_hate_unfairness.py,sha256=LcnJuePAwByoaXAQ5CVKnkO2IVCCRdVnFTUYbOyQTbs,6043
+azure/ai/evaluation/_evaluators/_content_safety/_self_harm.py,sha256=Nl_hTIRXO_UdAjUatPzbCTextsngkgib1ECzsmDHUvE,5280
+azure/ai/evaluation/_evaluators/_content_safety/_sexual.py,sha256=z8bDdkZHW09D6NTY9mlK2abNMOy9hRAJwwTQs5vjvAc,5520
+azure/ai/evaluation/_evaluators/_content_safety/_violence.py,sha256=z9dM3GOBSIw_WoEELPHRE3DSK3ol7MZbDkFJyuYENVk,5591
 azure/ai/evaluation/_evaluators/_eci/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 azure/ai/evaluation/_evaluators/_eci/_eci.py,sha256=a36sLZPHKi3YAdl0JvpL6vboZMqgGjnmz0qZ-o8vcWY,2934
 azure/ai/evaluation/_evaluators/_f1_score/__init__.py,sha256=aEVbO7iMoF20obdpLQKcKm69Yyu3mYnblKELLqu8OGI,260
-azure/ai/evaluation/_evaluators/_f1_score/_f1_score.py,sha256=YtPEG1ZT0jAPvEnOpD2Eaojm-8zS61bxOr3US6vvgqc,5779
+azure/ai/evaluation/_evaluators/_f1_score/_f1_score.py,sha256=nDUAz-vmkIR0Sj7JfMm3mVHfE3XnKrZnTzOUa6QKskk,5399
 azure/ai/evaluation/_evaluators/_fluency/__init__.py,sha256=EEJw39xRa0bOAA1rELTTKXQu2s60n_7CZQRD0Gu2QVw,259
 azure/ai/evaluation/_evaluators/_fluency/_fluency.py,sha256=mHQCismdL4cCeANcqWrDHCiVgr4UAWj0yIYJXt2pFDA,4399
 azure/ai/evaluation/_evaluators/_fluency/fluency.prompty,sha256=n9v0W9eYwgIO-JSsLTSKEM_ApJuxxuKWQpNblrTEkFY,4861
 azure/ai/evaluation/_evaluators/_gleu/__init__.py,sha256=Ae2EvQ7gqiYAoNO3LwGIhdAAjJPJDfT85rQGKrRrmbA,260
-azure/ai/evaluation/_evaluators/_gleu/_gleu.py,sha256=RaY_RZ5A3sMx4yE6uCyjvchB8rRoMvIv0JYYyMBXFM8,2696
+azure/ai/evaluation/_evaluators/_gleu/_gleu.py,sha256=E_HeUuDAW2pPhsbaWLHMMxqgUxPOgBv2Bnr_Z9M6AAs,3359
 azure/ai/evaluation/_evaluators/_groundedness/__init__.py,sha256=UYNJUeRvBwcSVFyZpdsf29un5eyaDzYoo3QvC1gvlLg,274
 azure/ai/evaluation/_evaluators/_groundedness/_groundedness.py,sha256=Zil5S7BXaVvW2wBUlsF3oGzZLOYrvSzGAY4TqKfFUX8,6876
 azure/ai/evaluation/_evaluators/_groundedness/groundedness_with_query.prompty,sha256=v7TOm75DyW_1gOU6gSiZoPcRnHcJ65DrzR2cL_ucWDY,5814
 azure/ai/evaluation/_evaluators/_groundedness/groundedness_without_query.prompty,sha256=8kNShdfxQvkII7GnqjmdqQ5TNelA2B6cjnqWZk8FFe4,5296
 azure/ai/evaluation/_evaluators/_meteor/__init__.py,sha256=209na3pPsdmcuYpYHUYtqQybCpc3yZkc93HnRdicSlI,266
-azure/ai/evaluation/_evaluators/_meteor/_meteor.py,sha256=UPNvWpNkMlx8NmOPuSkcXF1DA_daDdrRArhJAbbTQkc,3767
-azure/ai/evaluation/_evaluators/_multimodal/__init__.py,sha256=tPvsY0nv8T3VtiiAwJM6wT5A9FhKP2XXwUlCH994xl4,906
-azure/ai/evaluation/_evaluators/_multimodal/_content_safety_multimodal.py,sha256=x0l6eLQhxVP85jEyGfFCl27C2okMgD0S3aJ_qrgB3Q8,5219
-azure/ai/evaluation/_evaluators/_multimodal/_content_safety_multimodal_base.py,sha256=X2IVw0YvymDD3e4Vx-TfjqgqtYiAKVhUumjBowCpOmA,2441
-azure/ai/evaluation/_evaluators/_multimodal/_hate_unfairness.py,sha256=ral1AAbP5pfsygDe30MtuwajuydiXoXzzCeuLBzIkWc,3779
-azure/ai/evaluation/_evaluators/_multimodal/_protected_material.py,sha256=gMrfyn3KHcV6SoowuEjR7Fon9vVLN7GOPM4rkJRK6xU,4906
-azure/ai/evaluation/_evaluators/_multimodal/_self_harm.py,sha256=QwOCBb618ZXSs-OoVXyNM65N4ZEL7IZt-S1Nqd8xNbY,3703
-azure/ai/evaluation/_evaluators/_multimodal/_sexual.py,sha256=6zz89yzr_SdldqBVv-3wOErz3H5sBO6wYgNh39aHXmY,3668
-azure/ai/evaluation/_evaluators/_multimodal/_violence.py,sha256=t1h3bY6N7SwlSgP_1P-90KGTsq1oWvTYDJpy_uMvzjA,3694
+azure/ai/evaluation/_evaluators/_meteor/_meteor.py,sha256=OpugAjIgcTcNQ6g6Rks_8GVhcRiH524PbmBKH3bTefs,4369
 azure/ai/evaluation/_evaluators/_protected_material/__init__.py,sha256=eRAQIU9diVXfO5bp6aLWxZoYUvOsrDIfy1gnDOeNTiI,109
 azure/ai/evaluation/_evaluators/_protected_material/_protected_material.py,sha256=IABs1YMBZdIi1u57dPi-aQpSiPWIGxEZ4hyt97jvdNA,4604
 azure/ai/evaluation/_evaluators/_qa/__init__.py,sha256=bcXfT--C0hjym2haqd1B2-u9bDciyM0ThOFtU1Q69sk,244
-azure/ai/evaluation/_evaluators/_qa/_qa.py,sha256=kLkXwkmrXqgfBu7MJwEYAobeqGh4b4zE7cjIkD_1iwA,3854
+azure/ai/evaluation/_evaluators/_qa/_qa.py,sha256=HG4JiSt5S20D70LmzW8t24qWg5_uiIKwIxjJ13ygfDo,3670
 azure/ai/evaluation/_evaluators/_relevance/__init__.py,sha256=JlxytW32Nl8pbE-fI3GRpfgVuY9EG6zxIAn5VZGSwyc,265
 azure/ai/evaluation/_evaluators/_relevance/_relevance.py,sha256=S1J5BR1-ZyCLQOTbdAHLDzzY1ccVnPyy9uVUlivmCx0,5287
 azure/ai/evaluation/_evaluators/_relevance/relevance.prompty,sha256=VHKzVlC2Cv1xuholgIGmerPspspAI0t6IgJ2cxOuYDE,4811
@@ -76,14 +70,16 @@ azure/ai/evaluation/_evaluators/_retrieval/__init__.py,sha256=kMu47ZyTZ7f-4Yh6H3
 azure/ai/evaluation/_evaluators/_retrieval/_retrieval.py,sha256=fmd8zNOVSGQGT5icSAI6PwgnS7kKz_ZMKMnxKIchYl8,5085
 azure/ai/evaluation/_evaluators/_retrieval/retrieval.prompty,sha256=_YVoO4Gt_WD42bUcj5n6BDW0dMUqNf0yF3Nj5XMOX2c,16490
 azure/ai/evaluation/_evaluators/_rouge/__init__.py,sha256=kusCDaYcXogDugGefRP8MQSn9xv107oDbrMCqZ6K4GA,291
-azure/ai/evaluation/_evaluators/_rouge/_rouge.py,sha256=SV5rESLVARQqh1n0Pf6EMvJoJH3A0nNKM_U33q1LQoE,4026
+azure/ai/evaluation/_evaluators/_rouge/_rouge.py,sha256=xTsAF9em2QpWcgCHOmDCEcuRwrob5kPUUpkgul9E5jA,4642
 azure/ai/evaluation/_evaluators/_service_groundedness/__init__.py,sha256=0DODUGTOgaYyFbO9_zxuwifixDL3SIm3EkwP1sdwn6M,288
 azure/ai/evaluation/_evaluators/_service_groundedness/_service_groundedness.py,sha256=GPvufAgTnoQ2HYs6Xnnpmh23n5E3XxnUV0NGuwjDyU0,6648
 azure/ai/evaluation/_evaluators/_similarity/__init__.py,sha256=V2Mspog99_WBltxTkRHG5NpN5s9XoiTSN4I8POWEkLA,268
-azure/ai/evaluation/_evaluators/_similarity/_similarity.py,sha256=UVBIa1xIlOIJtPctCu-UCOWvXzE4ysaK_XFdokajCuA,5669
+azure/ai/evaluation/_evaluators/_similarity/_similarity.py,sha256=AeqJ_OJUAsdu9Cac4OLVPF2zbrBmpXD7_5oOs_cxKsk,4244
 azure/ai/evaluation/_evaluators/_similarity/similarity.prompty,sha256=eoludASychZoGL625bFCaZai-OY7DIAg90ZLax_o4XE,4594
 azure/ai/evaluation/_evaluators/_xpia/__init__.py,sha256=VMEL8WrpJQeh4sQiOLzP7hRFPnjzsvwfvTzaGCVJPCM,88
 azure/ai/evaluation/_evaluators/_xpia/xpia.py,sha256=Nv14lU7jN0yXKbHgHRXMHEy6pn1rXmesBOYI2Ge9ewk,5849
+azure/ai/evaluation/_safety_evaluation/__init__.py,sha256=Yx1Iq2GNKQ5lYxTotvPwkPL4u0cm6YVxUe-iVbu1clI,180
+azure/ai/evaluation/_safety_evaluation/_safety_evaluation.py,sha256=HtT6aKdZ4EP6Dx-JuExdysO_CBERFMwRWmGTWeHQ_NU,33890
 azure/ai/evaluation/_vendor/__init__.py,sha256=Yx1Iq2GNKQ5lYxTotvPwkPL4u0cm6YVxUe-iVbu1clI,180
 azure/ai/evaluation/_vendor/rouge_score/__init__.py,sha256=03OkyfS_UmzRnHv6-z9juTaJ6OXJoEJM989hgifIZbc,607
 azure/ai/evaluation/_vendor/rouge_score/rouge_scorer.py,sha256=DtNSeshHipzc6vFnvx7kbs5viXe4LNq-ZrgllFvfR4U,11299
@@ -92,15 +88,15 @@ azure/ai/evaluation/_vendor/rouge_score/tokenize.py,sha256=IyHVsWY6IFFZdB23cLiJs
 azure/ai/evaluation/_vendor/rouge_score/tokenizers.py,sha256=3_-y1TyvyluHuERhSJ5CdXSwnpcMA7aAKU6PCz9wH_Q,1745
 azure/ai/evaluation/simulator/__init__.py,sha256=JbrPZ8pvTBalyX94SvZ9btHNoovX8rbZV03KmzxxWys,552
 azure/ai/evaluation/simulator/_adversarial_scenario.py,sha256=9rpAPz594tYjxzM3XMeDq6CZSc2yvf5YaNaGC7nzYhM,1710
-azure/ai/evaluation/simulator/_adversarial_simulator.py,sha256=FPZ3OdpGuwCHDVoOZW-f_j7pyK71PfDN3JPh205tW0c,21706
+azure/ai/evaluation/simulator/_adversarial_simulator.py,sha256=2QEt5hHOKnNqMySt3SNsr-LY22DvFyusVnPkvAyur1I,22100
 azure/ai/evaluation/simulator/_constants.py,sha256=nCL7_1BnYh6k0XvxudxsDVMbiG9MMEvYw5wO9FZHHZ8,857
 azure/ai/evaluation/simulator/_direct_attack_simulator.py,sha256=FTtWf655dHJF5FLJi0xGSBgIlGWNiVWyqaLDJSud9XA,10199
 azure/ai/evaluation/simulator/_indirect_attack_simulator.py,sha256=nweIU_AkUIR50qLQpjmljf_OkpsCPth2Ebf4vusygCA,10226
-azure/ai/evaluation/simulator/_simulator.py,sha256=pWxVfy9ll6gmOyGEk6Ie7Y48X21wJ5DebqY8Re0SIOk,36213
+azure/ai/evaluation/simulator/_simulator.py,sha256=LBzez7qvObpVjTwmlGS_PfhDLo8pRknh5epra2yo9X8,36484
 azure/ai/evaluation/simulator/_tracing.py,sha256=frZ4-usrzINast9F4-ONRzEGGox71y8bYw0UHNufL1Y,3069
 azure/ai/evaluation/simulator/_utils.py,sha256=16NltlywpbMtoFtULwTKqeURguIS1kSKSo3g8uKV8TA,5181
-azure/ai/evaluation/simulator/_conversation/__init__.py,sha256=s8djzJ58_-CiIA8xHB-SbgeZaq1F7ftrc3qJbpUpUdg,17853
-azure/ai/evaluation/simulator/_conversation/_conversation.py,sha256=qdzGMtCPYMxeGpR91NZTEmmz2RtADTvQGj6C-3EUTw4,7402
+azure/ai/evaluation/simulator/_conversation/__init__.py,sha256=LOR5h7vSACrv_cmaS6r7KUqjJcHYg7PAgkDbdvOmd4g,17726
+azure/ai/evaluation/simulator/_conversation/_conversation.py,sha256=h8OHq0sWKiTH821tC5zF44CJ-QgutTgDnEYsFEA7Cw0,7635
 azure/ai/evaluation/simulator/_conversation/constants.py,sha256=3v7zkjPwJAPbSpJYIK6VOZZy70bJXMo_QTVqSFGlq9A,984
 azure/ai/evaluation/simulator/_data_sources/__init__.py,sha256=Yx1Iq2GNKQ5lYxTotvPwkPL4u0cm6YVxUe-iVbu1clI,180
 azure/ai/evaluation/simulator/_data_sources/grounding.json,sha256=jqdqHrCgS7hN7K2kXSEcPCmzFjV4cv_qcCSR-Hutwx4,1257075
@@ -109,15 +105,15 @@ azure/ai/evaluation/simulator/_helpers/_language_suffix_mapping.py,sha256=7BBLH7
 azure/ai/evaluation/simulator/_helpers/_simulator_data_classes.py,sha256=BOttMTec3muMiA4OzwD_iW08GTrhja7PL9XVjRCN3jM,3029
 azure/ai/evaluation/simulator/_model_tools/__init__.py,sha256=aMv5apb7uVjuhMF9ohhA5kQmo652hrGIJlhdl3y2R1I,835
 azure/ai/evaluation/simulator/_model_tools/_identity_manager.py,sha256=-hptp2vpJIcfjvtd0E2c7ry00LVh23LxuYGevsNFfgs,6385
-azure/ai/evaluation/simulator/_model_tools/_proxy_completion_model.py,sha256=Zg_SzqjCGJ3Wt8hktxz6Y1JEJCcV0V5jBC9N06jQP3k,8984
+azure/ai/evaluation/simulator/_model_tools/_proxy_completion_model.py,sha256=D2Am07l2gq_v-vztqDBTX2MxZTUumm5ysaVSL2L0rxQ,9118
 azure/ai/evaluation/simulator/_model_tools/_rai_client.py,sha256=40MGzIXGv7oVshWH7AbOPLCigI4HlMrqbF2Rq5jFMGo,8755
 azure/ai/evaluation/simulator/_model_tools/_template_handler.py,sha256=NQWqjE7csSzkhb2XdW82AoCA-DxixpTrfBxAnOt2Wlc,7075
 azure/ai/evaluation/simulator/_model_tools/models.py,sha256=bfVm0PV3vfH_8DkdmTMZqYVN-G51hZ6Y0TOO-NiysJY,21811
 azure/ai/evaluation/simulator/_prompty/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 azure/ai/evaluation/simulator/_prompty/task_query_response.prompty,sha256=2BzSqDDYilDushvR56vMRDmqFIaIYAewdUlUZg_elMg,2182
 azure/ai/evaluation/simulator/_prompty/task_simulate.prompty,sha256=NE6lH4bfmibgMn4NgJtm9_l3PMoHSFrfjjosDJEKM0g,939
-azure_ai_evaluation-1.1.0.dist-info/METADATA,sha256=zusuZTIcO7487bWQK5V7XZ-Pbqugm7HFzvcCV5yfTmk,28751
-azure_ai_evaluation-1.1.0.dist-info/NOTICE.txt,sha256=4tzi_Yq4-eBGhBvveobWHCgUIVF-ZeouGN0m7hVq5Mk,3592
-azure_ai_evaluation-1.1.0.dist-info/WHEEL,sha256=pL8R0wFFS65tNSRnaOVrsw9EOkOqxLrlUPenUYnJKNo,91
-azure_ai_evaluation-1.1.0.dist-info/top_level.txt,sha256=S7DhWV9m80TBzAhOFjxDUiNbKszzoThbnrSz5MpbHSQ,6
-azure_ai_evaluation-1.1.0.dist-info/RECORD,,
+azure_ai_evaluation-1.3.0.dist-info/METADATA,sha256=msxfA0EnNspmcqbsGuYtdEESA3kLPnEfWBpzf7yFHEw,32885
+azure_ai_evaluation-1.3.0.dist-info/NOTICE.txt,sha256=4tzi_Yq4-eBGhBvveobWHCgUIVF-ZeouGN0m7hVq5Mk,3592
+azure_ai_evaluation-1.3.0.dist-info/WHEEL,sha256=pL8R0wFFS65tNSRnaOVrsw9EOkOqxLrlUPenUYnJKNo,91
+azure_ai_evaluation-1.3.0.dist-info/top_level.txt,sha256=S7DhWV9m80TBzAhOFjxDUiNbKszzoThbnrSz5MpbHSQ,6
+azure_ai_evaluation-1.3.0.dist-info/RECORD,,

azure/ai/evaluation/_evaluators/_multimodal/__init__.py

@@ -1,20 +0,0 @@
-# ---------------------------------------------------------
-# Copyright (c) Microsoft Corporation. All rights reserved.
-# ---------------------------------------------------------
-from ._content_safety_multimodal import ContentSafetyMultimodalEvaluator
-from ._content_safety_multimodal_base import ContentSafetyMultimodalEvaluatorBase
-from ._hate_unfairness import HateUnfairnessMultimodalEvaluator
-from ._self_harm import SelfHarmMultimodalEvaluator
-from ._sexual import SexualMultimodalEvaluator
-from ._violence import ViolenceMultimodalEvaluator
-from ._protected_material import ProtectedMaterialMultimodalEvaluator
-
-__all__ = [
-    "ContentSafetyMultimodalEvaluator",
-    "ContentSafetyMultimodalEvaluatorBase",
-    "ViolenceMultimodalEvaluator",
-    "SexualMultimodalEvaluator",
-    "SelfHarmMultimodalEvaluator",
-    "HateUnfairnessMultimodalEvaluator",
-    "ProtectedMaterialMultimodalEvaluator",
-]