azure-ai-evaluation 1.0.0b1__py3-none-any.whl → 1.0.0b2__py3-none-any.whl

azure/ai/evaluation/_evaluate/_telemetry/__init__.py

@@ -6,21 +6,24 @@ import functools
 import inspect
 import json
 import logging
-from typing import Callable, Dict
+from typing import Callable, Dict, TypeVar
 
 import pandas as pd
-
 from promptflow._sdk.entities._flows import FlexFlow as flex_flow
 from promptflow._sdk.entities._flows import Prompty as prompty_sdk
 from promptflow._sdk.entities._flows.dag import Flow as dag_flow
 from promptflow.client import PFClient
 from promptflow.core import Prompty as prompty_core
+from typing_extensions import ParamSpec
 
 from ..._user_agent import USER_AGENT
 from .._utils import _trace_destination_from_project_scope
 
 LOGGER = logging.getLogger(__name__)
 
+P = ParamSpec("P")
+R = TypeVar("R")
+
 
 def _get_evaluator_type(evaluator: Dict[str, Callable]):
     """
@@ -82,7 +85,7 @@ def _get_evaluator_properties(evaluator, evaluator_name):
             name = str(evaluator)
             pf_type = "Unknown"
     except Exception as e:  # pylint: disable=broad-exception-caught
-        LOGGER.debug(f"Failed to get evaluator properties: {e}")
+        LOGGER.debug("Failed to get evaluator properties: %s", e)
         name = str(evaluator)
         pf_type = "Unknown"
 
@@ -95,15 +98,17 @@ def _get_evaluator_properties(evaluator, evaluator_name):
 
 
 # cspell:ignore isna
-def log_evaluate_activity(func) -> None:
+def log_evaluate_activity(func: Callable[P, R]) -> Callable[P, R]:
     """Decorator to log evaluate activity
 
     :param func: The function to be decorated
     :type func: Callable
+    :returns: The decorated function
+    :rtype: Callable[P, R]
     """
 
     @functools.wraps(func)
-    def wrapper(*args, **kwargs) -> Callable:
+    def wrapper(*args: P.args, **kwargs: P.kwargs) -> R:
         from promptflow._sdk._telemetry import ActivityType, log_activity
         from promptflow._sdk._telemetry.telemetry import get_telemetry_logger
 
@@ -119,7 +124,7 @@ def log_evaluate_activity(func) -> None:
             user_agent=USER_AGENT,
         )
 
-        track_in_cloud = bool(pf_client._config.get_trace_destination())
+        track_in_cloud = bool(pf_client._config.get_trace_destination())  # pylint: disable=protected-access
         evaluate_target = bool(kwargs.get("target", None))
         evaluator_config = bool(kwargs.get("evaluator_config", None))
         custom_dimensions = {
@@ -154,7 +159,7 @@ def log_evaluate_activity(func) -> None:
                         evaluator_info["failed_rows"] = failed_rows
                         evaluator_info["total_rows"] = total_rows
                     except Exception as e:  # pylint: disable=broad-exception-caught
-                        LOGGER.debug(f"Failed to collect evaluate failed row info for {evaluator_name}: {e}")
+                        LOGGER.debug("Failed to collect evaluate failed row info for %s: %s", evaluator_name, e)
                     evaluators_info.append(evaluator_info)
 
                 custom_dimensions = {"evaluators_info": json.dumps(evaluators_info)}
@@ -167,7 +172,7 @@ def log_evaluate_activity(func) -> None:
                 ):
                     pass
             except Exception as e:  # pylint: disable=broad-exception-caught
-                LOGGER.debug(f"Failed to collect evaluate usage info: {e}")
+                LOGGER.debug("Failed to collect evaluate usage info: %s", e)
 
             return result
 

azure/ai/evaluation/_evaluate/_utils.py

@@ -8,12 +8,13 @@ import re
 import tempfile
 from collections import namedtuple
 from pathlib import Path
+from typing import Dict
 
 import pandas as pd
 
-from azure.ai.evaluation._constants import DEFAULT_EVALUATION_RESULTS_FILE_NAME, Prefixes
-from azure.ai.evaluation._exceptions import EvaluationException, ErrorBlame, ErrorCategory, ErrorTarget
+from azure.ai.evaluation._constants import DEFAULT_EVALUATION_RESULTS_FILE_NAME, DefaultOpenEncoding, Prefixes
 from azure.ai.evaluation._evaluate._eval_run import EvalRun
+from azure.ai.evaluation._exceptions import ErrorBlame, ErrorCategory, ErrorTarget, EvaluationException
 
 LOGGER = logging.getLogger(__name__)
 
@@ -33,16 +34,16 @@ def extract_workspace_triad_from_trace_provider(trace_provider: str):  # pylint:
     match = re.match(AZURE_WORKSPACE_REGEX_FORMAT, trace_provider)
     if not match or len(match.groups()) != 5:
         raise EvaluationException(
-                message="Malformed trace provider string, expected azureml://subscriptions/<subscription_id>/"
-                "resourceGroups/<resource_group>/providers/Microsoft.MachineLearningServices/"
-                f"workspaces/<workspace_name>, got {trace_provider}",
-                internal_message="Malformed trace provider string, expected azureml://subscriptions/<subscription_id>/"
-                "resourceGroups/<resource_group>/providers/Microsoft.MachineLearningServices/"
-                "workspaces/<workspace_name>,",
-                target=ErrorTarget.UNKNOWN,
-                category=ErrorCategory.INVALID_VALUE,
-                blame=ErrorBlame.UNKNOWN,
-            )
+            message="Malformed trace provider string, expected azureml://subscriptions/<subscription_id>/"
+            "resourceGroups/<resource_group>/providers/Microsoft.MachineLearningServices/"
+            f"workspaces/<workspace_name>, got {trace_provider}",
+            internal_message="Malformed trace provider string, expected azureml://subscriptions/<subscription_id>/"
+            "resourceGroups/<resource_group>/providers/Microsoft.MachineLearningServices/"
+            "workspaces/<workspace_name>,",
+            target=ErrorTarget.UNKNOWN,
+            category=ErrorCategory.INVALID_VALUE,
+            blame=ErrorBlame.UNKNOWN,
+        )
     subscription_id = match.group(1)
     resource_group_name = match.group(3)
     workspace_name = match.group(5)
@@ -50,7 +51,7 @@ def extract_workspace_triad_from_trace_provider(trace_provider: str):  # pylint:
 
 
 def load_jsonl(path):
-    with open(path, "r", encoding="utf-8") as f:
+    with open(path, "r", encoding=DefaultOpenEncoding.READ) as f:
         return [json.loads(line) for line in f.readlines()]
 
 
@@ -99,7 +100,7 @@ def _log_metrics_and_instance_results(
         with tempfile.TemporaryDirectory() as tmpdir:
             tmp_path = os.path.join(tmpdir, artifact_name)
 
-            with open(tmp_path, "w", encoding="utf-8") as f:
+            with open(tmp_path, "w", encoding=DefaultOpenEncoding.WRITE) as f:
                 f.write(instance_results.to_json(orient="records", lines=True))
 
             ev_run.log_artifact(tmpdir, artifact_name)
@@ -155,11 +156,13 @@ def _write_output(path, data_dict):
     if os.path.isdir(path):
         p = p / DEFAULT_EVALUATION_RESULTS_FILE_NAME
 
-    with open(p, "w") as f:
+    with open(p, "w", encoding=DefaultOpenEncoding.WRITE) as f:
         json.dump(data_dict, f)
 
 
-def _apply_column_mapping(source_df: pd.DataFrame, mapping_config: dict, inplace: bool = False) -> pd.DataFrame:
+def _apply_column_mapping(
+    source_df: pd.DataFrame, mapping_config: Dict[str, str], inplace: bool = False
+) -> pd.DataFrame:
     """
     Apply column mapping to source_df based on mapping_config.
 
@@ -167,10 +170,11 @@ def _apply_column_mapping(source_df: pd.DataFrame, mapping_config: dict, inplace
     :param source_df: the data frame to be changed.
     :type source_df: pd.DataFrame
     :param mapping_config: The configuration, containing column mapping.
-    :type mapping_config: dict.
+    :type mapping_config: Dict[str, str].
     :param inplace: If true, the source_df will be changed inplace.
     :type inplace: bool
     :return: The modified data frame.
+    :rtype: pd.DataFrame
     """
     result_df = source_df
 
@@ -211,19 +215,22 @@ def _has_aggregator(evaluator):
     return hasattr(evaluator, "__aggregate__")
 
 
-def get_int_env_var(env_var_name, default_value=None):
+def get_int_env_var(env_var_name: str, default_value: int) -> int:
     """
-    The function `get_int_env_var` retrieves an integer environment variable value, with an optional
+    The function `get_int_env_var` retrieves an integer environment variable value, with a
     default value if the variable is not set or cannot be converted to an integer.
 
     :param env_var_name: The name of the environment variable you want to retrieve the value of
+    :type env_var_name: str
     :param default_value: The default value is the value that will be returned if the environment
-    variable is not found or if it cannot be converted to an integer
+        variable is not found or if it cannot be converted to an integer
+    :type default_value: int
     :return: an integer value.
+    :rtype: int
     """
     try:
-        return int(os.environ.get(env_var_name, default_value))
-    except Exception:
+        return int(os.environ[env_var_name])
+    except (ValueError, KeyError):
         return default_value
 
 

azure/ai/evaluation/_evaluators/_bleu/_bleu.py

@@ -2,8 +2,8 @@
 # Copyright (c) Microsoft Corporation. All rights reserved.
 # ---------------------------------------------------------
 from nltk.translate.bleu_score import SmoothingFunction, sentence_bleu
-
 from promptflow._utils.async_utils import async_run_allowing_running_loop
+
 from azure.ai.evaluation._common.utils import nltk_tokenize
 
 

azure/ai/evaluation/_evaluators/_chat/_chat.py

@@ -4,19 +4,18 @@
 import json
 import logging
 from concurrent.futures import as_completed
-from typing import Dict, List, Union
+from typing import Dict, List
 
 import numpy as np
-
 from promptflow.tracing import ThreadPoolExecutorWithContext as ThreadPoolExecutor
 
+from azure.ai.evaluation._exceptions import ErrorBlame, ErrorCategory, ErrorTarget, EvaluationException
+
 from .._coherence import CoherenceEvaluator
 from .._fluency import FluencyEvaluator
 from .._groundedness import GroundednessEvaluator
-from ..._model_configurations import AzureOpenAIModelConfiguration, OpenAIModelConfiguration
 from .._relevance import RelevanceEvaluator
 from .retrieval import RetrievalChatEvaluator
-from azure.ai.evaluation._exceptions import EvaluationException, ErrorBlame, ErrorCategory, ErrorTarget
 
 logger = logging.getLogger(__name__)
 
@@ -203,7 +202,7 @@ class ChatEvaluator:
             return score
         except Exception as e:  # pylint: disable=broad-exception-caught
             logger.warning(
-                f"Evaluator {evaluator.__class__.__name__} failed for turn {turn_num + 1} with exception: {e}"
+                "Evaluator %s failed for turn %s with exception: %s", evaluator.__class__.__name__, turn_num + 1, e
             )
             return {}
 
@@ -264,7 +263,10 @@ class ChatEvaluator:
                 )
 
             if "role" not in turn or "content" not in turn:
-                msg = f"Each turn in 'conversation' must have 'role' and 'content' keys. Turn number: {one_based_turn_num}"
+                msg = (
+                    "Each turn in 'conversation' must have 'role' and 'content' keys. "
+                    + f"Turn number: {one_based_turn_num}"
+                )
                 raise EvaluationException(
                     message=msg,
                     internal_message=msg,
@@ -272,7 +274,7 @@ class ChatEvaluator:
                     category=ErrorCategory.INVALID_VALUE,
                     blame=ErrorBlame.USER_ERROR,
                 )
-            
+
             if turn["role"] != expected_role:
                 msg = f"Expected role {expected_role} but got {turn['role']}. Turn number: {one_based_turn_num}"
                 raise EvaluationException(
@@ -305,7 +307,9 @@ class ChatEvaluator:
                     )
 
                 if "citations" not in turn["context"]:
-                    msg = f"Context in each assistant's turn must have 'citations' key. Turn number: {one_based_turn_num}"
+                    msg = (
+                        f"Context in each assistant's turn must have 'citations' key. Turn number: {one_based_turn_num}"
+                    )
                     raise EvaluationException(
                         message=msg,
                         internal_message=msg,
@@ -326,7 +330,10 @@ class ChatEvaluator:
 
                 for citation_num, citation in enumerate(turn["context"]["citations"]):
                     if not isinstance(citation, dict):
-                        msg = f"Each citation in 'citations' must be a dictionary. Turn number: {one_based_turn_num}, Citation number: {citation_num + 1}"
+                        msg = (
+                            "Each citation in 'citations' must be a dictionary. "
+                            + f"Turn number: {one_based_turn_num}, Citation number: {citation_num + 1}"
+                        )
                         raise EvaluationException(
                             message=msg,
                             internal_message=msg,

azure/ai/evaluation/_evaluators/_chat/retrieval/_retrieval.py

@@ -6,18 +6,12 @@ import json
 import logging
 import os
 import re
-from typing import Union
 
 import numpy as np
-
 from promptflow._utils.async_utils import async_run_allowing_running_loop
 from promptflow.core import AsyncPrompty
 
-from ...._model_configurations import AzureOpenAIModelConfiguration, OpenAIModelConfiguration
-from ...._common.utils import (
-    check_and_add_api_version_for_aoai_model_config,
-    check_and_add_user_agent_for_aoai_model_config,
-)
+from ...._common.utils import ensure_api_version_in_aoai_model_config, ensure_user_agent_in_aoai_model_config
 
 logger = logging.getLogger(__name__)
 
@@ -34,7 +28,7 @@ class _AsyncRetrievalChatEvaluator:
     DEFAULT_OPEN_API_VERSION = "2024-02-15-preview"
 
     def __init__(self, model_config: dict):
-        check_and_add_api_version_for_aoai_model_config(model_config, self.DEFAULT_OPEN_API_VERSION)
+        ensure_api_version_in_aoai_model_config(model_config, self.DEFAULT_OPEN_API_VERSION)
 
         prompty_model_config = {"configuration": model_config, "parameters": {"extra_headers": {}}}
 
@@ -42,7 +36,7 @@ class _AsyncRetrievalChatEvaluator:
         # https://github.com/encode/httpx/discussions/2959
         prompty_model_config["parameters"]["extra_headers"].update({"Connection": "close"})
 
-        check_and_add_user_agent_for_aoai_model_config(
+        ensure_user_agent_in_aoai_model_config(
             model_config,
             prompty_model_config,
             USER_AGENT,
@@ -92,7 +86,7 @@ class _AsyncRetrievalChatEvaluator:
 
             except Exception as e:  # pylint: disable=broad-exception-caught
                 logger.warning(
-                    f"Evaluator {self.__class__.__name__} failed for turn {turn_num + 1} with exception: {e}"
+                    "Evaluator %s failed for turn %s with exception: %s", self.__class__.__name__, turn_num + 1, e
                 )
 
                 per_turn_scores.append(np.nan)

azure/ai/evaluation/_evaluators/_coherence/_coherence.py

@@ -4,19 +4,14 @@
 
 import os
 import re
-from typing import Union
 
 import numpy as np
-
 from promptflow._utils.async_utils import async_run_allowing_running_loop
-from azure.ai.evaluation._exceptions import EvaluationException, ErrorBlame, ErrorCategory, ErrorTarget
 from promptflow.core import AsyncPrompty
 
-from ..._model_configurations import AzureOpenAIModelConfiguration, OpenAIModelConfiguration
-from ..._common.utils import (
-    check_and_add_api_version_for_aoai_model_config,
-    check_and_add_user_agent_for_aoai_model_config,
-)
+from azure.ai.evaluation._exceptions import ErrorBlame, ErrorCategory, ErrorTarget, EvaluationException
+
+from ..._common.utils import ensure_api_version_in_aoai_model_config, ensure_user_agent_in_aoai_model_config
 
 try:
     from ..._user_agent import USER_AGENT
@@ -31,7 +26,7 @@ class _AsyncCoherenceEvaluator:
     DEFAULT_OPEN_API_VERSION = "2024-02-15-preview"
 
     def __init__(self, model_config: dict):
-        check_and_add_api_version_for_aoai_model_config(model_config, self.DEFAULT_OPEN_API_VERSION)
+        ensure_api_version_in_aoai_model_config(model_config, self.DEFAULT_OPEN_API_VERSION)
 
         prompty_model_config = {"configuration": model_config, "parameters": {"extra_headers": {}}}
 
@@ -39,7 +34,7 @@ class _AsyncCoherenceEvaluator:
         # https://github.com/encode/httpx/discussions/2959
         prompty_model_config["parameters"]["extra_headers"].update({"Connection": "close"})
 
-        check_and_add_user_agent_for_aoai_model_config(
+        ensure_user_agent_in_aoai_model_config(
             model_config,
             prompty_model_config,
             USER_AGENT,

azure/ai/evaluation/_evaluators/_content_safety/_content_safety.py

@@ -5,8 +5,6 @@ from concurrent.futures import as_completed
 
 from promptflow.tracing import ThreadPoolExecutorWithContext as ThreadPoolExecutor
 
-from azure.ai.evaluation._model_configurations import AzureAIProject
-
 try:
     from ._hate_unfairness import HateUnfairnessEvaluator
     from ._self_harm import SelfHarmEvaluator

azure/ai/evaluation/_evaluators/_content_safety/_content_safety_base.py

@@ -6,8 +6,7 @@ from abc import ABC
 
 from azure.ai.evaluation._common.constants import EvaluationMetrics
 from azure.ai.evaluation._common.rai_service import evaluate_with_rai_service
-from azure.ai.evaluation._exceptions import EvaluationException, ErrorBlame, ErrorCategory, ErrorTarget
-from azure.ai.evaluation._model_configurations import AzureAIProject
+from azure.ai.evaluation._exceptions import ErrorBlame, ErrorCategory, ErrorTarget, EvaluationException
 
 
 class ContentSafetyEvaluatorBase(ABC):

azure/ai/evaluation/_evaluators/_content_safety/_content_safety_chat.py

@@ -7,9 +7,8 @@ from typing import Dict, List
 
 import numpy as np
 from promptflow.tracing import ThreadPoolExecutorWithContext as ThreadPoolExecutor
-from azure.ai.evaluation._exceptions import EvaluationException, ErrorBlame, ErrorCategory, ErrorTarget
 
-from azure.ai.evaluation._model_configurations import AzureAIProject 
+from azure.ai.evaluation._exceptions import ErrorBlame, ErrorCategory, ErrorTarget, EvaluationException
 
 try:
     from ._hate_unfairness import HateUnfairnessEvaluator
@@ -165,7 +164,10 @@ class ContentSafetyChatEvaluator:
             return score
         except Exception as e:  # pylint: disable=broad-exception-caught
             logger.warning(
-                f"Evaluator {evaluator.__class__.__name__} failed for turn {turn_num + 1} with exception: {e}"
+                "Evaluator %s failed for turn %s with exception: %s",
+                evaluator.__class__.__name__,
+                turn_num + 1,
+                e,
             )
             return {}
 
@@ -235,7 +237,10 @@ class ContentSafetyChatEvaluator:
                 )
 
             if "role" not in turn or "content" not in turn:
-                msg = f"Each turn in 'conversation' must have 'role' and 'content' keys. Turn number: {one_based_turn_num}"
+                msg = (
+                    "Each turn in 'conversation' must have 'role' and 'content' keys. "
+                    + f"Turn number: {one_based_turn_num}"
+                )
                 raise EvaluationException(
                     message=msg,
                     internal_message=msg,

azure/ai/evaluation/_evaluators/_content_safety/_hate_unfairness.py

@@ -2,8 +2,8 @@
 # Copyright (c) Microsoft Corporation. All rights reserved.
 # ---------------------------------------------------------
 from promptflow._utils.async_utils import async_run_allowing_running_loop
+
 from azure.ai.evaluation._common.constants import EvaluationMetrics
-from azure.ai.evaluation._model_configurations import AzureAIProject
 
 try:
     from ._content_safety_base import ContentSafetyEvaluatorBase

azure/ai/evaluation/_evaluators/_content_safety/_self_harm.py

@@ -2,8 +2,8 @@
 # Copyright (c) Microsoft Corporation. All rights reserved.
 # ---------------------------------------------------------
 from promptflow._utils.async_utils import async_run_allowing_running_loop
+
 from azure.ai.evaluation._common.constants import EvaluationMetrics
-from azure.ai.evaluation._model_configurations import AzureAIProject
 
 try:
     from ._content_safety_base import ContentSafetyEvaluatorBase

azure/ai/evaluation/_evaluators/_content_safety/_sexual.py

@@ -2,8 +2,8 @@
 # Copyright (c) Microsoft Corporation. All rights reserved.
 # ---------------------------------------------------------
 from promptflow._utils.async_utils import async_run_allowing_running_loop
+
 from azure.ai.evaluation._common.constants import EvaluationMetrics
-from azure.ai.evaluation._model_configurations import AzureAIProject
 
 try:
     from ._content_safety_base import ContentSafetyEvaluatorBase

azure/ai/evaluation/_evaluators/_content_safety/_violence.py

@@ -2,8 +2,8 @@
 # Copyright (c) Microsoft Corporation. All rights reserved.
 # ---------------------------------------------------------
 from promptflow._utils.async_utils import async_run_allowing_running_loop
+
 from azure.ai.evaluation._common.constants import EvaluationMetrics
-from azure.ai.evaluation._model_configurations import AzureAIProject
 
 try:
     from ._content_safety_base import ContentSafetyEvaluatorBase

azure/ai/evaluation/_evaluators/_eci/_eci.py

@@ -2,10 +2,10 @@
 # Copyright (c) Microsoft Corporation. All rights reserved.
 # ---------------------------------------------------------
 from promptflow._utils.async_utils import async_run_allowing_running_loop
+
 from azure.ai.evaluation._common.constants import _InternalEvaluationMetrics
 from azure.ai.evaluation._common.rai_service import evaluate_with_rai_service
-from azure.ai.evaluation._exceptions import EvaluationException, ErrorBlame, ErrorCategory, ErrorTarget
-from azure.ai.evaluation._model_configurations import AzureAIProject
+from azure.ai.evaluation._exceptions import ErrorBlame, ErrorCategory, ErrorTarget, EvaluationException
 
 
 class _AsyncECIEvaluator:

azure/ai/evaluation/_evaluators/_f1_score/_f1_score.py

@@ -6,7 +6,8 @@ from collections import Counter
 from typing import List
 
 from promptflow._utils.async_utils import async_run_allowing_running_loop
-from azure.ai.evaluation._exceptions import EvaluationException, ErrorBlame, ErrorCategory, ErrorTarget
+
+from azure.ai.evaluation._exceptions import ErrorBlame, ErrorCategory, ErrorTarget, EvaluationException
 
 
 class _AsyncF1ScoreEvaluator:

azure/ai/evaluation/_evaluators/_fluency/_fluency.py

@@ -4,19 +4,14 @@
 
 import os
 import re
-from typing import Union
 
 import numpy as np
-
 from promptflow._utils.async_utils import async_run_allowing_running_loop
-from azure.ai.evaluation._exceptions import EvaluationException, ErrorBlame, ErrorCategory, ErrorTarget
 from promptflow.core import AsyncPrompty
 
-from ..._model_configurations import AzureOpenAIModelConfiguration, OpenAIModelConfiguration
-from ..._common.utils import (
-    check_and_add_api_version_for_aoai_model_config,
-    check_and_add_user_agent_for_aoai_model_config,
-)
+from azure.ai.evaluation._exceptions import ErrorBlame, ErrorCategory, ErrorTarget, EvaluationException
+
+from ..._common.utils import ensure_api_version_in_aoai_model_config, ensure_user_agent_in_aoai_model_config
 
 try:
     from ..._user_agent import USER_AGENT
@@ -31,7 +26,7 @@ class _AsyncFluencyEvaluator:
     DEFAULT_OPEN_API_VERSION = "2024-02-15-preview"
 
     def __init__(self, model_config: dict):
-        check_and_add_api_version_for_aoai_model_config(model_config, self.DEFAULT_OPEN_API_VERSION)
+        ensure_api_version_in_aoai_model_config(model_config, self.DEFAULT_OPEN_API_VERSION)
 
         prompty_model_config = {"configuration": model_config, "parameters": {"extra_headers": {}}}
 
@@ -39,7 +34,7 @@ class _AsyncFluencyEvaluator:
         # https://github.com/encode/httpx/discussions/2959
         prompty_model_config["parameters"]["extra_headers"].update({"Connection": "close"})
 
-        check_and_add_user_agent_for_aoai_model_config(
+        ensure_user_agent_in_aoai_model_config(
             model_config,
             prompty_model_config,
             USER_AGENT,

azure/ai/evaluation/_evaluators/_gleu/_gleu.py

@@ -2,8 +2,8 @@
 # Copyright (c) Microsoft Corporation. All rights reserved.
 # ---------------------------------------------------------
 from nltk.translate.gleu_score import sentence_gleu
-
 from promptflow._utils.async_utils import async_run_allowing_running_loop
+
 from azure.ai.evaluation._common.utils import nltk_tokenize
 
 

azure/ai/evaluation/_evaluators/_groundedness/_groundedness.py

@@ -4,19 +4,14 @@
 
 import os
 import re
-from typing import Union
 
 import numpy as np
-
 from promptflow._utils.async_utils import async_run_allowing_running_loop
-from azure.ai.evaluation._exceptions import EvaluationException, ErrorBlame, ErrorCategory, ErrorTarget
 from promptflow.core import AsyncPrompty
 
-from ..._model_configurations import AzureOpenAIModelConfiguration, OpenAIModelConfiguration
-from ..._common.utils import (
-    check_and_add_api_version_for_aoai_model_config,
-    check_and_add_user_agent_for_aoai_model_config,
-)
+from azure.ai.evaluation._exceptions import ErrorBlame, ErrorCategory, ErrorTarget, EvaluationException
+
+from ..._common.utils import ensure_api_version_in_aoai_model_config, ensure_user_agent_in_aoai_model_config
 
 try:
     from ..._user_agent import USER_AGENT
@@ -31,7 +26,7 @@ class _AsyncGroundednessEvaluator:
     DEFAULT_OPEN_API_VERSION = "2024-02-15-preview"
 
     def __init__(self, model_config: dict):
-        check_and_add_api_version_for_aoai_model_config(model_config, self.DEFAULT_OPEN_API_VERSION)
+        ensure_api_version_in_aoai_model_config(model_config, self.DEFAULT_OPEN_API_VERSION)
 
         prompty_model_config = {"configuration": model_config, "parameters": {"extra_headers": {}}}
 
@@ -39,7 +34,7 @@ class _AsyncGroundednessEvaluator:
         # https://github.com/encode/httpx/discussions/2959
         prompty_model_config["parameters"]["extra_headers"].update({"Connection": "close"})
 
-        check_and_add_user_agent_for_aoai_model_config(
+        ensure_user_agent_in_aoai_model_config(
             model_config,
             prompty_model_config,
             USER_AGENT,

azure/ai/evaluation/_evaluators/_meteor/_meteor.py

@@ -4,6 +4,7 @@
 import nltk
 from nltk.translate.meteor_score import meteor_score
 from promptflow._utils.async_utils import async_run_allowing_running_loop
+
 from azure.ai.evaluation._common.utils import nltk_tokenize
 
 

azure/ai/evaluation/_evaluators/_protected_material/_protected_material.py

@@ -2,10 +2,10 @@
 # Copyright (c) Microsoft Corporation. All rights reserved.
 # ---------------------------------------------------------
 from promptflow._utils.async_utils import async_run_allowing_running_loop
+
 from azure.ai.evaluation._common.constants import EvaluationMetrics
 from azure.ai.evaluation._common.rai_service import evaluate_with_rai_service
-from azure.ai.evaluation._exceptions import EvaluationException, ErrorBlame, ErrorCategory, ErrorTarget
-from azure.ai.evaluation._model_configurations import AzureAIProject
+from azure.ai.evaluation._exceptions import ErrorBlame, ErrorCategory, ErrorTarget, EvaluationException
 
 
 class _AsyncProtectedMaterialEvaluator:

azure/ai/evaluation/_evaluators/_protected_materials/_protected_materials.py

@@ -2,10 +2,10 @@
 # Copyright (c) Microsoft Corporation. All rights reserved.
 # ---------------------------------------------------------
 from promptflow._utils.async_utils import async_run_allowing_running_loop
+
 from azure.ai.evaluation._common.constants import EvaluationMetrics
 from azure.ai.evaluation._common.rai_service import evaluate_with_rai_service
-from azure.ai.evaluation._exceptions import EvaluationException, ErrorBlame, ErrorCategory, ErrorTarget
-from azure.ai.evaluation._model_configurations import AzureAIProject
+from azure.ai.evaluation._exceptions import ErrorBlame, ErrorCategory, ErrorTarget, EvaluationException
 
 
 class _AsyncProtectedMaterialsEvaluator: