azure-ai-evaluation 1.11.0__tar.gz → 1.11.1__tar.gz

{azure_ai_evaluation-1.11.0 → azure_ai_evaluation-1.11.1}/CHANGELOG.md

@@ -1,5 +1,10 @@
 # Release History
 
+## 1.11.1 (2025-09-17)
+
+### Bugs Fixed
+- Pinning duckdb version to 1.3.2 for redteam extra to fix error `TypeError: unhashable type: '_duckdb.typing.DuckDBPyType'`
+
 ## 1.11.0 (2025-09-02)
 
 ### Features Added

{azure_ai_evaluation-1.11.0/azure_ai_evaluation.egg-info → azure_ai_evaluation-1.11.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: azure-ai-evaluation
-Version: 1.11.0
+Version: 1.11.1
 Summary: Microsoft Azure Evaluation Library for Python
 Home-page: https://github.com/Azure/azure-sdk-for-python
 Author: Microsoft Corporation
@@ -28,13 +28,14 @@ Requires-Dist: nltk>=3.9.1
 Requires-Dist: azure-storage-blob>=12.10.0
 Requires-Dist: httpx>=0.25.1
 Requires-Dist: pandas<3.0.0,>=2.1.2
-Requires-Dist: openai>=1.78.0
+Requires-Dist: openai>=1.108.0
 Requires-Dist: ruamel.yaml<1.0.0,>=0.17.10
 Requires-Dist: msrest>=0.6.21
 Requires-Dist: Jinja2>=3.1.6
 Requires-Dist: aiohttp>=3.0
 Provides-Extra: redteam
 Requires-Dist: pyrit==0.8.1; extra == "redteam"
+Requires-Dist: duckdb==1.3.2; extra == "redteam"
 Dynamic: author
 Dynamic: author-email
 Dynamic: classifier
@@ -412,6 +413,11 @@ This project has adopted the [Microsoft Open Source Code of Conduct][code_of_con
 
 # Release History
 
+## 1.11.1 (2025-09-17)
+
+### Bugs Fixed
+- Pinning duckdb version to 1.3.2 for redteam extra to fix error `TypeError: unhashable type: '_duckdb.typing.DuckDBPyType'`
+
 ## 1.11.0 (2025-09-02)
 
 ### Features Added

{azure_ai_evaluation-1.11.0 → azure_ai_evaluation-1.11.1}/azure/ai/evaluation/_evaluate/_evaluate_aoai.py

@@ -272,8 +272,33 @@ def _get_single_run_results(
     for row_result in all_results:
         listed_results["index"].append(row_result.datasource_item_id)
         for single_grader_row_result in row_result.results:
-            grader_name = run_info["grader_name_map"][single_grader_row_result["name"]]
-            for name, value in single_grader_row_result.items():
+            if isinstance(single_grader_row_result, dict):
+                result_dict = single_grader_row_result
+            elif hasattr(single_grader_row_result, "model_dump"):
+                result_dict = single_grader_row_result.model_dump()
+            elif hasattr(single_grader_row_result, "dict"):
+                result_dict = single_grader_row_result.dict()
+            elif hasattr(single_grader_row_result, "__dict__"):
+                result_dict = vars(single_grader_row_result)
+            else:
+                raise EvaluationException(
+                    message=("Unsupported AOAI evaluation result type: " f"{type(single_grader_row_result)!r}."),
+                    blame=ErrorBlame.UNKNOWN,
+                    category=ErrorCategory.FAILED_EXECUTION,
+                    target=ErrorTarget.AOAI_GRADER,
+                )
+
+            grader_result_name = result_dict.get("name", None)
+            if grader_result_name is None:
+                raise EvaluationException(
+                    message="AOAI evaluation response missing grader result name; unable to map to original grader.",
+                    blame=ErrorBlame.UNKNOWN,
+                    category=ErrorCategory.FAILED_EXECUTION,
+                    target=ErrorTarget.AOAI_GRADER,
+                )
+
+            grader_name = run_info["grader_name_map"][grader_result_name]
+            for name, value in result_dict.items():
                 if name in ["name"]:
                     continue
                 if name.lower() == "passed":

{azure_ai_evaluation-1.11.0 → azure_ai_evaluation-1.11.1}/azure/ai/evaluation/_version.py

@@ -3,4 +3,4 @@
 # ---------------------------------------------------------
 # represents upcoming version
 
-VERSION = "1.11.0"
+VERSION = "1.11.1"

{azure_ai_evaluation-1.11.0 → azure_ai_evaluation-1.11.1/azure_ai_evaluation.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: azure-ai-evaluation
-Version: 1.11.0
+Version: 1.11.1
 Summary: Microsoft Azure Evaluation Library for Python
 Home-page: https://github.com/Azure/azure-sdk-for-python
 Author: Microsoft Corporation
@@ -28,13 +28,14 @@ Requires-Dist: nltk>=3.9.1
 Requires-Dist: azure-storage-blob>=12.10.0
 Requires-Dist: httpx>=0.25.1
 Requires-Dist: pandas<3.0.0,>=2.1.2
-Requires-Dist: openai>=1.78.0
+Requires-Dist: openai>=1.108.0
 Requires-Dist: ruamel.yaml<1.0.0,>=0.17.10
 Requires-Dist: msrest>=0.6.21
 Requires-Dist: Jinja2>=3.1.6
 Requires-Dist: aiohttp>=3.0
 Provides-Extra: redteam
 Requires-Dist: pyrit==0.8.1; extra == "redteam"
+Requires-Dist: duckdb==1.3.2; extra == "redteam"
 Dynamic: author
 Dynamic: author-email
 Dynamic: classifier
@@ -412,6 +413,11 @@ This project has adopted the [Microsoft Open Source Code of Conduct][code_of_con
 
 # Release History
 
+## 1.11.1 (2025-09-17)
+
+### Bugs Fixed
+- Pinning duckdb version to 1.3.2 for redteam extra to fix error `TypeError: unhashable type: '_duckdb.typing.DuckDBPyType'`
+
 ## 1.11.0 (2025-09-02)
 
 ### Features Added

{azure_ai_evaluation-1.11.0 → azure_ai_evaluation-1.11.1}/azure_ai_evaluation.egg-info/requires.txt

@@ -5,7 +5,7 @@ nltk>=3.9.1
 azure-storage-blob>=12.10.0
 httpx>=0.25.1
 pandas<3.0.0,>=2.1.2
-openai>=1.78.0
+openai>=1.108.0
 ruamel.yaml<1.0.0,>=0.17.10
 msrest>=0.6.21
 Jinja2>=3.1.6
@@ -13,3 +13,4 @@ aiohttp>=3.0
 
 [redteam]
 pyrit==0.8.1
+duckdb==1.3.2

{azure_ai_evaluation-1.11.0 → azure_ai_evaluation-1.11.1}/setup.py

@@ -75,13 +75,13 @@ setup(
         "httpx>=0.25.1",
         # Dependencies added since Promptflow will soon be made optional
         "pandas>=2.1.2,<3.0.0",
-        "openai>=1.78.0",
+        "openai>=1.108.0",
         "ruamel.yaml>=0.17.10,<1.0.0",
         "msrest>=0.6.21",
         "Jinja2>=3.1.6",
         "aiohttp>=3.0",
     ],
-    extras_require={"redteam": ["pyrit==0.8.1"]},
+    extras_require={"redteam": ["pyrit==0.8.1", "duckdb==1.3.2"]},
     project_urls={
         "Bug Reports": "https://github.com/Azure/azure-sdk-for-python/issues",
         "Source": "https://github.com/Azure/azure-sdk-for-python",

{azure_ai_evaluation-1.11.0 → azure_ai_evaluation-1.11.1}/tests/unittests/test_aoai_score_model_grader.py

@@ -20,6 +20,22 @@ from azure.ai.evaluation._evaluate._evaluate_aoai import (
 )
 
 
+def _sampling_params_as_dict(value):
+    """Normalize sampling params to a plain dictionary for assertions."""
+
+    if value is None:
+        return {}
+    if isinstance(value, dict):
+        return value
+    if hasattr(value, "model_dump"):
+        return value.model_dump(exclude_none=True)
+    if hasattr(value, "dict"):
+        return value.dict(exclude_none=True)
+    if hasattr(value, "__dict__"):
+        return {k: v for k, v in vars(value).items() if v is not None and not k.startswith("_")}
+    return value
+
+
 @pytest.fixture
 def mock_aoai_model_config():
     """Mock Azure OpenAI model configuration for testing."""
@@ -200,8 +216,9 @@ class TestAzureOpenAIScoreModelGrader:
         assert config.input[0].role == "system"
         assert config.input[1].role == "user"
         assert config.range == [0.0, 1.0]
-        assert config.sampling_params["temperature"] == 0.0
-        assert config.sampling_params["max_tokens"] == 100
+        sampling_params = _sampling_params_as_dict(config.sampling_params)
+        assert sampling_params["temperature"] == 0.0
+        assert sampling_params["max_tokens"] == 100
         assert grader.pass_threshold == 0.5
 
     def test_different_score_ranges(self, mock_aoai_model_config):
@@ -534,7 +551,7 @@ class TestAzureOpenAIScoreModelGraderEdgeCases:
             sampling_params=complex_params,
         )
 
-        assert grader._grader_config.sampling_params == complex_params
+        assert _sampling_params_as_dict(grader._grader_config.sampling_params) == complex_params
 
     def test_grader_with_unicode_content(self, mock_aoai_model_config):
         """Test grader with Unicode and special characters in content."""