Update: modify 24 file(s)

Author: vawsgit

examples/scripts/aggregate_metrics_demo.py

@@ -212,7 +212,7 @@ def demonstrate_deprecation_warning():
         warnings.simplefilter("always")
 
         # This will trigger a deprecation warning
-        field = ComparableField(aggregate=True, comparator=ExactComparator())
+        ComparableField(aggregate=True, comparator=ExactComparator())
 
         if w:
             print(f"   Warning: {w[0].message}")

examples/scripts/bulk_evaluation_demo.py

@@ -249,7 +249,7 @@ def demo_evaluation_with_output():
         evaluator.update(gt_doc, pred_doc, doc_id)
 
     # Get results and save metrics
-    result = evaluator.compute()
+    evaluator.compute()
     evaluator.save_metrics(metrics_file)
 
     print("\n💾 Output Files Created:")
@@ -341,7 +341,7 @@ def main():
     print("=" * 60)
 
     # Demo 1: Basic bulk evaluation
-    evaluator = demo_basic_bulk_evaluation()
+    demo_basic_bulk_evaluation()
 
     # Demo 2: Batch processing
     demo_batch_processing()

src/stickler/comparators/__init__.py

@@ -16,15 +16,19 @@
 
 # Import BERTComparator if evaluate is available
 try:
-    from stickler.comparators.bert import BERTComparator
+    from stickler.comparators.bert import BERTComparator  # noqa: F401
 
     BERT_AVAILABLE = True
 except ImportError:
     BERT_AVAILABLE = False
 
 # Import FuzzyComparator and Fuzz alias only if rapidfuzz is available
 try:
-    from stickler.comparators.fuzzy import RAPIDFUZZ_AVAILABLE, Fuzz, FuzzyComparator
+    from stickler.comparators.fuzzy import (  # noqa: F401
+        RAPIDFUZZ_AVAILABLE,
+        Fuzz,
+        FuzzyComparator,
+    )
 except ImportError:
     RAPIDFUZZ_AVAILABLE = False
 

src/stickler/reporting/html/__init__.py

@@ -9,5 +9,7 @@
 
 __all__ = [
     "EvaluationHTMLReporter",
-    "ReportConfig"
+    "ReportConfig",
+    "SectionGenerator",
+    "VisualizationEngine"
 ]

src/stickler/structured_object_evaluator/models/configuration_helper.py

@@ -5,9 +5,14 @@
 """
 
 import inspect
-from typing import Any, Dict, Union, get_args, get_origin
+from typing import TYPE_CHECKING, Any, Dict, Union, get_args, get_origin
 
 from stickler.comparators.levenshtein import LevenshteinComparator
+
+if TYPE_CHECKING:
+    from stickler.structured_object_evaluator.models.comparison_info import (
+        ComparableFieldConfig,
+    )
 from stickler.comparators.structured import StructuredModelComparator
 
 

tests/common/comparators/test_llm.py

@@ -167,7 +167,7 @@ def test_compare_with_custom_prompt(self, mock_bedrock):
         mock_bedrock.return_value = mock_client
 
         custom_prompt = "Custom prompt {value1} vs {value2}"
-        comparator = LLMComparator(
+        LLMComparator(
             model_name="test-model", prompt_template=custom_prompt
         )
 
@@ -380,7 +380,7 @@ def test_get_comparison_details_error_handling(self):
 
         assert "error" in details
         assert "comparison_result" in details
-        assert details["comparison_result"] == False
+        assert not details["comparison_result"]
 
     def test_string_representation(self):
         """Test string representations for serialization."""
@@ -477,20 +477,20 @@ def test_get_comparison_details_comprehensive_error_handling(self):
         self.mock_agent.side_effect = NoCredentialsError()
         details = self.comparator.get_comparison_details("value1", "value2")
         assert "error" in details
-        assert details["comparison_result"] == False
+        assert not details["comparison_result"]
 
         # Test ClientError
         error_response = {'Error': {'Code': 'ThrottlingException', 'Message': 'Rate exceeded'}}
         self.mock_agent.side_effect = ClientError(error_response, 'InvokeModel')
         details = self.comparator.get_comparison_details("value1", "value2")
         assert "error" in details
-        assert details["comparison_result"] == False
+        assert not details["comparison_result"]
 
         # Test generic exception
         self.mock_agent.side_effect = Exception("Generic error")
         details = self.comparator.get_comparison_details("value1", "value2")
         assert "error" in details
-        assert details["comparison_result"] == False
+        assert not details["comparison_result"]
 
     def test_model_initialization_error(self):
         """Test error handling during model initialization."""

tests/structured_object_evaluator/test_aggregate_contact_issue.py

@@ -80,7 +80,6 @@ def test_contact_object_level_metrics_not_rollup():
     expected_contact_fa = 0  # No false alarms (both sides have 1 contact)
     expected_contact_fp = 1  # fp = fa + fd = 0 + 1 = 1
     expected_contact_tp = 0  # No true positives
-    expected_contact_tn = 0  # No true negatives
     expected_contact_fn = 0  # No false negatives
 
     # CURRENT BUG: These will fail because it's rolling up nested field metrics (fd=1, fa=1, fp=2)

tests/structured_object_evaluator/test_comparable_field_fix.py

@@ -242,7 +242,7 @@ class TestModel(StructuredModel):
         assert isinstance(config.comparator, CustomTestComparator)
         assert config.threshold == 0.75
         assert config.weight == 2.5
-        assert config.clip_under_threshold == False
+        assert not config.clip_under_threshold
 
     def test_multiple_custom_comparators(self):
         """Test that multiple different custom comparators can coexist."""

tests/structured_object_evaluator/test_comparators.py

@@ -15,7 +15,7 @@
     FUZZY_AVAILABLE = False
 
 try:
-    from stickler.comparators.semantic import SemanticComparator
+    from stickler.comparators.semantic import SemanticComparator  # noqa: F401
 
     SEMANTIC_AVAILABLE = True
 except ImportError:

tests/structured_object_evaluator/test_confusion_matrix_metrics.py

@@ -356,7 +356,7 @@ def test_nested_structured_models():
 
     # Create test nested models
     details1 = SimpleModel(name="Details 1", count=1, description="First details")
-    details2 = SimpleModel(name="Details 2", count=2, description=None)
+    SimpleModel(name="Details 2", count=2, description=None)
 
     # Similar but not exact nested model
     details1_similar = SimpleModel(name="Details 1", count=1, description="First")