Added more tests

src/scoring.py

@@ -48,6 +48,10 @@ def check_errors(goldstandard, predictions):
     if unknown_labels:
         errors.append(f"Unknown labels found in predictions: {set(unknown_labels)}")
 
+    unknown_keys = [key for key in predictions.keys() if key not in goldstandard.keys()]
+    if unknown_keys:
+        errors.append(f"Unknown keys found in predictions: {set(unknown_keys)}")
+
     if errors:
         print("Errors found in predictions:")
         for error in errors:

tests/test_check_errors.py

@@ -81,4 +81,86 @@ def mock_print(msg):  # Capture print statements
 
     builtins.print = original_print  # Restore original print
 
-    assert "Missing predictions for IDs:" in errors[1]  # Expect an error message
+    assert "Missing predictions for IDs:" in errors[1]  # Expect an error message
+
+def test_check_errors_DuplicatePrediction():
+    gold_df = pd.DataFrame({
+        'id': ['de_0', 'en_0', 'fr_0'],
+        'label': [0, 1, 0]
+    })
+    gold_df = gold_df.set_index('id')
+    predictions = pd.DataFrame({
+        'id': ['de_0', 'en_0', 'fr_0', 'fr_0'],
+        'predicted_label': [0, 1, 0, 1]
+    })
+    predictions = predictions.set_index('id')['predicted_label'].to_dict()
+    errors = []
+
+    def mock_print(msg):  # Capture print statements
+        errors.append(msg)
+
+    # Monkey-patch print to capture output
+    import builtins
+    original_print = builtins.print
+    builtins.print = mock_print
+
+    check_errors(gold_df, predictions)
+
+    builtins.print = original_print  # Restore original print
+
+    assert "Duplicate prediction entries for IDs:" in errors[1]  # Expect an error message
+
+def test_check_errors_UnknownLabels():
+    gold_df = pd.DataFrame({
+        'id': ['de_0', 'en_0', 'fr_0'],
+        'label': [0, 1, 0]
+    })
+    gold_df = gold_df.set_index('id')
+    predictions = pd.DataFrame({
+        'id': ['de_0', 'en_0', 'fr_0'],
+        'predicted_label': [0, 1, 2]
+    })
+    predictions = predictions.set_index('id')['predicted_label'].to_dict()
+    errors = []
+
+    def mock_print(msg):  # Capture print statements
+        errors.append(msg)
+
+    # Monkey-patch print to capture output
+    import builtins
+    original_print = builtins.print
+    builtins.print = mock_print
+
+    check_errors(gold_df, predictions)
+
+    builtins.print = original_print  # Restore original print
+
+    assert "Unknown labels found in predictions:" in errors[1]  # Expect error
+
+
+def test_check_errors_UnknownKeys():
+    gold_df = pd.DataFrame({
+        'id': ['de_0', 'en_0', 'fr_0'],
+        'label': [0, 1, 0]
+    })
+    gold_df = gold_df.set_index('id')
+    predictions = pd.DataFrame({
+        'id': ['de_0', 'en_0', 'fr_0', 'en_1'],
+        'predicted_label': [0, 1, 0, 1]
+    })
+    predictions = predictions.set_index('id')['predicted_label'].to_dict()
+    errors = []
+
+    def mock_print(msg):  # Capture print statements
+        errors.append(msg)
+
+    # Monkey-patch print to capture output
+    import builtins
+    original_print = builtins.print
+    builtins.print = mock_print
+
+    check_errors(gold_df, predictions)
+
+    builtins.print = original_print  # Restore original print
+
+    assert "Unknown keys found in predictions:" in errors[1]  # Expect error