unique_toolkit 1.12.0__py3-none-any.whl → 1.12.1__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- unique_toolkit/agentic/evaluation/evaluation_manager.py +9 -5
- unique_toolkit/agentic/evaluation/hallucination/hallucination_evaluation.py +9 -1
- {unique_toolkit-1.12.0.dist-info → unique_toolkit-1.12.1.dist-info}/METADATA +4 -1
- {unique_toolkit-1.12.0.dist-info → unique_toolkit-1.12.1.dist-info}/RECORD +6 -6
- {unique_toolkit-1.12.0.dist-info → unique_toolkit-1.12.1.dist-info}/LICENSE +0 -0
- {unique_toolkit-1.12.0.dist-info → unique_toolkit-1.12.1.dist-info}/WHEEL +0 -0
@@ -123,6 +123,15 @@ class EvaluationManager:
|
|
123
123
|
self._evaluation_passed = False
|
124
124
|
evaluation_results_unpacked.append(unpacked_evaluation_result)
|
125
125
|
|
126
|
+
for evaluation_name, evaluation_result in zip(
|
127
|
+
selected_evaluation_names, evaluation_results_unpacked
|
128
|
+
):
|
129
|
+
evaluation_instance = self.get_evaluation_by_name(evaluation_name)
|
130
|
+
if evaluation_instance:
|
131
|
+
await self._show_message_assessment(
|
132
|
+
evaluation_instance, evaluation_result, assistant_message_id
|
133
|
+
)
|
134
|
+
|
126
135
|
return evaluation_results_unpacked
|
127
136
|
|
128
137
|
async def execute_evaluation_call(
|
@@ -143,11 +152,6 @@ class EvaluationManager:
|
|
143
152
|
evaluation_metric_result: EvaluationMetricResult = (
|
144
153
|
await evaluation_instance.run(loop_response)
|
145
154
|
)
|
146
|
-
# show results to the user
|
147
|
-
await self._show_message_assessment(
|
148
|
-
evaluation_instance, evaluation_metric_result, assistant_message_id
|
149
|
-
)
|
150
|
-
|
151
155
|
return evaluation_metric_result
|
152
156
|
|
153
157
|
return EvaluationMetricResult(
|
@@ -78,11 +78,19 @@ class HallucinationEvaluation(Evaluation):
|
|
78
78
|
if not evaluation_result.error
|
79
79
|
else ChatMessageAssessmentStatus.ERROR
|
80
80
|
)
|
81
|
+
explanation = evaluation_result.reason
|
82
|
+
|
83
|
+
if status == ChatMessageAssessmentStatus.ERROR:
|
84
|
+
title = "Hallucination Check Error"
|
85
|
+
label = ChatMessageAssessmentLabel.RED
|
86
|
+
explanation = (
|
87
|
+
"An unrecoverable error occurred while evaluating the response."
|
88
|
+
)
|
81
89
|
|
82
90
|
return EvaluationAssessmentMessage(
|
83
91
|
status=status,
|
84
92
|
title=title,
|
85
|
-
explanation=
|
93
|
+
explanation=explanation,
|
86
94
|
label=label,
|
87
95
|
type=self.get_assessment_type(),
|
88
96
|
)
|
@@ -1,6 +1,6 @@
|
|
1
1
|
Metadata-Version: 2.1
|
2
2
|
Name: unique_toolkit
|
3
|
-
Version: 1.12.
|
3
|
+
Version: 1.12.1
|
4
4
|
Summary:
|
5
5
|
License: Proprietary
|
6
6
|
Author: Cedric Klinkert
|
@@ -118,6 +118,9 @@ All notable changes to this project will be documented in this file.
|
|
118
118
|
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
|
119
119
|
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
|
120
120
|
|
121
|
+
## [1.12.1] - 2025-10-07
|
122
|
+
- Fix bug where failed evaluations did not show an error to the user.
|
123
|
+
|
121
124
|
## [1.12.0] - 2026-10-07
|
122
125
|
- Add the `OpenAIUserMessageBuilder` for complex user messages with images
|
123
126
|
- More examples with documents/images on the chat
|
@@ -31,10 +31,10 @@ unique_toolkit/agentic/evaluation/config.py,sha256=Fer-y1aP8kmnPWXQydh12i9f_XU7K
|
|
31
31
|
unique_toolkit/agentic/evaluation/context_relevancy/prompts.py,sha256=EdHFUOB581yVxcOL8482KUv_LzaRjuiem71EF8udYMc,1331
|
32
32
|
unique_toolkit/agentic/evaluation/context_relevancy/schema.py,sha256=lZd0TPzH43ifgWWGg3WO6b1AQX8aK2R9y51yH0d1DHM,2919
|
33
33
|
unique_toolkit/agentic/evaluation/context_relevancy/service.py,sha256=fkPGq4Nnn5las1waYDICqHl6xC-rR5iOpT24YifGO20,9654
|
34
|
-
unique_toolkit/agentic/evaluation/evaluation_manager.py,sha256=
|
34
|
+
unique_toolkit/agentic/evaluation/evaluation_manager.py,sha256=IPx4BVUgkjFOP1BGLi0BlB6UujpXlZ0KGuSXDRemQhY,8143
|
35
35
|
unique_toolkit/agentic/evaluation/exception.py,sha256=7lcVbCyoN4Md1chNJDFxpUYyWbVrcr9dcc3TxWykJTc,115
|
36
36
|
unique_toolkit/agentic/evaluation/hallucination/constants.py,sha256=0HyvI5zu7JmjHLe9lKJSeAWMvfQfpmR6MLHJ4HPX1hc,2063
|
37
|
-
unique_toolkit/agentic/evaluation/hallucination/hallucination_evaluation.py,sha256=
|
37
|
+
unique_toolkit/agentic/evaluation/hallucination/hallucination_evaluation.py,sha256=yMcfA7iMNXkneNrFxJuoDIoB37mK8IRXEKnPsK_UDOk,3454
|
38
38
|
unique_toolkit/agentic/evaluation/hallucination/prompts.py,sha256=O3Hi_rOzZlujvnO2wn2jhoPmrYLjzVtRWwxn5Q81m9Y,3405
|
39
39
|
unique_toolkit/agentic/evaluation/hallucination/service.py,sha256=Ut-f768HY4E9zEhfMoKYnGTFRZVkxWGiSTGOpgfZWYM,2447
|
40
40
|
unique_toolkit/agentic/evaluation/hallucination/utils.py,sha256=QLsYvgAyQ5XnKEzn7ko7bXfzePD4De99TWnMKglMpds,8178
|
@@ -149,7 +149,7 @@ unique_toolkit/short_term_memory/schemas.py,sha256=OhfcXyF6ACdwIXW45sKzjtZX_gkcJ
|
|
149
149
|
unique_toolkit/short_term_memory/service.py,sha256=5PeVBu1ZCAfyDb2HLVvlmqSbyzBBuE9sI2o9Aajqjxg,8884
|
150
150
|
unique_toolkit/smart_rules/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
151
151
|
unique_toolkit/smart_rules/compile.py,sha256=Ozhh70qCn2yOzRWr9d8WmJeTo7AQurwd3tStgBMPFLA,1246
|
152
|
-
unique_toolkit-1.12.
|
153
|
-
unique_toolkit-1.12.
|
154
|
-
unique_toolkit-1.12.
|
155
|
-
unique_toolkit-1.12.
|
152
|
+
unique_toolkit-1.12.1.dist-info/LICENSE,sha256=GlN8wHNdh53xwOPg44URnwag6TEolCjoq3YD_KrWgss,193
|
153
|
+
unique_toolkit-1.12.1.dist-info/METADATA,sha256=mBeCg71Dak88SQUNJpaiQwTuh9MuKmA9vDPaakm3AQg,36092
|
154
|
+
unique_toolkit-1.12.1.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
|
155
|
+
unique_toolkit-1.12.1.dist-info/RECORD,,
|
File without changes
|
File without changes
|