chore/update documentation and dev dependencies

langtest/langtest.py

@@ -1539,7 +1539,7 @@ def __single_dataset_generate(self, dataset: list):
                         setattr(sample, "expected_results", v(sample.original))
                         for sample in m_data
                     ]
-                    (testcases[k]) = TestFactory.transform(
+                    testcases[k] = TestFactory.transform(
                         self.task, dataset, tests, m_data=m_data
                     )
 
@@ -1556,7 +1556,7 @@ def __single_dataset_generate(self, dataset: list):
                     testcases = DataFactory.filter_curated_bias(tests_to_filter, dataset)
                     if len(tests.keys()) > 2:
                         tests = {k: v for k, v in tests.items() if k != "bias"}
-                        (other_testcases) = TestFactory.transform(
+                        other_testcases = TestFactory.transform(
                             self.task, dataset, tests, m_data=m_data
                         )
                         testcases.extend(other_testcases)

langtest/metrics/prometheus_eval.py

@@ -222,8 +222,7 @@ def get_prompt(self) -> str:
             The prompt for the model.
         """
         s, f = self.get_score_rubric()
-        prompt = dedent(
-            """
+        prompt = dedent("""
         ###Task Description:
         An instruction (might include an Input inside it), a response to evaluate, a reference answer that gets from {formatted_criteria_keys}, and a score rubric representing a evaluation criteria are given.
         1. Write a detailed feedback that assess the quality of the response strictly based on the given score rubric, not evaluating in general.
@@ -244,8 +243,7 @@ def get_prompt(self) -> str:
         {score_rubric}
 
         ###Feedback:
-        """
-        )
+        """)
         return prompt.format(
             instruction=self.instruction,
             response=self.response,
@@ -302,8 +300,7 @@ def get_prompt(self) -> str:
             The prompt for the model.
         """
         s, f = self.get_score_rubric()
-        prompt = dedent(
-            """
+        prompt = dedent("""
         ###Task Description:
         An instruction (might include an Input inside it), a response to evaluate, and a score rubric representing a evaluation criteria are given.
         1. Write a detailed feedback that assess the quality of two responses strictly based on the given score rubric, not evaluating in general.
@@ -327,8 +324,7 @@ def get_prompt(self) -> str:
         {score_rubric}
 
         ###Feedback:
-        """
-        )
+        """)
         return prompt.format(
             instruction=self.instruction,
             response_a=self.response_a,

langtest/transform/clinical.py

@@ -958,8 +958,7 @@ def transform(sample_list: List[Sample], *args, **kwargs):
             else:
                 from textwrap import dedent
 
-                error_message = dedent(
-                    """
+                error_message = dedent("""
                     Attack model information is not provided in Configuration. Please provide the attack model information.
                     {
                         "medfuzz": {
@@ -970,8 +969,7 @@ def transform(sample_list: List[Sample], *args, **kwargs):
                             }
                         }
                     }
-                """
-                ).strip()
+                """).strip()
 
                 raise ValueError(error_message)
 

langtest/transform/robustness.py

@@ -34,7 +34,6 @@
 from ..utils.number_to_word import ConvertNumberToWord
 from collections import defaultdict
 
-
 inverted_ocr_typo_dict = defaultdict(list)
 for k, v in ocr_typo_dict.items():
     inverted_ocr_typo_dict[v].append(k)

langtest/transform/utils.py

@@ -266,7 +266,7 @@ def get_ethnicity_representation_dict(data: List[Sample]) -> Dict[str, int]:
 
     @staticmethod
     def get_entity_representation_proportions(
-        entity_representation: Dict[str, int]
+        entity_representation: Dict[str, int],
     ) -> Dict[str, float]:
         """Calculates the proportions of each entity in the representation.
 

langtest/utils/config_utils.py

@@ -3,7 +3,6 @@
 # from pkg_resources import resource_filename
 from importlib import resources
 
-
 LLM_DEFAULTS_CONFIG = {
     "azure-openai": str(
         resources.files("langtest").joinpath(

pyproject.toml

@@ -116,7 +116,7 @@ ipdb = "^0.13.13"
 pytest = "^7.4.0"
 taskipy = "^1.11.0"
 en-core-web-sm = { url = "https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.8.0/en_core_web_sm-3.8.0.tar.gz" }
-black = "^24.10.0"
+black = ">=26.3.1"
 pyproject-flake8 = "^7.0.0"
 pre-commit = "^3.3.3"
 pydocstyle = "^6.3.0"

tests/test_nlp_tasks.py

@@ -3,7 +3,6 @@
 from langtest import Harness
 from langtest.utils.custom_types import Sample
 
-
 task_configurations = [
     {
         "task": "ner",