pre-commit format

lm_eval/api/task.py

@@ -71,12 +71,12 @@ class GroupConfig(dict):
     group_alias: Optional[str] = None
     task: Optional[Union[str, list]] = None
     tag_to_task: Optional[str] = False
-    aggregate_metric: Optional[
-        Union[List[AggMetricConfig], AggMetricConfig, dict]
-    ] = None
-    metadata: Optional[
-        dict
-    ] = None  # by default, not used in the code. allows for users to pass arbitrary info to tasks
+    aggregate_metric: Optional[Union[List[AggMetricConfig], AggMetricConfig, dict]] = (
+        None
+    )
+    metadata: Optional[dict] = (
+        None  # by default, not used in the code. allows for users to pass arbitrary info to tasks
+    )
 
     def __getitem__(self, item):
         return getattr(self, item)

lm_eval/evaluator.py

@@ -688,10 +688,10 @@ def evaluate(
                                     if "N/A" in stderrs:
                                         results[group_or_task][stderr] = "N/A"
                                     else:
-                                        results[group_or_task][
-                                            stderr
-                                        ] = lm_eval.api.metrics.pooled_sample_stderr(
-                                            stderrs, sizes
+                                        results[group_or_task][stderr] = (
+                                            lm_eval.api.metrics.pooled_sample_stderr(
+                                                stderrs, sizes
+                                            )
                                         )
                                         # TODO: allow GroupConfigs to choose which variance formula is used, for back-compatibility
                                         # To use the old (likely incorrect) variance formula, comment out the above and uncomment this line:

lm_eval/evaluator_utils.py

@@ -344,9 +344,9 @@ def consolidate_results(
                 metric_key
             ]
             results[task_output.task_id]["samples"] = task_output.sample_len
-            results[task_output.task_id][
-                f"{metric}_stderr,{filter_key}"
-            ] = task_output.agg_metrics[f"{metric}_stderr,{filter_key}"]
+            results[task_output.task_id][f"{metric}_stderr,{filter_key}"] = (
+                task_output.agg_metrics[f"{metric}_stderr,{filter_key}"]
+            )
     return results, samples, configs, versions, num_fewshot, higher_is_better