clean up diff

lm_eval/api/task.py

@@ -137,7 +137,6 @@ class ConfigurableGroup(abc.ABC):
         config: Optional[dict] = None,
     ) -> None:
         self._config = GroupConfig(**config)
-        # self._task_id = self._config.group
 
     @property
     def group(self):
@@ -155,14 +154,6 @@ class ConfigurableGroup(abc.ABC):
     def config(self):
         return self._config.to_dict()
 
-    # @property
-    # def task_id(self) -> Any:
-    #     return self._task_id
-
-    # @task_id.setter
-    # def task_id(self, value):
-    #     self._task_id = value
-
     @property
     def group_name(self) -> Any:
         return self._config.group
@@ -360,8 +351,6 @@ class Task(abc.ABC):
         self._fewshot_docs: Optional[list] = None
         self._instances: Optional[List[Instance]] = None
 
-        # Create a unique identifier ID
-        # self._task_id = shortuuid.uuid()[:8]
         self._config: TaskConfig = TaskConfig({**config}) if config else TaskConfig()
 
         self._filters = [build_filter_ensemble("none", [["take_first", None]])]
@@ -818,14 +807,6 @@ class Task(abc.ABC):
         )
         return doc_iterator
 
-    # @property
-    # def task_id(self) -> Any:
-    #     return self._task_id
-
-    # @task_id.setter
-    # def task_id(self, value):
-    #     self._task_id = value
-
 
 class ConfigurableTask(Task):
     VERSION = "Yaml"
@@ -839,9 +820,6 @@ class ConfigurableTask(Task):
         download_mode=None,
         config: Optional[dict] = None,
     ) -> None:  # TODO no super() call here
-        # Create a unique identifier ID
-        # self._task_id = shortuuid.uuid()[:8]
-
         # Get pre-configured attributes
         self._config = self.CONFIG
 

lm_eval/evaluator.py

@@ -41,7 +41,7 @@ from lm_eval.utils import (
 
 if TYPE_CHECKING:
     from lm_eval.api.model import LM
-    from lm_eval.tasks import Task
+    from lm_eval.api.task import Task
 
 
 @positional_deprecated

lm_eval/evaluator_utils.py

@@ -44,7 +44,6 @@ class TaskOutput:
         self,
         task=None,
         task_name=None,
-        # task_id=None,
         task_config=None,
         version=None,
         group_name=None,
@@ -56,7 +55,6 @@ class TaskOutput:
         self.task = task
         self.task_config = task_config
         self.task_name = task_name
-        # self.task_id = task_id
         self.group_name = group_name
         self.version = version
         self.n_shot = n_shot
@@ -82,7 +80,6 @@ class TaskOutput:
                 task=task, task_name=task_name, is_group=is_group, group_name=group_name
             )
         version = task.VERSION
-        # task_id = task.task_id
         task_config = dict(task.dump_config())
         if (n_shot := task_config.get("num_fewshot")) == 0:
             n_shot = task_config.get("metadata", {}).get("num_fewshot", 0)
@@ -91,7 +88,6 @@ class TaskOutput:
         return cls(
             task=task,
             task_name=task_name,
-            # task_id=task_id,
             task_config=task_config,
             group_name=group_name,
             version=version,
@@ -321,7 +317,6 @@ def consolidate_results(
     higher_is_better = collections.defaultdict(dict)
 
     for task_output in eval_tasks:
-        # results[task_output.task_id]["task"] = task_output.task_name
         if "task_alias" in (task_config := task_output.task_config):
             results[task_output.task_name]["alias"] = task_config["task_alias"]
         else: