RecursionError: maximum recursion depth exceeded

[Previous line repeated 962 more times] ！！！！

The following is the terminal error message

Traceback (most recent call last):
File "E:\anaconda3\envs\pytorch1.10.0\lib\site-packages\fairseq_cli\hydra_train.py", line 27, in hydra_main
_hydra_main(cfg)
File "E:\anaconda3\envs\pytorch1.10.0\lib\site-packages\fairseq_cli\hydra_train.py", line 56, in _hydra_main
distributed_utils.call_main(cfg, pre_main, **kwargs)
File "E:\anaconda3\envs\pytorch1.10.0\lib\site-packages\fairseq\distributed\utils.py", line 369, in call_main
main(cfg, **kwargs)
File "E:\anaconda3\envs\pytorch1.10.0\lib\site-packages\fairseq_cli\train.py", line 96, in main
model = task.build_model(cfg.model)
File "E:\anaconda3\envs\pytorch1.10.0\lib\site-packages\fairseq\tasks\fairseq_task.py", line 343, in build_model
model = models.build_model(cfg, self, from_checkpoint)
File "E:\anaconda3\envs\pytorch1.10.0\lib\site-packages\fairseq\models\__init__.py", line 106, in build_model
return model.build_model(cfg, task)
File "E:\Code\notebooks\data2vec\models\data2vec2.py", line 392, in build_model
return cls(cfg, modalities, task=task, skip_ema=cfg.skip_ema)
File "E:\Code\notebooks\data2vec\models\data2vec2.py", line 251, in __init__
self.ema = self.make_ema_teacher(cfg.ema_decay)
File "E:\anaconda3\envs\pytorch1.10.0\lib\site-packages\torch\autograd\grad_mode.py", line 28, in decorate_context
return func(*args, **kwargs)
File "E:\Code\notebooks\data2vec\models\data2vec2.py", line 302, in make_ema_teacher
return EMAModule(
File "E:\anaconda3\envs\pytorch1.10.0\lib\site-packages\fairseq\modules\ema_module.py", line 42, in __init__
self.model = copy.deepcopy(model)
File "E:\anaconda3\envs\pytorch1.10.0\lib\copy.py", line 172, in deepcopy
y = _reconstruct(x, memo, *rv)
File "E:\anaconda3\envs\pytorch1.10.0\lib\copy.py", line 270, in _reconstruct
state = deepcopy(state, memo)
File "E:\anaconda3\envs\pytorch1.10.0\lib\copy.py", line 146, in deepcopy
y = copier(x, memo)
File "E:\anaconda3\envs\pytorch1.10.0\lib\copy.py", line 230, in _deepcopy_dict
y[deepcopy(key, memo)] = deepcopy(value, memo)
File "E:\anaconda3\envs\pytorch1.10.0\lib\copy.py", line 172, in deepcopy
y = _reconstruct(x, memo, *rv)
File "E:\anaconda3\envs\pytorch1.10.0\lib\copy.py", line 270, in _reconstruct
state = deepcopy(state, memo)
File "E:\anaconda3\envs\pytorch1.10.0\lib\copy.py", line 146, in deepcopy
y = copier(x, memo)
File "E:\anaconda3\envs\pytorch1.10.0\lib\copy.py", line 230, in _deepcopy_dict
y[deepcopy(key, memo)] = deepcopy(value, memo)
File "E:\anaconda3\envs\pytorch1.10.0\lib\copy.py", line 172, in deepcopy
y = _reconstruct(x, memo, *rv)
File "E:\anaconda3\envs\pytorch1.10.0\lib\copy.py", line 271, in _reconstruct
if hasattr(y, '__setstate__'):
File "E:\anaconda3\envs\pytorch1.10.0\lib\site-packages\fairseq\tasks\fairseq_task.py", line 40, in __getattr__
if name not in self._state and name in self._factories:
File "E:\anaconda3\envs\pytorch1.10.0\lib\site-packages\fairseq\tasks\fairseq_task.py", line 40, in __getattr__
if name not in self._state and name in self._factories:
File "E:\anaconda3\envs\pytorch1.10.0\lib\site-packages\fairseq\tasks\fairseq_task.py", line 40, in __getattr__
if name not in self._state and name in self._factories:
[Previous line repeated 962 more times]
RecursionError: maximum recursion depth exceeded

Start with models/data2vec2.py, especially build_model, __init__, and make_ema_teacher, then inspect fairseq/modules/ema_module.py and fairseq/tasks/fairseq_task.py around the traceback. Reproduce model construction in the reported environment and determine why deepcopy repeatedly triggers task attribute lookup. Done means model initialization completes without the RecursionError and the relevant regression is covered by a reproducible test or documented validation.

Written by the indexing model from the issue text.