训练问题

[xings-sdnu]

Traceback (most recent call last):
File "/home/sx/nanodet-main/tools/train.py", line 155, in
main(args)
File "/home/sx/nanodet-main/tools/train.py", line 150, in main
trainer.fit(task, train_dataloader, val_dataloader, ckpt_path=model_resume_path)
File "/home/sx/.conda/envs/nanodet/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 545, in fit
call._call_and_handle_interrupt(
File "/home/sx/.conda/envs/nanodet/lib/python3.8/site-packages/pytorch_lightning/trainer/call.py", line 43, in _call_and_handle_interrupt
return trainer.strategy.launcher.launch(trainer_fn, *args, trainer=trainer, **kwargs)
File "/home/sx/.conda/envs/nanodet/lib/python3.8/site-packages/pytorch_lightning/strategies/launchers/subprocess_script.py", line 102, in launch
return function(*args, **kwargs)
File "/home/sx/.conda/envs/nanodet/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 581, in _fit_impl
self._run(model, ckpt_path=ckpt_path)
File "/home/sx/.conda/envs/nanodet/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 938, in _run
_verify_loop_configurations(self)
File "/home/sx/.conda/envs/nanodet/lib/python3.8/site-packages/pytorch_lightning/trainer/configuration_validator.py", line 37, in _verify_loop_configurations
__verify_train_val_loop_configuration(trainer, model)
File "/home/sx/.conda/envs/nanodet/lib/python3.8/site-packages/pytorch_lightning/trainer/configuration_validator.py", line 81, in __verify_train_val_loop_configuration
raise NotImplementedError(
NotImplementedError: Support for training_epoch_end has been removed in v2.0.0. TrainingTask implements this method. You can use the on_train_epoch_end hook instead. To access outputs, save them in-memory as instance attributes. You can find migration examples in Lightning-AI/pytorch-lightning#16520.

请问作者这是怎么回事？应该怎么修改?谢谢！