🐛 Bug

在Pytorch Lightning的AimLogger适配器中调用finalize()后，日志记录器无法重新连接到Aim运行。例如，在以下训练循环中发生这种情况：

trainer = Trainer(logger=aim_logger, ...)
trainer.fit(...)
trainer.test(...)

完整的回溯信息

File "/home/cas/Documents/repositories/Yggdrasil/yggdrasil/apps/train.py", line 172, in train_cli
    aim_logger.experiment.add_tag("Interrupted")
  File "/home/cas/local/conda/envs/yggdrasil/lib/python3.9/site-packages/lightning_fabric/loggers/logger.py", line 117, in experiment
    return get_experiment() or _DummyExperiment()
  File "/home/cas/local/conda/envs/yggdrasil/lib/python3.9/site-packages/lightning_utilities/core/rank_zero.py", line 27, in wrapped_fn
    return fn(*args, **kwargs)
  File "/home/cas/local/conda/envs/yggdrasil/lib/python3.9/site-packages/lightning_fabric/loggers/logger.py", line 115, in get_experiment
    return fn(self)
  File "/home/cas/local/conda/envs/yggdrasil/lib/python3.9/site-packages/aim/sdk/adapters/pytorch_lightning.py", line 80, in experiment
    self._run = Run(
  File "/home/cas/local/conda/envs/yggdrasil/lib/python3.9/site-packages/aim/ext/exception_resistant.py", line 70, in wrapper
    _SafeModeConfig.exception_callback(e, func)
  File "/home/cas/local/conda/envs/yggdrasil/lib/python3.9/site-packages/aim/ext/exception_resistant.py", line 47, in reraise_exception
    raise e
  File "/home/cas/local/conda/envs/yggdrasil/lib/python3.9/site-packages/aim/ext/exception_resistant.py", line 68, in wrapper
    return func(*args, **kwargs)
  File "/home/cas/local/conda/envs/yggdrasil/lib/python3.9/site-packages/aim/sdk/run.py", line 828, in __init__
    super().__init__(run_hash, repo=repo, read_only=read_only, experiment=experiment, force_resume=force_resume)
  File "/home/cas/local/conda/envs/yggdrasil/lib/python3.9/site-packages/aim/sdk/run.py", line 276, in __init__
    super().__init__(run_hash, repo=repo, read_only=read_only, force_resume=force_resume)
  File "/home/cas/local/conda/envs/yggdrasil/lib/python3.9/site-packages/aim/sdk/base_run.py", line 50, in __init__
    self._lock.lock(force=force_resume)
  File "/home/cas/local/conda/envs/yggdrasil/lib/python3.9/site-packages/aim/storage/lock_proxy.py", line 38, in lock
    return self._rpc_client.run_instruction(self._hash, self._handler, 'lock', (force,))
  File "/home/cas/local/conda/envs/yggdrasil/lib/python3.9/site-packages/aim/ext/transport/client.py", line 258, in run_instruction
    return self._run_read_instructions(queue_id, resource, method, args)
  File "/home/cas/local/conda/envs/yggdrasil/lib/python3.9/site-packages/aim/ext/transport/client.py", line 283, in _run_read_instructions
    raise_exception(status_msg.header.exception)
  File "/home/cas/local/conda/envs/yggdrasil/lib/python3.9/site-packages/aim/ext/transport/message_utils.py", line 76, in raise_exception
    raise exception(*args) if args else exception()
TypeError: __init__() missing 1 required positional argument: 'lock_file'

重现问题

aim_repo = ...  # Remote repo
logger = AimLogger(experiment="test", repo=aim_repo)

logger.log_metrics({"A": 1, "B": 2}, step=0)
logger.log_metrics({"A": 10, "B": 20}, step=1)

logger.finalize()

logger.experiment.add_tag("working_or_not")  # raises the TypeError

预期行为

访问.experiment属性应该重新连接到远程Aim运行。

环境

• Aim版本(例如，3.0.1):3.17.3
• Python版本：3.9.16
• pip版本：23.0.1
• OS(例如，Linux):Linux

其他上下文信息

通过以下猴子补丁，异常不会被引发。具体来说：设置force_resume=True。