window server multi-GPU don't working

when i run train_network.py by multi GPU, it show :

NOTE: Redirects are currently not supported in Windows or MacOs.
[E C:\actions-runner\_work\pytorch\pytorch\builder\windows\pytorch\torch\csrc\distributed\c10d\socket.cpp:793] [c10d] The client socket has timed out after 900s while trying to connect to (127.0.0.1, 29500).

the accelerate config is:

How many different machines will you use (use more than 1 for multi-node training)? [1]: 1
Do you wish to optimize your script with torch dynamo?[yes/NO]:no
Do you want to use DeepSpeed? [yes/NO]: no
Do you want to use FullyShardedDataParallel? [yes/NO]: no
Do you want to use Megatron-LM ? [yes/NO]: no
How many GPU(s) should be used for distributed training? [1]:2
What GPU(s) (by id) should be used for training on this machine as a comma-seperated list? [all]:[1,2]

How can i solve it ?? help help help.

Start with train_network.py and the provided Accelerate configuration, then reproduce the two-GPU run on Windows using the reported settings. Investigate the 127.0.0.1:29500 socket timeout and confirm that multi-GPU training starts without the distributed connection failure.

Written by the indexing model from the issue text.