how to use learner.distributed(), in self supervised pretrain code ？

[lilei]

how to use self supervised pretrain code to train on multiple GPUS or multi-node ?

I want to user revised the code for multinode or multiple GPUS for large dataset and large parameters.

But not successed