§ קg…ãó@—ddlZddlmZGd„dej¦«ZdS)éNcóP—eZdZdZejdejdejfd„¦«ZdS)ÚSharderap This is an interface which allows user to create more advanced sharding strategies that are not easily be composed by the `ShardingSpec`. :class:`torch.distributed._shard.sharding_plan.ShardingPlan` could take an object of the `Sharder` and call `shard` to shard the module, then replace the original module with sharded module returned. ÚmoduleÚreturncó—dS)aq Shard a module base on the implementation of this method, and return the sharded version of the module. Args: module (:class:`torch.nn.Module`): The module to apply sharding to. Returns: A :class:`torch.nn.Module` object that represents a module that's already been sharded. N©)Úselfrs ú\/var/www/html/ai-engine/env/lib/python3.11/site-packages/torch/distributed/_shard/sharder.pyÚshardz Sharder.shards€€€óN) Ú__name__Ú __module__Ú __qualname__Ú__doc__ÚabcÚabstractmethodÚnnÚModuler rr r rrsT€€€€€ððð Ôð  ˜BœIð  ¨"¬)ð  ð  ð  ñÔð  ð  ð  r r)rÚtorch.nnrÚABCrrr r úrsVðØ € € € àÐÐÐÐÐð ð ð ð ð ˆcŒgñ ô ð ð ð r