Will Feng 6bddfb9546 [FSDP2] Add cache for FSDP wrapper class (#134135) ... Currently, `fully_shard` will create a new `FSDPMyModuleClass` class for each `MyModuleClass` module **object**, which causes Dynamo to guard-fail on every module object's type checking. This PR fixes the issue by caching and reusing previously created FSDP wrapper class. Pull Request resolved: https://github.com/pytorch/pytorch/pull/134135 Approved by: https://github.com/awgu		2024-08-22 00:41:30 +00:00
..
fsdp	[FSDP2] Add cache for FSDP wrapper class (#134135)	2024-08-22 00:41:30 +00:00
__init__.py
checkpoint_activation.py	Add None return type to init (#132335)	2024-08-01 15:26:45 +00:00
contract.py	Add None return type to init (#132335)	2024-08-01 15:26:45 +00:00
fully_shard.py	[BE] mypy: disallow untyped decorators (#131428)	2024-07-23 21:50:55 +00:00
replicate.py	[DDP][FSDP2] keep DTensor params for replicate(fully_shard) (#133059)	2024-08-09 18:38:05 +00:00