HELSON d7ecaf362b [zero] fix init bugs in zero context (#686) ...

* adapt model weight initialization for methods in Pytorch nn.init

2022-04-07 17:38:45 +08:00

..

init_ctx

[zero] fix init bugs in zero context (#686)

2022-04-07 17:38:45 +08:00

shard_utils

[zero] initialize a stateful tensor manager (#614)

2022-04-06 16:18:49 +08:00

sharded_model

polish docstring of zero (#612)

2022-04-01 14:50:56 +08:00

sharded_optim

[hotfix] fix bugs for unsharded parameters when restore data (#664)

2022-04-03 22:02:11 +08:00

sharded_param

[zero] initialize a stateful tensor manager (#614)

2022-04-06 16:18:49 +08:00

__init__.py

[refactor] remove old zero code (#517)

2022-03-25 14:54:39 +08:00