ColossalAI/colossalai/zero
HELSON e6d50ec107
[zero] adapt zero for unsharded parameters (#561)
* support existing sharded and unsharded parameters in zero

* add unitest for moe-zero model init

* polish moe gradient handler
2022-03-31 18:34:11 +08:00
..
init_ctx [zero] adapt zero for unsharded parameters (#561) 2022-03-31 18:34:11 +08:00
shard_utils [zero] non model data tracing (#545) 2022-03-29 15:45:48 +08:00
sharded_model [zero] adapt zero for unsharded parameters (#561) 2022-03-31 18:34:11 +08:00
sharded_optim [zero] trace states of fp16/32 grad and fp32 param (#571) 2022-03-31 16:26:54 +08:00
sharded_param [zero] trace states of fp16/32 grad and fp32 param (#571) 2022-03-31 16:26:54 +08:00
__init__.py [refactor] remove old zero code (#517) 2022-03-25 14:54:39 +08:00