ver217 fce9432f08 sync before creating empty grad

2022-03-16 14:24:09 +08:00

..

init_ctx

use double buffer to handle grad

2022-03-16 14:24:09 +08:00

shard_utils

polish code

2022-03-14 15:48:55 +08:00

sharded_model

sync before creating empty grad

2022-03-16 14:24:09 +08:00

sharded_optim

[zero] refactory ShardedOptimV2 init method (#416)

2022-03-15 10:45:55 +08:00

sharded_param

use double buffer to handle grad

2022-03-16 14:24:09 +08:00

__init__.py

added buffer sync to naive amp model wrapper (#291)

2022-03-11 15:50:28 +08:00