oslo
oslo copied to clipboard
Integration ZeroDDP and ShardedModelv2 from colossal AI
Describe a TODO feature
There are two version of Zero support from ZeroDDP and ShardedModelv2
- check the possibility to merge two into one
- Otherwise, it has a function to choose one of them based on the flag (not shown to users directly)
Assignees
- Dongsung and Hyen
https://github.com/hpcaitech/ColossalAI/blob/6e51d296f07c0ad34d7f85cf9a70d4ceee15ede7/colossalai/zero/sharded_optim/low_level_optim.py#L29
There is updates about Zero-DP in Colossal AI. Zero 1 -> Low_level_optimizer Zero 2 -> Low_level_optimizer (partitioned gradient =True) Zero 3 -> Sharded optimizer v2, sharded modelV2