You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

Jiarui Fang 05e33b2578 [zero] fix grad offload (#528) ... * [zero] fix grad offload * polish code		3 years ago
..
data_sampler	…
gradient_accumulation	…
memory_tracer	[zero] refactor model data tracing (#522)	3 years ago
memory_utils	[zero] fix grad offload (#528)	3 years ago
multi_tensor_apply	…
profiler	fixed error when no collective communication in CommProfiler	3 years ago
tensor_detector	Added tensor detector (#393)	3 years ago
__init__.py	[memory] add model data tensor moving api (#503)	3 years ago
activation_checkpoint.py	Added activation offload (#331)	3 years ago
checkpointing.py	fixed mkdir conflict and align yapf config with flake (#220)	3 years ago
common.py	[MOE] remove old MoE legacy (#493)	3 years ago
cuda.py	…
moe.py	[polish] polish singleton and global context (#500)	3 years ago
timer.py	[profiler] primary memory tracer	3 years ago