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

..

common.py

[zero] adapt zero for unsharded parameters (#561)

2022-03-31 18:34:11 +08:00

test_init_context.py

[polish] rename col_attr -> colo_attr (#558)

2022-03-31 12:25:45 +08:00

test_shard_model_v2.py

[zero] polish ZeroInitContext (#540)

2022-03-29 09:09:04 +08:00

test_shard_param.py

[zero] trace states of fp16/32 grad and fp32 param (#571)

2022-03-31 16:26:54 +08:00

test_sharded_optim_v2.py

[zero] non model data tracing (#545)

2022-03-29 15:45:48 +08:00

test_sharded_optim_with_sync_bn.py

[zero] polish ZeroInitContext (#540)

2022-03-29 09:09:04 +08:00

test_state_dict.py

[zero] polish ZeroInitContext (#540)

2022-03-29 09:09:04 +08:00

test_zero_engine.py

[zero] polish ZeroInitContext (#540)

2022-03-29 09:09:04 +08:00