ColossalAI/tests/test_moe
Xuanlei Zhao f71e63b0f3
[moe] support optimizer checkpoint (#5015)
* Refactor MoE Manager setup method

* unshard optim ckpt

* optim io

* update transformer version

* update requirements

* update ckpt

* update ckpt

* update ckpt

* fix engine

* fix engine
2023-11-08 15:07:03 +00:00
..
moe_utils.py [moe] merge moe into main (#4978) 2023-11-02 02:21:24 +00:00
test_grad_handler.py [moe] support optimizer checkpoint (#5015) 2023-11-08 15:07:03 +00:00
test_kernel.py [moe] support optimizer checkpoint (#5015) 2023-11-08 15:07:03 +00:00
test_moe_checkpoint.py [moe] support optimizer checkpoint (#5015) 2023-11-08 15:07:03 +00:00
test_moe_ep_tp.py [moe] support optimizer checkpoint (#5015) 2023-11-08 15:07:03 +00:00
test_moe_group.py [moe] support optimizer checkpoint (#5015) 2023-11-08 15:07:03 +00:00
test_moe_hybrid_zero.py [moe] support optimizer checkpoint (#5015) 2023-11-08 15:07:03 +00:00
test_moe_load_balance.py [moe] support optimizer checkpoint (#5015) 2023-11-08 15:07:03 +00:00
test_moe_router.py [moe] merge moe into main (#4978) 2023-11-02 02:21:24 +00:00
test_moe_zero_fwd_bwd.py [moe] support optimizer checkpoint (#5015) 2023-11-08 15:07:03 +00:00
test_moe_zero_optim.py [moe] support optimizer checkpoint (#5015) 2023-11-08 15:07:03 +00:00