duanjunwen d0ec221b38 [fix\ fix fail case test_shard_llama

2024-10-25 02:28:55 +00:00

..

schedule

[fix\ fix fail case test_shard_llama

2024-10-25 02:28:55 +00:00

__init__.py

[feat] add zerobubble pp (just a frame now); add POC test for dx_dw; add test for zerobubble;

2024-08-22 10:25:34 +00:00

p2p.py

fix object_to_tensor usage when torch>=2.3.0 (#5820)

2024-07-16 13:59:25 +08:00

stage_manager.py

[fix\ fix fail case test_shard_llama

2024-10-25 02:28:55 +00:00

weight_grad_store.py

[feat] support meta cache, meta_grad_send, meta_tensor_send; fix runtime too long in Recv Bwd; benchmark for llama + Hybrid(tp+pp);

2024-10-24 07:30:19 +00:00