Ziyue Jiang 8b045b3c1f [Pipeline Middleware] Reduce comm redundancy by getting accurate output (#2232) ...

* move to cpu to avoid dead lock

* get output by offsets

Co-authored-by: Ziyue Jiang <ziyue.jiang@gmail.com>

2023-01-03 13:43:57 +08:00

..

_C

[optimizer] add div_scale for optimizers (#2117)

2022-12-12 17:58:57 +08:00

amp

[builder] unified cpu_optim fused_optim inferface (#2190)

2022-12-23 20:57:41 +08:00

auto_parallel

[autoparallel] patch torch.flatten metainfo for autoparallel (#2247)

2023-01-02 15:51:03 +08:00

builder

…

cli

[cli] updated installation cheheck with more inforamtion (#2050)

2022-11-30 17:53:55 +08:00

communication

…

context

[hotfix] Fixing the bug related to ipv6 support

2022-12-27 12:42:46 +08:00

device

[device] update flatten device mesh usage (#2079)

2022-12-05 16:16:07 +08:00

engine

…

fx

[autoparallel] record parameter attribute in colotracer (#2217)

2022-12-28 19:29:08 +08:00

gemini

[example] update gpt example for larger model scale (#2211)

2022-12-28 13:54:08 +08:00

kernel

[hotfix] fix error for torch 2.0 (#2243)

2022-12-30 23:11:55 +08:00

logging

[logger] hotfix, missing _FORMAT (#2231)

2022-12-29 22:59:39 +08:00

nn

[zero] fix error for BEiT models (#2169)

2022-12-26 15:03:54 +08:00

pipeline

[Pipeline Middleware] Reduce comm redundancy by getting accurate output (#2232)

2023-01-03 13:43:57 +08:00

registry

…

tensor

[hotfix] fix error for torch 2.0 (#2243)

2022-12-30 23:11:55 +08:00

testing

[zero] test gradient accumulation (#1964)

2022-11-29 13:00:30 +08:00

trainer

[polish] remove useless file _mem_tracer_hook.py (#1963)

2022-11-16 15:55:10 +08:00

utils

[builder] unified cpu_optim fused_optim inferface (#2190)

2022-12-23 20:57:41 +08:00

zero

[example] add zero1, zero2 example in GPT examples (#2146)

2022-12-20 14:30:27 +08:00

__init__.py

[setup] supported conda-installed torch (#2048)

2022-11-30 16:45:15 +08:00

constants.py

…

core.py

…

global_variables.py

…

initialize.py

…