.. |
__init__.py
|
b112c99ea8
Fix loading a universal checkpoint (#5263)
|
7 months ago |
bwc.py
|
08e0733e4a
Support MoE for pipeline models (#5338)
|
6 months ago |
comms_logging.py
|
0b507253e5
fix comm logging for inference (#4043)
|
1 year ago |
debug.py
|
40342055ce
Remove hooks on gradient accumulation on engine/optimizer destroy (#4858)
|
9 months ago |
exceptions.py
|
b361c72761
Update DeepSpeed copyright license to Apache 2.0 (#3111)
|
1 year ago |
groups.py
|
08e0733e4a
Support MoE for pipeline models (#5338)
|
6 months ago |
init_on_device.py
|
b361c72761
Update DeepSpeed copyright license to Apache 2.0 (#3111)
|
1 year ago |
logging.py
|
b361c72761
Update DeepSpeed copyright license to Apache 2.0 (#3111)
|
1 year ago |
mixed_precision_linkage.py
|
18179807f5
Remove optimizer step on initialization (#5104)
|
8 months ago |
numa.py
|
5dadf68771
support HBM in utils/numa.py (#3918)
|
1 year ago |
nvtx.py
|
b361c72761
Update DeepSpeed copyright license to Apache 2.0 (#3111)
|
1 year ago |
tensor_fragment.py
|
b112c99ea8
Fix loading a universal checkpoint (#5263)
|
7 months ago |
timer.py
|
697f945a05
Split is_synchronized_device api to multiple apis (#5026)
|
8 months ago |
torch.py
|
6dcced1d5c
Cleanup required_torch_version code and references. (#5370)
|
6 months ago |
types.py
|
0a61d5d664
Hybrid Engine Refactor and Llama Inference Support (#3425)
|
1 year ago |
z3_leaf_module.py
|
19e0dc39ba
Delay reduce-scatter for ZeRO3 leaf modules (#5008)
|
8 months ago |
zero_to_fp32.py
|
2518cc429d
Support `exclude_frozen_parameters` for `zero_to_fp32.py` script (#4979)
|
8 months ago |