Logan Adams 5ce448d326 Switch hasattr to check for compiler and not compile since compile was introduced in torch 2.0 but compiler was introduced in torch 2.1, this fixes issues for those building with torch 2.0 8 月之前
..
autotuning c052545122 DS #4993 #662 : autotune single node hostfile bugfix (#4996) 8 月之前
checkpoint 8998707a2f Universal Checkpoint for Sequence Parallelism (#4752) 10 月之前
comm c3cfe96bb3 Enable torch.compile with ZeRO (Experimental) (#4878) 8 月之前
compression 389bf69319 fix: Remove duplicate word the (#4051) 1 年之前
elasticity d2e9adce39 Fix error report of DSElasticAgent._set_master_addr_port() (#4985) 9 月之前
inference 1d35db76a0 Refactor the Qwen positional emebdding config code (#4955) 9 月之前
launcher 8f6277001a launcher_helper: enable fds passing (#5042) 8 月之前
model_implementations d5a7c1e0b4 Capture short kernel sequences to graph (#4318) 10 月之前
module_inject e212845e39 Add backwards compatibility w/ older versions of diffusers (<0.25.0) (#5083) 8 月之前
moe 9922270f47 Further refactor deepspeed.moe.utils + deepspeed.moe.layer type hints (#5060) 8 月之前
monitor 604d701e35 Introduce pydantic_v1 compatibility module for pydantic>=2.0.0 support (#4407) 1 年之前
nebula cd4e473ee6 fix typo with deepspeed/ (#3547) 1 年之前
ops fd0a52c1ac use all_gather_into_tensor instead of all_gather (#4705) 10 月之前
pipe b361c72761 Update DeepSpeed copyright license to Apache 2.0 (#3111) 1 年之前
profiling 61391229c9 Update flops profiler to recurse (#4374) 11 月之前
runtime 5ce448d326 Switch hasattr to check for compiler and not compile since compile was introduced in torch 2.0 but compiler was introduced in torch 2.1, this fixes issues for those building with torch 2.0 8 月之前
sequence 2afa1c7f2f Communication Optimization for Large-Scale Training (#4695) 11 月之前
utils 19e0dc39ba Delay reduce-scatter for ZeRO3 leaf modules (#5008) 8 月之前
__init__.py c3cfe96bb3 Enable torch.compile with ZeRO (Experimental) (#4878) 8 月之前
accelerator 9548d48f48 Abstract accelerator (step 2) (#2560) 1 年之前
constants.py 706a72562a Allow env var for timeout (#4405) 1 年之前
env_report.py c1ba6a104f [CANN] Support cpu offload optimizer for Ascend NPU (#4568) 11 月之前
git_version_info.py 57a27b0803 add type checker ignore to resolve that pylance can't resolved noqa annotation (#4102) 1 年之前
pydantic_v1.py 604d701e35 Introduce pydantic_v1 compatibility module for pydantic>=2.0.0 support (#4407) 1 年之前