Reza Yazdani 2afa1c7f2f Communication Optimization for Large-Scale Training (#4695) | 11 月之前 | |
---|---|---|
.. | ||
__init__.py | b361c72761 Update DeepSpeed copyright license to Apache 2.0 (#3111) | 1 年之前 |
experts.py | b361c72761 Update DeepSpeed copyright license to Apache 2.0 (#3111) | 1 年之前 |
layer.py | 2afa1c7f2f Communication Optimization for Large-Scale Training (#4695) | 11 月之前 |
mappings.py | b361c72761 Update DeepSpeed copyright license to Apache 2.0 (#3111) | 1 年之前 |
sharded_moe.py | b354c28b76 polishing timers and log_dist (#3996) | 1 年之前 |
utils.py | b361c72761 Update DeepSpeed copyright license to Apache 2.0 (#3111) | 1 年之前 |