mirror of https://github.com/hpcaitech/ColossalAI
![]() * Fix state_dict output for ZeroDDP duplicated parameters * Rewrite state_dict based on get_static_torch_model * Modify get_static_torch_model to be compatible with the lower version (ZeroDDP) |
||
---|---|---|
.. | ||
_ops | ||
layer | ||
loss | ||
lr_scheduler | ||
metric | ||
optimizer | ||
parallel | ||
__init__.py | ||
init.py |