github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-09-10 05:20:33 +00:00
Code Issues Packages Projects Releases Wiki Activity

Support TP-compatible Torch AMP and Update trainer API (#27)

Browse Source 
* Add gradient accumulation, fix lr scheduler

* fix FP16 optimizer and adapted torch amp with tensor parallel (#18)

* fixed bugs in compatibility between torch amp and tensor parallel and performed some minor fixes

* fixed trainer

* Revert "fixed trainer"

This reverts commit 2e0b0b7699.

* improved consistency between trainer, engine and schedule (#23)

Co-authored-by: 1SAA <c2h214748@gmail.com>

Co-authored-by: 1SAA <c2h214748@gmail.com>
Co-authored-by: ver217 <lhx0217@gmail.com>
This commit is contained in:
Frank Lee
2021-11-18 19:45:06 +08:00
committed by GitHub
parent 2b05de4c64
commit 3defa32aee
80 changed files with 2194 additions and 1584 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
colossalai/utils/common.py
View File

@@ -27,7 +27,7 @@ def sync_model_param_in_dp(model):
:param model: A pyTorch nn.model on whose parameters you check the consistency
'''
if gpc.is_initialized(ParallelMode.DATA) and gpc.get_world_size(ParallelMode.DATA) > 2:
if gpc.is_initialized(ParallelMode.DATA) and gpc.get_world_size(ParallelMode.DATA) > 1:
for param in model.parameters():
ranks = gpc.get_ranks_in_group(ParallelMode.DATA)
dist.broadcast(param, src=ranks[0], group=gpc.get_group(ParallelMode.DATA))