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

[hotfix] set return_outputs=False in examples and polish code (#5404)

Browse Source 
* fix: simplify merge_batch

* fix: use return_outputs=False to eliminate extra memory consumption

* feat: add return_outputs warning

* style: remove `return_outputs=False` as it is the default value
This commit is contained in:
Wenhao Chen
2024-03-25 12:31:09 +08:00
committed by GitHub
parent 5fcd7795cd
commit bb0a668fee
24 changed files with 28 additions and 36 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
docs/source/en/features/pipeline_parallel.md
View File

@@ -198,8 +198,7 @@ def train_epoch(epoch: int, model: nn.Module, optimizer: Optimizer, _criterion:
model,
_criterion,
optimizer,
return_loss=True,
return_outputs=True)
return_loss=True)
# Backward and optimize
if is_pp_last_stage:
loss = outputs['loss']

2
docs/source/en/features/shardformer.md
View File

@@ -271,7 +271,7 @@ However, if pipeline parallel is enabled, there are several usages different fro
3. Do forward and backward passing through calling `Booster.execute_pipeline` method:
```python
outputs = booster.execute_pipeline(
train_dataloader_iter, model, _criterion, optimizer, return_loss=True, return_outputs=True
train_dataloader_iter, model, _criterion, optimizer, return_loss=True
)
```
Backward passing has been completed by this method, so there is no need to call `loss.backward()` after executing this method.