wassname/pytorch-lightning
1
0
Fork 0
mirror of https://github.com/wassname/pytorch-lightning.git synced 2026-09-09 11:32:07 +08:00
Code Issues Packages Projects Releases Wiki Activity

Removes need to unsqueeze from dp (#1319)

Browse Source 
* removes need to unsqueeze from dp

* removes need to unsqueeze from dp

* fixed examples

* added auto unsqueeze

* added auto unsqueeze

* added auto unsqueeze

* added auto unsqueeze

* Update pytorch_lightning/overrides/data_parallel.py

Co-Authored-By: Adrian Wälchli <adrian.waelchli@students.unibe.ch>

* fixed dp parse

* fixed dp parse

Co-authored-by: Adrian Wälchli <adrian.waelchli@students.unibe.ch>
This commit is contained in:
William Falcon
2020-04-02 11:46:20 -04:00
committed by GitHub
co-authored by Adrian Wälchli
parent 6b41b5c589
commit 3cb149f4f4
6 changed files with 24 additions and 37 deletions
Download Patch File Download Diff File Expand all files Collapse all files
pl_examples/basic_examples/lightning_module_template.py
-9
View File
@@ -111,10 +111,6 @@ class LightningTemplateModel(LightningModule):
# calculate loss
loss_val = self.loss(y, y_hat)
# in DP mode (default) make sure if result is scalar, there's another dim in the beginning
if self.trainer.use_dp or self.trainer.use_ddp2:
loss_val = loss_val.unsqueeze(0)
tqdm_dict = {'train_loss': loss_val}
output = OrderedDict({
'loss': loss_val,
@@ -145,11 +141,6 @@ class LightningTemplateModel(LightningModule):
if self.on_gpu:
val_acc = val_acc.cuda(loss_val.device.index)
# in DP mode (default) make sure if result is scalar, there's another dim in the beginning
if self.trainer.use_dp or self.trainer.use_ddp2:
loss_val = loss_val.unsqueeze(0)
val_acc = val_acc.unsqueeze(0)
output = OrderedDict({
'val_loss': loss_val,
'val_acc': val_acc,