New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Upgrade lightning to 2.0 #3419
Upgrade lightning to 2.0 #3419
Conversation
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
Job PR-3419-05d0c20 is done. |
Job PR-3419-059fd91 is done. |
timeseries/setup.py
Outdated
@@ -30,7 +30,7 @@ | |||
"torch", # version range defined in `core/_setup_utils.py` | |||
"statsmodels>=0.13.0,<0.15", | |||
"gluonts>=0.13.1,<0.14", | |||
"pytorch-lightning>=1.7.4,<1.10.0", | |||
"pytorch-lightning>=2.0.0,<2.1", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Changes to timeseries
look good to me, I validated that everything works as expected locally with pytorch-lightning
v2.0.
Feel free to merge once the changes to multimodal
are approved.
@@ -5,6 +5,7 @@ | |||
import torch | |||
import torch.nn.functional as F | |||
import torchmetrics | |||
from pytorch_lightning.utilities import grad_norm |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should we at some point consider switching to the new API (as described here) or should we wait until the deprecation notice is released for the old API?
import lightning as L
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Good call. We can switch to the new namespace when it becomes more stable.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Changes to _setup_utils
look good to me
Job PR-3419-7acac86 is done. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
Job PR-3419-87c9101 is done. |
* 'master' of https://github.com/awslabs/autogluon: Upgrade lightning to 2.0 (autogluon#3419)
Issue #, if available:
Description of changes:
track_grad_norm
. Refer to Remove Trainer'strack_grad_norm
argument Lightning-AI/pytorch-lightning#16745auto
instead of None, which is required by lightning 2.0.strategy_name
attribute since some strategies like DDPStrategy no longer has it.TODOs:
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.