Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Fix Mixing hparams and arguments in LightningModule #1505

Merged
merged 9 commits into from
Apr 19, 2020
Merged

Fix Mixing hparams and arguments in LightningModule #1505

Changes from 4 commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
181b053
Attempt to fix #1468
HenryJia Apr 16, 2020
8603508
Remove the if statement, it doesn't actually make any difference
HenryJia Apr 16, 2020
7ffa62d
Update docs
HenryJia Apr 16, 2020
d16a5e2
Correct warnings I caused in the last commit
HenryJia Apr 16, 2020
c01158f
Add to changelog
HenryJia Apr 17, 2020
52d2eec
Actually add to changelog
HenryJia Apr 17, 2020
1528fa0
Resolve merge conflicts
HenryJia Apr 17, 2020
298911b
Clarify documentation and examples
HenryJia Apr 18, 2020
9b1a1f8
Update CHANGELOG.md
Borda Apr 19, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 3 additions & 4 deletions pytorch_lightning/core/lightning.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1427,6 +1427,8 @@ def load_from_checkpoint(
it stores the hyperparameters in the checkpoint if you initialized your :class:`LightningModule`
with an argument called ``hparams`` which is a :class:`~argparse.Namespace`
(output of :meth:`~argparse.ArgumentParser.parse_args` when parsing command line arguments).
Any additional arguments that are not included in hparams can be passed to the model through
\*args and \*\*kwargs.

Example:
.. code-block:: python
Expand Down Expand Up @@ -1537,10 +1539,7 @@ def _load_model_state(cls, checkpoint: Dict[str, Any], *args, **kwargs) -> 'Ligh

# load the state_dict on the model automatically
model_args = [hparams] if hparams else []
if len(model_args) > 0:
model = cls(*model_args)
else:
model = cls(*args, **kwargs)
model = cls(*model_args, *args, **kwargs)
model.load_state_dict(checkpoint['state_dict'])

# give model a chance to load something
Expand Down