Add materialisation of new modules before loading checkpoint #1030
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
The |
|
Is this ready for review? |
|
Yes |
| ) | ||
|
|
||
| model_state_dict = self.model.state_dict() | ||
| params = { |
There was a problem hiding this comment.
Can you add a comment explaining this?
There was a problem hiding this comment.
I wanna test this with a couple configs and then I will approve it
sophie-xhonneux
approved these changes
Oct 13, 2025
There was a problem hiding this comment.
This is a good fix. It is not fully complete as I think with new encoders/decoders that have the same parameter shapes as the existing model there will be issues, but this cannot be solved until embedding engines, prediction heads, etc are named according to the stream names they belong to. This issue is for a new PR.
|
Note I test with pre-training on ERA5 and continuing with ERA5, NPPATMS, & SYNOP on 2 GPUs |
sophie-xhonneux
pushed a commit
that referenced
this pull request
Oct 17, 2025
* Add materialisation of new modules before loading checkpoint * Initialize new modules in load_model * Fix adding new embedding networks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
Fixes crash when using
--forecast_finetuneflag by adding materialization of new modules before loading checkpointIssue Number
Closes #1029
Checklist before asking for review
./scripts/actions.sh lint./scripts/actions.sh unit-test./scripts/actions.sh integration-testlaunch-slurm.py --time 60