Save state on train end #1168
Merged
Save state on train end #1168
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
Owner
|
Thank you! This makes a big sense. |
nana0304
pushed a commit
to nana0304/sd-scripts
that referenced
this pull request
Jun 4, 2025
Save state on train end
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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.
Based on the current implementation, the state is saved as frequently as it is with LoRA.
However, given the significant file size of the state, even with the use of save_last_n_XX to manage and delete old states, it remains cumbersome, especially for those who need to save a large number of step versions (approximately 100-200 steps) to pick up the optimal training outcomes.
This minor modification introduces
--save_state_on_train_endallowing the preservation of only the final state at the end of training. This facilitates the continuation of training for under-trained models.It has been observed that, compared to resuming with weights, resuming from the state provides a more stable and similar descent curve to the original training process.