Implemented the possibility to load predictions from details files and continue evaluating from there #488
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
…d continue evaluating from there.
|
This is very nice, thanks! We'll still have to add a saving step at the end of an evaluation type. For example, saving at the end of loglikelihood evals, or generative ones. (We should be able to safely assume that once the first item of a given type's batch runs, the rest will run fine too since they are sorted by inverse length) |
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
|
Sorry, I am not sure I fully follow. Please let me know what changes are necessary. |
This was referenced Jan 11, 2025
NathanHB
approved these changes
Jan 29, 2025
hynky1999
pushed a commit
that referenced
this pull request
May 22, 2025
…d continue evaluating from there (#488) * Implemented the possibility to load predictions from details files and continue evaluating from there. * Run model as fallback when no details can be loaded. * Improved loading speed and added more useful error messages. * Fixed typo. * Fixed gnarly bug with details loading to prevent loading too many examples. * Unpacking predictions to fix issue with weirdly saved predictions. * Made bulk loading easier by also allowing first timestamp more generally. * Made loading details more robust against tensors being saved in the details files.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Fixes #467.