Allow backbone to be any functional, preprocessor any callable #1900
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
Consolidate saving into a saving class to match loading. Plenty more cleanup to do there probably, but this will at least bring our saving and loading routines into a common flow.
mattdangerw
commented
Oct 1, 2024
| return preprocessor | ||
|
|
||
|
|
||
| class KerasPresetSaver: |
Member
Author
There was a problem hiding this comment.
We might want to clean this up in the future, for now I'm just consolidating this to match the loading helper we have.
mattdangerw
force-pushed
the
flexible-backbone-preprocessor
branch
from
4abb29a to
e8219e1
Compare
Member
Author
|
Merging this! The last failure is unrelated, and has to do with image preprocessing on the torch backend and gpu <> cpu errors. I will fix up on the other PR. |
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Outdated suggestions cannot be applied.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.
The main change this allows is passing arbitrary functional models as a backbone to a task and arbitrary callables as preprocessing for a task. This should allow a lot more flexible API usage.
Rescalinglayer or a Pipeline as preprocessing for aImageClassifiertask.I also moved our saving routines to a
PresetSaverclass, but this is really just to be parallel with thePresetLoaderwe added for transformers/timm conversion. You can still overridesave_to_preseton a model if you need to. Probably more cleanups to do.