refactor: Simplify model export #73
Merged
Conversation
This file contains 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
Replaces OutputMapperLayer with a custom serving function. This means we don't need to save two models (training/serving) anymore, as the (ex training) model will do the "output mapping" through the custom serving function. With this approach, we could even have multiple serving functions with different output mapping strategies.
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
|
I'd love to get feedback on the proposed refactoring. It's not a fluid as I'd like, but my feeling is it's going in the right direction (?). I put a few FIXME in the comments where things are still a bit rough. Comments welcome! |
streino
changed the title
|
Ready for rewiew. Hopefully it's at least as readable as pre-refactoring 😅 There are inconsistencies in the run timestamps because I had to fix a memory crash in the "Test" section and didn't want to re-run the full 2h of model training. It's only cosmetic. |
streino
commented
Jun 13, 2022
AnisZakari
requested changes
Jun 29, 2022
AnisZakari
approved these changes
Jul 6, 2022
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Outdated suggestions cannot be applied.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.
Replaces OutputMapperLayer with a custom serving function.
This means we don't need to save two models (training/serving)
anymore, as the (ex training) model will do the "output mapping"
through the custom serving function.
With this approach, we could even have multiple serving functions with
different output mapping strategies.