Fix the finetuning script's loss and metric config #176
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
Contributor
chenmoneygithub
commented
May 10, 2022
chenmoneygithub
commented
- Model output should be probability instead of logits.
- Metric should be SparseCategoricalAccuracy instead of accuracy.
mattdangerw
approved these changes
May 10, 2022
examples/bert/run_glue_finetuning.py
Outdated
| num_classes, | ||
| name="logits", | ||
| name="probability", | ||
| activation="softmax", |
There was a problem hiding this comment.
Given that the pre training script is doing logit outputs, I think we might want to stay consistent.
https://github.com/keras-team/keras-nlp/blob/master/examples/bert/run_pretraining.py#L119
So stick with logits here, and pass from_logits=True to the loss below.
chenmoneygithub
force-pushed
the
fix-finetune
branch
from
643115b to
2bfef5d
Compare
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Outdated suggestions cannot be applied.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.