[wip] brier score #782
Closed
[wip] brier score #782
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
bmosaicml
force-pushed
the
brier_score
branch
from
1e65cdc
to
6013a30
Compare
bmosaicml
commented
Dec 6, 2023
| cont_tok_logits = output_logits[batch_idx].index_select(dim=0, index=cont_idx - 1) | ||
| # labels have been shifted left by one index, so the cont_idx needs to be shifted as well. | ||
| cont_tok_targ = labels[batch_idx].index_select(dim=0, index=cont_idx - 1) | ||
| mean_logit_of_targ_tok = cont_tok_logits.index_select(dim=1, index=cont_tok_targ).diagonal().mean() |
There was a problem hiding this comment.
this is the part I need some help double checking. This computes the mean output logit of the target tokens, then below we compute exp(-mean_logit). On line 45 we normalize across the C choices
|
Feel free to reopen if you are still doing this |
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Outdated suggestions cannot be applied.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.
Brier score seems of questionable usefulness. COPA results:
First number for each model is Brier score. Below we find that accuracy AND brier score both go up with model size (not good because brier score is lower = better)