Revert calibration post-processing (measured regression on gate rounds) - #172
Merged
Conversation
Remove _apply_calibration and its four constants (_MARKET_ANCHOR_WEIGHT, _EXTREMITY_FLOOR, _EXTREMITY_CEIL, _DATASET_SHRINKAGE) from eval.py. Gate round scores dropped from 61.2/61.4 to 59.6/60.8 with calibration active, a -1.6 to -0.6 Brier Index regression. The calibration was added by gen-0 outer loop commits and has been empirically shown to hurt overall performance. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
lukeinglis
marked this pull request as ready for review
August 24, 2026 15:39
lukeinglis
added a commit
that referenced
this pull request
Aug 24, 2026
Gate round validation confirmed gen-0 calibration was a net regression (-1.6 to -0.6 BI). Calibration reverted (PR #172). Updated costs to reflect full 82-hour, ~$521 follow-up cycle. Added outcome section documenting what survived and what was reverted. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes the calibration revert task.
Changes
_apply_calibrationfunction and its four constants (_MARKET_ANCHOR_WEIGHT,_EXTREMITY_FLOOR,_EXTREMITY_CEIL,_DATASET_SHRINKAGE) fromeval.py_apply_calibration(forecasts, questions)call fromrun_eval.get()guard (PR Fix KeyError crash when non-eval JSON files exist in results/ #170), and thequestion_by_idmerge fix (PR Fix multi-round forecast matching for multi-horizon dataset questions #171) are preservedVerification