fix: support continuous reward scores (int truncation + falsy float)#20

Merged

Yif-Yang merged 3 commits into

microsoft:mainfrom

1s1x:fix-continuous-reward-scores

May 30, 2026

Commits on May 29, 2026

fix(trainer): support continuous reward scores in bucket aggregation
1s1x
committed
fix(reflect): support continuous reward scores in failure filtering
1s1x
committed
fix(scoring): use float() instead of int() for continuous reward scores
1s1x
committed