fix(trainer): support continuous reward scores in bucket aggregation

Browse Source

int() truncates any float in [0,1) to 0. Replace with float().
Also fix falsy float check in failure detection.
Backward compatible with binary hard=0/1.

...

This commit is contained in:

zq

2026-05-29 19:03:52 +08:00

parent 75b5c7f31c

commit afb552008b

1 changed files with 1912 additions and 1912 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

skillopt/engine/trainer.py

+1912 -1912

File diff suppressed because it is too large Load Diff