jev-bench
Running-Dolphins/jev-bench
Measure accuracy and calibration of Jev (TypeSafe AI's decision model) on public datasets: 12 business-like tasks, 7 experiments, one Python file.
- Stars
- 0
- Forks
- 0
Why it is listed
jev-bench is an independent accuracy/calibration benchmark on public datasets; its README emphasizes that confidence behavior differs across tasks.