PyPI page
Home page
Author:
Saif Punjwani
Summary:
Statistically valid evaluation for LLM-judged benchmarks: bias-corrected scores, honest intervals, defensible comparisons
Latest version:
0.7.4
Required dependencies:
numpy
|
scipy
Optional dependencies:
hypothesis
|
mypy
|
numba
|
ppi-python
|
pytest
|
pytest-xdist
|
ruff
|
scipy-stubs
Downloads last day:
0
Downloads last week:
24
Downloads last month:
41
7-day smoothing