PyPI Stats

Search

All packages
Top packages

Track packages

holdout-evals


PyPI page
Home page
Author: Jordan Baillie
License: MIT
Summary: An independent significance referee for LLM & agent evals — is your improvement real, or noise?
Latest version: 0.1.0
Required dependencies: numpy
Optional dependencies: pytest

Downloads last day: 2
Downloads last week: 6
Downloads last month: 11
7-day smoothing