aai-institute/kyle
A library for calibrating classifiers and computing calibration metrics
When you're using a predictive model that outputs probabilities (like "there's an 80% chance of rain"), Kyle helps you check if those probabilities are actually trustworthy. It takes your model's probability predictions and the actual outcomes, then tells you how accurate those probabilities are. This is useful for data scientists and machine learning engineers who need to ensure their models' confidence scores are reliable.
No commits in the last 6 months.
Use this if you need to evaluate and improve the trustworthiness of probability predictions from your machine learning models.
Not ideal if you are looking for tools to build the predictive models themselves, rather than analyze their probability outputs.
Stars
14
Forks
1
Language
Jupyter Notebook
License
—
Category
Last pushed
Nov 28, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/aai-institute/kyle"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
facebookincubator/MCGrad
MCGrad is a scalable and easy-to-use tool for multicalibration. It ensures your ML model...
dholzmueller/probmetrics
Post-hoc calibration methods and metrics for classification
gpleiss/temperature_scaling
A simple way to calibrate your neural network.
yfzhang114/Generalization-Causality
关于domain generalization，domain adaptation，causality，robutness，prompt，optimization，generative...
Affirm/splinator
Splinator: probabilistic calibration with regression splines