uncbiag/LTS
Local Temperature Scaling for Probability Calibration
When performing automated image analysis, especially in medical imaging or autonomous driving, it's crucial that your segmentation model's confidence scores are accurate for every part of the image. This tool improves the reliability of those confidence scores, particularly in specific local regions. You feed in the raw confidence predictions from your segmentation model, and it outputs a refined, more trustworthy set of probabilities, helping scientists, medical professionals, or robotics engineers ensure their automated systems make safer, more accurate decisions.
No commits in the last 6 months.
Use this if you need highly accurate, localized confidence scores from your image segmentation models, and global calibration methods aren't sufficient for complex or safety-critical applications.
Not ideal if your application only requires overall segmentation accuracy and global probability calibration is sufficient for your needs.
Stars
22
Forks
7
Language
Python
License
Apache-2.0
Category
Last pushed
Nov 26, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/uncbiag/LTS"
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