Shypd.aiThe largest open index of AI toolstemperature_scaling is an open-source Python module for calibrating neural networks. It provides a simple way to adjust model confidence scores post-training, based on temperature scaling.
Trending
Also listed in
About
temperature_scaling is an open-source Python module designed to calibrate neural networks by adjusting their confidence scores. Originally created as a demonstration for PyTorch 0.3, it implements temperature scaling, a post-processing technique that divides logits by a learned scalar parameter to minimize negative log-likelihood on a validation set. This helps address the common issue of neural networks outputting overconfident probabilities, ensuring that confidence scores better match true correctness likelihood. While the repository is unmaintained, it offers a clear example of how to integrate temperature scaling into a project for improved model calibration.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending
