Anchoring
- Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
This software provides methods and functions for training deep image classification models based on the principle of anchoring. It features a user-friendly PyTorch wrapper that facilitates the easy conversion of any model into an anchored model. The software supports various standard datasets and includes scripts for conducting evaluations. Developed with PyTorch, it is compatible with common neural network architectures used for image data. Additionally, it offers tools for computing evaluation metrics to assess model performance.
- Short Name / Acronym:
- anchoring
- Site Accession Number:
- LLNL-CODE-2003995
- Software Type:
- Scientific
- License(s):
- MIT License
- Research Organization:
- Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
- Sponsoring Organization:
- USDOE National Nuclear Security Administration (NNSA)Primary Award/Contract Number:AC52-07NA27344
- DOE Contract Number:
- AC52-07NA27344
- Code ID:
- 153374
- OSTI ID:
- code-153374
- Country of Origin:
- United States
Similar Records
DECIDER
Know Your Space: Inlier and Outlier Construction for Calibrating Medical OOD Detectors
Graphene U-Net v1
Software
·
Thu Sep 19 20:00:00 EDT 2024
·
OSTI ID:code-156111
Know Your Space: Inlier and Outlier Construction for Calibrating Medical OOD Detectors
Software
·
Sun Jun 04 20:00:00 EDT 2023
·
OSTI ID:code-108775
Graphene U-Net v1
Software
·
Mon Jun 07 20:00:00 EDT 2021
·
OSTI ID:code-60277