Abstract
The feature interpretability code is a python module that interprets and analyzes neural networks trained on hydrodynamic simulation output in the form of numpy arrays. The code takes trained neural networks and extracts internal model states in the form of images. Additionally, tools for covariance analysis of network weights and predictions are provided. This code is built on the TensorFlow and PyTorch python libraries, and includes trained networks and example input data for demonstration purposes.
- Developers:
- Release Date:
- 2023-12-19
- Project Type:
- Open Source, Publicly Available Repository
- Software Type:
- Scientific
- Licenses:
-
BSD 3-clause "New" or "Revised" License
- Sponsoring Org.:
-
USDOE National Nuclear Security Administration (NNSA)Primary Award/Contract Number:AC52-06NA25396
- Code ID:
- 121161
- Site Accession Number:
- O4675
- Research Org.:
- Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
- Country of Origin:
- United States
Citation Formats
Callis, Skylar.
Feature Interpretability.
Computer Software.
https://github.com/lanl/feature_interpretability.
USDOE National Nuclear Security Administration (NNSA).
19 Dec. 2023.
Web.
doi:10.11578/dc.20240126.6.
Callis, Skylar.
(2023, December 19).
Feature Interpretability.
[Computer software].
https://github.com/lanl/feature_interpretability.
https://doi.org/10.11578/dc.20240126.6.
Callis, Skylar.
"Feature Interpretability." Computer software.
December 19, 2023.
https://github.com/lanl/feature_interpretability.
https://doi.org/10.11578/dc.20240126.6.
@misc{
doecode_121161,
title = {Feature Interpretability},
author = {Callis, Skylar},
abstractNote = {The feature interpretability code is a python module that interprets and analyzes neural networks trained on hydrodynamic simulation output in the form of numpy arrays. The code takes trained neural networks and extracts internal model states in the form of images. Additionally, tools for covariance analysis of network weights and predictions are provided. This code is built on the TensorFlow and PyTorch python libraries, and includes trained networks and example input data for demonstration purposes.},
doi = {10.11578/dc.20240126.6},
url = {https://doi.org/10.11578/dc.20240126.6},
howpublished = {[Computer Software] \url{https://doi.org/10.11578/dc.20240126.6}},
year = {2023},
month = {dec}
}