Sign in

DIANNA

Deep Insight And Neural Network Analysis, DIANNA is the only Explainable AI, XAI library for scientists supporting Open Neural Network Exchange, ONNX - the de facto standard models format.

11
mentions
8
contributors

Cite this software

DOI:

10.5281/zenodo.5592606

What DIANNA can do for you

  • Provides an easy-to-use interface for non (X)AI experts
  • Implements well-known XAI methods (LIME, RISE and Kernal SHAP) chosen by systematic and objective evaluation criteria
  • Supports the de-facto standard of neural network models - ONNX
  • Supports both images and text data modalities, time series and tabular data are to be added
  • Comes with simple intuitive image and text benchmarks
  • Easily extendable to other XAI methods

Modern scientific challenges are often tackled with (Deep) Neural Networks (DNN). Despite their high predictive accuracy, DNNs lack inherent explainability. Many scientists do not harvest DNNs power because of lack of trust and understanding of their working. Meanwhile, the eXplainable AI (XAI) research offers some post-hoc (after training) interpretability methods that provide insight into the DNN reasoning by quantifying the relevance of individual features (image pixels, words in text, etc.) with respect to the prediction. These relevance heatmaps indicate how the network has reached its decision directly in the input modality (images, text, speech etc.) of the scientific data. Representing visually the captured knowledge by the AI system can become a source of scientific insights. There are many Open Source Software (OSS) implementations of these methods, alas, supporting a single DNN format, while standards like Open Neural Network eXchange (ONNX) exist. The libraries are known mostly by the AI experts. For the adoption by the wide scientific community, understanding of the XAI methods, and well-documented and standardized OSS are needed. The DIANNA library supports the best XAI methods in the context of scientific usage providing their OSS implementation based on the ONNX standard and demonstrations on benchmark datasets. DIANNA supports images and text data, while time-series and embedding support is currently being worked on.

Logo of DIANNA
Keywords
Programming language
  • Jupyter Notebook 77%
  • Python 20%
  • TeX 3%
License
  • Apache-2.0
</>Source code

Participating organisations

Netherlands eScience Center
SURF

Mentions

How to find your Artificial Intelligence explainer

Author(s): Elena Ranguelova
Published in 2022

Contributors

Elena Ranguelova
Elena Ranguelova
Project lead
Netherlands eScience Center
Christiaan Meijer
Christiaan Meijer
Scrum master, developer
Netherlands eScience Center
LO
Leon Oostrum
developer
Netherlands eScience Center
YL
Yang Liu
developer
Netherlands eScience Center
LO
Laura Ootes
develoepr dashboard
Netherlands eScience Center
PC
Pranav Chandramouli
developer
Netherlands eScience Center
Patrick Bos
Patrick Bos
developer, advisor
Netherlands eScience Center
GC
Giulia Crocioni
develoepr dashboard
Netherlands eScience Center

Related projects

DIANNA - Deep Insight and Neural Networks Analysis

Explainable AI tool for scientists

Updated 3 weeks ago
Running

Understanding visually grounded spoken language via multi-tasking

An alternative approach for intelligent systems to understand human speech

Updated 2 months ago
Finished

eScience Technology to Boost Quantum Dot Energy Conversion

More efficient lighting and solar energy conversion devices

Updated 2 months ago
Finished