TX$^2$: Transformer eXplainability and eXploration
The Transformer eXplainability and eXploration (Martindale & Stewart, 2021), or TX2 software package, is a library designed for artificial intelligence researchers to better understand the performance of transformer models (Vaswani et al., 2017) used for sequence classification. The tool is capable of integrating with a trained transformer model and a dataset split into training and testing populations to produce an ipywidget (Project Jupyter Contributors, 2021) dashboard with a number of visualizations to understand model performance with an emphasis on explainability and interpretability. The TX2 package is primarily intended to integrate into a workflow centered around Jupyter Notebooks (Kluyver et al., 2016), and currently assumes the use of PyTorch (Paszke et al., 2019) and Hugging Face transformers library (Wolf et al., 2020). The dashboard includes visualization and data exploration features to aid researchers, including an interactive UMAP embedding graph (McInnes et al., 2018) to understand classification clusters, a word salience map that can be updated as researchers alter textual entries in near real time, a set of tools to understand word frequency and importance based on the clusters in the UMAP embedding graph, and a set of traditional confusion matrix analysis tools.
- Research Organization:
- Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC)
- Grant/Contract Number:
- AC05-00OR22725
- OSTI ID:
- 1837407
- Journal Information:
- Journal of Open Source Software, Journal Name: Journal of Open Source Software Journal Issue: 68 Vol. 6; ISSN 2475-9066
- Publisher:
- Open Source Initiative - NumFOCUSCopyright Statement
- Country of Publication:
- United States
- Language:
- English
VisBERT: Hidden-State Visualizations for Transformers
|
conference | April 2020 |
The Language Interpretability Tool: Extensible, Interactive Visualizations and Analysis for NLP Models
|
conference | January 2020 |
A Multiscale Visualization of Attention in the Transformer Model
|
conference | January 2019 |
UMAP: Uniform Manifold Approximation and Projection
|
journal | September 2018 |
Similar Records
Dimensionally Aligned Signal Projection Algorithms Library
Supporting Greater Interactivity in the IPython Visualization Ecosystem