This is the repository used to generate and host the following paper accepted to 2nd Workshop on Visualization for AI Explainability (VisxAI):
Selecting the right tool for the job: understanding drawbacks and biases in visualization methods
Daniel B Burkhardt*, Scott A Gigante*, Smita Krishnaswamy.
Yale University, New Haven, CT.
Click here to view the interactive notebook on Binder
If you have any questions or want to get in touch, feel free to add an Issue or send us an email by clicking on one of the author names above!
This repository is broken into several components.
Most of the interactive components of the notebook are powered by the blog_tools package. In this package are several modules for generating data, convenience wrappers for running each embedding tool, and interactive IPython and Plotly widgets that facilitate exploring the Selecting_the_right_tool_for_the_job.ipynb notebook. The setup.py and requirements.txt files in the root of this repository are for the blog_tools package. We wrote these tools specifically for this workshop paper, so you may not find them immediately applicable to a new project. However, we hope they serve as an example of how Plotly and Jupyter can be used to create an interactive explorable notebook.
This directory holds some data used in generating the figures in our article, and when you run some of the helper notebooks, data will be downloaded into this folder.
Source directories for images and markdown text that appears in the article respectively.
This folder contain Jupyter notebooks that were used to generate the figures used in the final article. The notebooks names refer to the dataset that is analyzed within that notebook. More details about the methods can be found within the notebooks.
This folder holds Python scripts that were used to generate the images for the parameter sensitivity analysis.
This is the notebook that is opened in the binder link above and contains all of the workshop paper accepted to VisxAI.