A language for specifying visual analysis insights, objectives and tasks.
npm install @leibatt/pyxisThis project was published as a full paper in the TVCG journal (presented at VIS 2024). Preliminary work was published as a short paper at VIS 2023. If you would like to cite this work, please use the following:
```
@article{battle2024what,
author={Battle, Leilani and Ottley, Alvitta},
journal={IEEE Transactions on Visualization and Computer Graphics},
title={What Do We Mean When We Say "Insight"? A Formal Synthesis of Existing Theory},
year={2024},
volume={30},
number={9},
pages={6075-6088},
doi={10.1109/TVCG.2023.3326698}
}
You will need to have both TypeScript and Yarn installed prior to setting up Pyxis. The following instructions assume you have already installed TypeScript and Yarn, and are currently in the Pyxis root folder.
To install all package dependencies for Pyxis run:
``
yarn
Assuming the installation was successful, to build everything and copy the datasets to the build folder run:
``
yarn build `
To build the source code only run: `
yarn build-src `
To build the examples only run: `
yarn build-examples `
To copy the datasets to the build folder run: `
yarn build-datasets
yarn test
`
All tests should pass. You may see a few warnings from eslint, but this is fine.Running the Examples
The codebase contains examples of how to specify all of the major components of Pyxis, including concepts, instances, domain knowledge nodes, data transformations, data relationships, analytic knowledge nodes, insights, and tasks.
To run any of our existing examples, you can do the following (assuming you are currently in the root folder for Pyxis):
`
node build-test/examples/[path to example].js
`Here is an example of running the Amar et al. usage scenario from the paper:
`
node build-test/examples/use_cases/amar.js
``