pyparasol 1.0b3

Web-based visualization tool for hyper-parameters based on Parasol (parallel coordinates)

Parasol

Parasol is a JavaScript library for visualization of multi-objective optimization problems such as hyper-parameter tuning and is based on parallel coordinates.

Parallel coordinates make it easy to visualize different parameter configurations

For an interactive example, check out the original authors site for the javascript library:

https://parasoljs.github.io/demo/lrgv.html

PyParasol

PyParasol is a Command line interface and API written in python that makes it easy to use Parasol without the need to write javascript code. It creates all HTML and javascript automatically and includes a HTTP server to let you view your data with ease.

Quickstart:

PyParasol requires a python version of at least 3.6. Download python at https://www.python.org/downloads/

Install pyparasol pip install --user pyparasol

Use command line interface to view a csv file

$ pyparasol data\lrgv.csv  # replace with path to your csv file(s)

 * Serving Flask app "lrgv" (lazy loading)
 * Environment: production
   WARNING: This is a development server. Do not use it in a production deployment.
   Use a production WSGI server instead.
 * Debug mode: off
 * Running on http://localhost:5000/ (Press CTRL+C to quit)

Now navigate to http://localhost:5000/

API usage example:

For an example how to use the API, have a look at the example.py script.

Acknowledgements

The original implementation PyParasol which this fork based on and Parasol itself were created by the Kasprzyk Research Group at the University of Colorado Boulder.

For more information about parasol, refer to https://github.com/ParasolJS/parasol-es