AlexandrSinitsyn / ad-visualization

Automatic differentiation visualization

Home Page:https://alexandrsinitsyn.github.io/ad-visualization/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Automatic differentiation visualization

icon

How to use

Injection

You can use it both as a standalone application and as a part of your server. This is a one-page web app using only HTML5 & JS.

  • To be standalone app, you can use ./server-runner.py,that provides a simple python server. It set the application to localhost on port 9000. This server is just a proxy to your local files (you can not use the app without any server, because browsers restrict local file dependency → it means that you can not import one js file into another if they are not a server response)
  • While as a single page to your existing server it can be added without any additional libraries or packages (npm, etc...). Everything is already imported by index.html itself as <script src="..."></script>

Compilation

  • This application is written on TypeScript, but browsers allows only JS, so the whole code need to be compiled to JS (there is already compiled code in ./static/js)
  • The same thing is with SCSS which should be compiled to CSS before using
  • Also, you should run to compile nearley.js grammar to TS code (and then to JS)

    nearleyc ./parser/grammar.ne -o ./parser/grammar.ts

Requirements

  • Node.js
  • Nearley.js

Authors

  • @AlexandrSinitsyn
  • @Rabysh239

About

Automatic differentiation visualization

https://alexandrsinitsyn.github.io/ad-visualization/

License:MIT License


Languages

Language:TypeScript 43.6%Language:JavaScript 40.9%Language:CSS 5.8%Language:HTML 5.5%Language:SCSS 2.3%Language:Nearley 1.6%Language:Python 0.4%