stat-knife - A Convenient Knife for Statistical Operations

Introduction

stat-knife is a convenient knife for statistical operations. It will also provide many features supported by modern Web browsers in many development environments.

Features

Use the interactive JavaScript (ES6) console on Electron
Implement the simple panel manager for summaries or plots with D3
Save the contents of a panel in HTML/SVG/PNG
Support the simple data manipulations
Support the simple random number generators
Support the markdown rendering with markdown-it and the LaTeX rendering with KaTeX
Load the user configuration script at $HOME/.stat-knife/init.js

Showcases

iris = StatKnife.DataSet.readIris()
iris.summary().show()
iris.filter([0, 50, 100]).show()
StatKnife.attach(iris).histogram('SepalLength').groupBy('Species').show()
StatKnife.attach(iris).plot('SepalWidth', 'SepalLength').groupBy('Species').show()

dice = new StatKnife.Dice(20190216)
sampleNorm = Array(100000).fill(0.0).map((_, i) => dice.normalDist()())
sampleNorm.summary().show()
StatKnife.attach(sampleNorm).histogram('value').thresholds(100).show()

canvas = (new StatKnife.CanvasPanel()).show()
canvas.context.fillStyle = '#ff000088'; canvas.context.fillRect(50, 50, 250, 250)
canvas.context.fillStyle = '#0000ff88'; canvas.context.fillRect(200, 200, 250, 250)

StatKnife.markdown().text(String.raw`
Everything starts with complex numbers and Euler's formula: $e^{i\theta} = \cos\theta + i\sin\theta$.

We can show that the well-known relations:

~~~tex
\begin{aligned}
  1 &= e^0 = e^{i\theta}e^{-i\theta} = \cos^2\theta + \sin^2\theta \\
  \sin(\alpha + \beta) &= \mathrm{Im}(e^{i\alpha}e^{i\beta}) = \sin\alpha\cos\beta + \cos\alpha\sin\alpha
\end{aligned}
~~~

...
`).show()
StatKnife.markdown().textFromFile('./README.md').show()

Installation and Development

The latest Visual Studio Code (VSCode) is recommended for developing stat-knife. VSCode provides a compact and fast development environment, including the terminal!

At first, install node.js. Then, download stat-knife in your terminal as follows:

$ git clone https://github.com/hiro4bbh/stat-knife

Install the packages depended on by stat-knife in your terminal at ./stat-knife as follows:

$ npm install

Start stat-knife in your terminal at ./stat-knife as follows:

$ npm start

TODO

Prepare documentations.
Write unit tests.
Prepare stat-knife logo.
Implement more statistical functions.
Tweak visual styles and plot options.
Prepare release packages.