stat_learn_js

Stat.JS

Provides a Rich Models for Statistical Computing using JavaScript

How to use

This package is developed with the concern of simple plug and play experience without worrying the under the hood issues

Simple Linear Regression

var { SLR } = require('stat_learn_js');

var ht = [152, 166, 174, 179, 183, 172] // X (independent variable)
  , wt = [62, 64, 85, 72, 78, 68]; // y (dependent variable)

SLR.fit(ht, wt); // training starts

SLR.predict(166); // predict the new weight with the new height value

Multi Linear Regression

// lets predict with multiple input columns
const {MLR} = require('stat_learn_js');

var cc = [1010, 1280, 1280, 795, 1461];
var hp = [88, 94, 95, 84, 108];
var weight = [1200, 1350, 1400, 850, 1550];
var milage = [23.0, 18.5, 19.3, 21.5, 17.0];

var X = [] // (Array of Arrays)
X.push(cc);
X.push(hp);
X.push(weight);

MLR.fit(X, milage); // Training starts
MLR.predict([1000, 95, 1008]); // predicting with the new input
MLR.coeff_(); //coefficient values or beta values