ML in JavaScript

Learning machine learning in Python might be reasonable if you are first in machine learning. Python's eco-environment is plentiful, and the community is widespread.

Building machine learning models in JavaScript is beneficial, especially in mono-language application systems like React.js and Node.js run in JavaScript; no need to make another backend server in Python.

"TensorFlow.js" is a popular library for machine learning in JavaScript, able to use ML directly in the browser or Node.js.

Step 1 - Set Up a Node

Initialize the Project

$ npm init -y

Configure the TypeScript Compiler

npm install --save-dev typescript

Install ambient Node.js types for TypeScript

npm install @types/node --save-dev

Install the pure JavaScript version

npm install @tensorflow/tfjs

Create a tsconfig.json file

touch tsconfig.json

Add following JSON to tsconfig.json

{
  "compilerOptions": {
    "module": "commonjs",
    "esModuleInterop": true,
    "target": "es5",
    "moduleResolution": "node",
    "resolveJsonModule": true,
    "sourceMap": true,
    "rootDir": "src",
    "outDir": "dist",
    "allowJs": true,
    "strict": true,
    "noImplicitAny": true,
    "skipLibCheck": true,
  },
  "lib": ["es6"]
}

Create a Minimal TypeScript Express Server

npm install --save express
npm install -save-dev @types/express

Install Environment Variables

npm install dotenv --save

Create a file called .env

NODE_ENV = development
PORT = 3000

create an src folder in the root directory

mkdir src

create a TypeScript file named index.ts

touch src/index.ts

Open up the index.ts file and paste in the following code snippet

import express from 'express';
const app = express();

if (process.env.NODE_ENV !== 'production') {
  require('dotenv').config();
}

const port = parseInt(process.env.PORT) || 3000;

app.get('/', (req, res) => {
  res.send('Hello World!');
});

app.listen(port, () => {
  return console.log(`Express is listening at http://localhost:${port}`);
});

Install nodemon

npm install --save-dev ts-node nodemon

Add a nodemon.json config

{
  "watch": ["src"],
  "ext": ".ts,.js",
  "ignore": [],
  "exec": "ts-node ./src/index.ts"
}

Add a script to package.json

"start:dev": "nodemon",

Run the server

npm run start:dev

Install rimraf, a cross-platform tool

npm install --save-dev rimraf

Add a script to package.json

"build": "rimraf ./build && tsc",
"start": "npm run build && node build/index.js"

Install ESLint

npm install --save-dev eslint @typescript-eslint/parser @typescript-eslint/eslint-plugin

Create a .eslintrc file

touch .eslintrc

Add a JSON to .eslintrc

  "root": true,
  "parser": "@typescript-eslint/parser",
  "plugins": [
    "@typescript-eslint"
  ],
  "extends": [
    "eslint:recommended",
    "plugin:@typescript-eslint/eslint-recommended",
    "plugin:@typescript-eslint/recommended"
  ],
  "rules": { 
    "no-console": 0
  }
}

Create an .eslintignore

touch .eslintignore

Add the thing to be ignored to .eslintignore

node_modules
dist

Add a script to the package.json

"lint": "eslint . --ext .ts",
"lint:fix": "eslint . --ext .ts --fix"

Run lint

npm run lint