ac-learn/dist/aboutLabels.js

ML platform for All Contributors

"use strict";

const ck = require('chalk');

const labels = require('./labels.json');

const cats = require('./categories.json');

const len = labels.length;
const categorisedData = labels.filter(l => !!l.category);
const nullCategorisedData = labels.filter(l => l.category == 'null');
const validCategorisedData = labels.filter(l => cats.includes(l.category));
const badData = labels.filter(l => !cats.includes(l.category)); //source: https://gist.github.com/antoniocsoares/fa794527a0a61f5decd515f502224af7#gistcomment-2360675

const duplicates = arr => {
  const seen = new Set();
  const store = [];
  arr.filter(item => seen.size === seen.add(item).size && !store.includes(item) && store.push(item));
  return store.filter(el => el !== undefined);
};

const dups = duplicates(labels); //full

const dupLabels = duplicates(labels.map(l => l.labels)); //only the labels (not categories)

const categoriesUsed = [...new Set(categorisedData.map(x => x.category))];
const unusedCategories = cats.filter(c => !categoriesUsed.includes(c));
/* eslint-disable no-console */

console.log(ck`# of labels: {cyan ${len}} 
# of categorised labels: {cyan ${categorisedData.length}} {greenBright (${categorisedData.length / len * 100}%)}
# of labels categorised "null": {cyan ${nullCategorisedData.length}} {greenBright (${nullCategorisedData.length / len * 100}%)}
# of labels with a valid category: {cyan ${validCategorisedData.length}} {greenBright (${validCategorisedData.length / len * 100}%)}
# of labels to fix: {red ${badData.length}} {yellow (${badData.length / len * 100}%)}
# of duplicates to remove: {red ${dups.length}} {yellow (${dups.length / len * 100}%)}
# of label duplicates to remove: {red ${dupLabels.length}} {yellow (${dupLabels.length / len * 100}%)}
# of unused labels: {red ${unusedCategories.length}} {yellow (${unusedCategories.length / cats.length * 100}%)}`);
/* eslint-enable no-console */