feat(lib): Replace shuffling lib with own implementation (#174)

Author: isair

README.md

@@ -8,7 +8,7 @@
 
 A library that aims to remove the overhead of creating tensors from CSV files completely; allowing you to dive right into the fun parts of your ML project.
 
-- Lightweight.
+- [Lightweight](https://bundlephobia.com/result?p=tensorflow-load-csv).
 - Fast.
 - Flexible.
 - TypeScript compatible.
@@ -21,18 +21,21 @@ You can find the docs [here](https://barissencan.com/tensorflow-load-csv/).
 ## Installation
 
 NPM:
+
 ```sh
 npm install tensorflow-load-csv
 ```
 
 Yarn:
+
 ```sh
 yarn add tensorflow-load-csv
 ```
 
 ## Usage
 
 Simple usage:
+
 ```js
 import loadCsv from 'tensorflow-load-csv';
 
@@ -46,6 +49,7 @@ labels.print();
 ```
 
 Advanced usage:
+
 ```js
 import loadCsv from 'tensorflow-load-csv';
 

package-lock.json

@@ -71,8 +71,5 @@
   },
   "peerDependencies": {
     "@tensorflow/tfjs": "^2.0.1"
-  },
-  "dependencies": {
-    "shuffle-seed": "^1.1.6"
   }
 }

package.json

@@ -1,12 +1,12 @@
 import fs from 'fs';
 
 import * as tf from '@tensorflow/tfjs';
-import { shuffle } from 'shuffle-seed';
 
 import { CsvReadOptions, CsvTable } from './loadCsv.models';
 import filterColumns from './filterColumns';
 import splitTestData from './splitTestData';
 import applyMappings from './applyMappings';
+import shuffle from './shuffle';
 
 const defaultShuffleSeed = 'mncv9340ur';
 

src/loadCsv.ts

@@ -0,0 +1,51 @@
+const mulberry32 = (a: number) => () => {
+  let t = (a += 0x6d2b79f5);
+  t = Math.imul(t ^ (t >>> 15), t | 1);
+  t ^= t + Math.imul(t ^ (t >>> 7), t | 61);
+  return ((t ^ (t >>> 14)) >>> 0) / 4294967296;
+};
+
+const cyrb53 = (str: string, seed = 0) => {
+  let h1 = 0xdeadbeef ^ seed,
+    h2 = 0x41c6ce57 ^ seed;
+  for (let i = 0, ch; i < str.length; i++) {
+    ch = str.charCodeAt(i);
+    h1 = Math.imul(h1 ^ ch, 2654435761);
+    h2 = Math.imul(h2 ^ ch, 1597334677);
+  }
+  h1 =
+    Math.imul(h1 ^ (h1 >>> 16), 2246822507) ^
+    Math.imul(h2 ^ (h2 >>> 13), 3266489909);
+  h2 =
+    Math.imul(h2 ^ (h2 >>> 16), 2246822507) ^
+    Math.imul(h1 ^ (h1 >>> 13), 3266489909);
+  return 4294967296 * (2097151 & h2) + (h1 >>> 0);
+};
+
+function shuffle<T>(array: T[], seed: number | string = 0) {
+  if (typeof seed === 'string') {
+    seed = cyrb53(seed);
+  }
+  const random = mulberry32(seed);
+
+  const output = new Array(array.length);
+
+  for (let i = 0; i < array.length; i++) {
+    output[i] = array[i];
+  }
+
+  let m = output.length;
+
+  while (m) {
+    const i = Math.floor(random() * m--);
+
+    const t = output[m];
+    output[m] = output[i];
+    output[i] = t;
+    ++seed;
+  }
+
+  return output;
+}
+
+export default shuffle;

src/shuffle.ts

@@ -37,18 +37,18 @@ test('Shuffling should work and preserve feature - label pairs', () => {
   // @ts-ignore
   expect(features.arraySync()).toBeDeepCloseTo(
     [
-      [102, -164],
       [5, 40.34],
       [0.234, 1.47],
       [-93.2, 103.34],
+      [102, -164],
     ],
     3
   );
   expect(labels.arraySync()).toMatchObject([
-    ['Landotzka'],
     ['Landistan'],
     ['SomeCountria'],
     ['SomeOtherCountria'],
+    ['Landotzka'],
   ]);
 });
 
@@ -61,18 +61,18 @@ test('Shuffling with a custom seed should work', () => {
   // @ts-ignore
   expect(features.arraySync()).toBeDeepCloseTo(
     [
-      [5, 40.34],
+      [-93.2, 103.34],
       [102, -164],
+      [5, 40.34],
       [0.234, 1.47],
-      [-93.2, 103.34],
     ],
     3
   );
   expect(labels.arraySync()).toMatchObject([
-    ['Landistan'],
+    ['SomeOtherCountria'],
     ['Landotzka'],
+    ['Landistan'],
     ['SomeCountria'],
-    ['SomeOtherCountria'],
   ]);
 });
 

tests/loadCsv.test.ts

@@ -0,0 +1,24 @@
+import shuffle from '../src/shuffle';
+
+const data = [1, 2, 3, 4];
+
+test('Shuffling without a seed should change order', () => {
+  expect(shuffle(data)).toEqual([4, 3, 1, 2]);
+});
+
+test('Shuffling should not modify the original array', () => {
+  expect(shuffle(data)).not.toEqual(data);
+});
+
+test('Shuffling with a number seed should change order', () => {
+  expect(shuffle(data, 7)).toEqual([3, 2, 4, 1]);
+});
+
+test('Shuffling with a string seed should change order', () => {
+  expect(shuffle(data, 'hello')).toEqual([2, 4, 3, 1]);
+});
+
+test('Shuffling with different seeds should produce different results', () => {
+  const results = [shuffle(data, 7), shuffle(data, 'hello')];
+  expect(results[0]).not.toEqual(results[2]);
+});