Generate all possible combinations, which are all the unordered samples of size k, without replacement, from a set of n objects. The number of k-combinations is equal to the binomial coefficient:
Very low memory footprint even if the number of combinations to generate is high. Thanks to generators, you can iterate over all possible samples, without creating a very large array.
$ npm install ml-combinations
// the package exports a generator function
const combinations = require('ml-combinations');
const options = { mode: 'index' };
// the generator function returns an iterator
var gen = combinations(2, 4, options);
// You can loop thr the iterator
for (let combination of gen) {
console.log(combination);
}
// Or use destructuring, if you want to manipulate the array with all possible sample combinations
console.log([...gen]); // [ [ 3, 2 ], [ 0, 2 ], [ 1, 2 ], [ 1, 2 ], [ 0, 2 ], [ 0, 1 ] ]
console.log(Array.from(gen)); // Same thing if destructuring is not supported
// Use mask mode instead of index mode (index mode is the default)
// The mask mode generates a list of arrays, each of which has a length equal to the number of object
// Each index has a 0 or 1 to indicate the absence / presence of the object
options.mode = 'mask';
gen = combinations(2, 4, options);
console.log(Array.from(gen)); // [ [ 0, 0, 1, 1 ][ 1, 0, 0, 1 ],[ 0, 1, 0, 1 ],[ 0, 1, 1, 0 ],[ 1, 0, 1, 0 ],[ 1, 1, 0, 0 ] ]Phillip J Chase, `Algorithm 382: Combinations of M out of N Objects [G6]', Communications of the Association for Computing Machinery 13:6:368 (1970). To the article