Skip to content
Quantum Tensors - NPM package for sparse matrix operations; Quantum Game engine
TypeScript JavaScript
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.vscode
demos
imgs
src
tests
.eslintrc.js
.gitignore
.prettierrc.js
README.md
index.js
main.ts
package-lock.json
package.json
tsconfig.json
yarn.lock

README.md

Quantum Tensors

It is a JavaScript / TypeScript package for sparse tensor operations on complex numbers. For example for quantum computing, quantum information, and well - the Quantum Game.

(1.00 +0.00i) |1,2,>,V⟩ -> (-0.71 +0.00i) |1,2,>,H⟩ + (0.71 +0.00i) |1,2,>,V⟩

Developed at the Centre of Quantum Technologies, National University of Singapore, by Piotr Migdał, Philippe Cochin et al. It is a part of the Quantum Game 2 project.

We base the philosophy of this package on:

  • Sparse operations (both for vectors and matrices)
  • Complex numbers
  • Tensor structure
  • Named tensor dimensions (vide Tensors considered harmful): there is a difference between a 2x2 operator on spin and on polarization. It helps with catching errors.

Documentation: p.migdal.pl/quantum-tensors (generated by TypeDoc).

Quantum Tensors logo

Usage

Right now, you can install this package from this GitHub repository. If you use NPM:

npm install stared/quantum-tensors#master

And if you use yarn:

yarn add stared/quantum-tensors#master

It will download it and run scripts to process TypeScript files, so it generates JavaScript. Right now, we intend to use it with TypeScript only - if you tried to use it with JavaScript, let us know!

And then in your project write:

import * as qt from 'quantum-tensors'

Once it gets more mature, I will push it to the NPM repository.

Why

For "Quantum Game 2" we needed a fast and precise way to simulate quantum stuff, so after tinkering with mathjs and TensorFlow.js we decided to code what we need from scratch. At some point, we may want to use one of these libraries for backend, if we discover that it helps.

Also, https://github.com/stared/thinking-in-tensors-writing-in-pytorch by Piotr Migdał.

Contributing

Interesting ideas

  • A better notation (e.g. ⨂(op1, opt2, op3))
  • An equation viewer
You can’t perform that action at this time.