numqi 0.3.0

a numpy-based quantum information toolbox

numqi: a numpy-based quantum information package

WARNING: no backward compatibility guarantee until version 1.0.0

keyword: quantum information, numpy, pytorch, manifold optimization, trivialization

Quickstart

more detailed installation instructions see docs/installation

pip install numqi

A simple example to detect whether Bell state wiki is entangle or not using positive partial transpose (PPT) criteria.

import numqi
bell_state = numqi.state.Werner(d=2, alpha=1)
print(bell_state) #a numpy array
# [[ 0.   0.   0.   0. ]
#  [ 0.   0.5 -0.5  0. ]
#  [ 0.  -0.5  0.5  0. ]
#  [ 0.   0.   0.   0. ]]
print(numqi.entangle.is_ppt(bell_state, (2,2))) #True if seperable, False if entangled
# False

Module structure

mindmap
   root((numqi))
      core
         manifold
         random
         optimize
         state
         gate
            pauli
         sim
            QCircuit simulator
            Clifford simulator
         dicke
         gellmann
         channel
         matrix space
            geometric measure
            hierarchical method
         group
      application
         entangle
            PureB ext
            irrep symext
            PPT
            CHA
         qec
            VarQEC
         maximum entropy
         unique determine
            UDA
            UDP
         query
            VarQQA
         optimal control
            GRAPE

core modules

1. numqi.sim: numpy based quantum simulator wiki/quantum-simulator, support gradient back-propagation and variational quantum circuits, Clifford circuit simulator
2. numqi.dicke: Dicke basis, bosonic states
3. numqi.random: sample random quantum states, matrices, channels, etc.
4. numqi.state: various quantum states, like Werner state, Bell state, etc. and their properties like entanglement measure, etc.
5. numqi.gate: various quantum gate wiki/quantum-logic-gate, like Pauli gate (various representation), Clifford gate, etc.
6. numqi.gellmann: convert matrix into gellmann basis and reversally wiki/gellmann wiki/generalized-gellmann
7. numqi.channel: utilities related to quantum channel, like conversion between different representations of a quantum channel (Kraus operator, super-operator, Choi state, etc.) wiki/quantum-channel
8. numqi.manifold: trivialzation of various smooth manifolds and quantum objects
   • smooth manifold: sphere, positive semi-definite matrix, special orthogonal / unitary group, Stiefel manifold etc.
   • quantum object: denotes quantum state, quantum gate, Hamiltonian, quantum channel etc.
9. numqi.group: some basic utility for group theory wiki/group, like irreducible representation, symmetric group, Symplectic group over finite field Sp(2n, F2), etc.
10. numqi.optimize: wrapper of scipy.optimize and SGD/Adam optimizers for manifold optimization
11. numqi.matrix_space: Numerical range, linear space of matrices, geometric measure, hierarchical method, etc.
    • Quantifying Subspace Entanglement with Geometric Measures arxiv-link
12. numqi.utils: some utility functions, like, calculate fidelity, purity, entropy etc.

application modules

1. numqi.entangle: detect entanglement wiki/quantum-entanglement
   • PPT (positive partial transpose)
   • convex hull approximation
   • Pure Bosonic extension arxiv-link
   • irreducible representation of symmetric extension doi-link
   • unextendible product state, bound extangled state, etc.
2. numqi.maximum_entropy: relation between Hamiltonian, groud state, and marginal problem
   • Maximum entropy methods for quantum state compatibility problems arxiv-link
3. numqi.qec: quantum error correction code (QECC) wiki
   • Variational learning for quantum error-correcting codes doi-link
   • circuits to encode some well-known QECC, like ((5,2,3)), ((8,8,3))
4. numqi.unique_determine: unique determinedness in quantum state tomography
   • Variational Approach to Unique Determinedness in Pure-state Tomography arxiv-link
5. numqi.query: quantum query
   • Variational learning algorithms for quantum query complexity arxiv-link
6. numqi.optimal_control: optimal control for quantum system, e.g. finding the optimal control pulse to implement a quantum gate
   • Gradient ascent pulse engineering algorithm (GRAPE) (gradient ascent pulse engineering)

Folder structure

1. python/numqi/: source code
2. docs/: kinds of markdown files for generating website
3. tests/: unit tests
4. examples/: single file scripts to demonstrate how to use numqi
5. project/: some projects based on numqi, whether finished or not
6. misc
   • README.md: this file
   • pyproject.toml: to make a pip-installable package
   • mkdocs.yml: to generate website
   • LICENSE
   • .gitignore

How to contribute

see docs/installation "Guide-for-contributors" section

mermaid diagram

Since mkdocs-material not supported the mermaid-10.x mkdocs-material/issue, while github web support it, let's put all these mermaid diagram here and make a screenshot on the mermaid-live-editor

mmdc -i numqi-module.mmd -o numqi-module.png

module structure

flowchart TD
   id1(density matrix)
   id2(((SEP)))
   id3(((ENT)))
   id4(PPT?)
   id5(CHA?)
   id6(bosonic extension SDP?)
   id7(larger k PureB-ext?)
   id8{{dotted line: not guaranteed, probably}}
   id1 --> id4
   id4 -->|no| id3
   id4 -->|yes| id5
   id5 -.->|large loss| id3
   id5 -->|large loss| id6
   id6 -->|no| id3
   id6 -->|yes| id7
   id7 -.->|zero loss| id2
   id7 -.->|large loss| id3
   id5 -->|zero loss| id2

Recommended courses

1. point-set topology
   • youtube-link Marius Furter
2. smooth manifold
   • youtube-link The Bright Side of Mathematics
   • youtube-link qncubed3
3. Riemannian manifold
4. Differential geometry
5. algebraic topology
   • youtube-link Presented by Dr. Anthony Bosman.