dann5 3.0.1

A quantum programming integrated platform for fast learning of quantum computing concepts and development of quantum programs

Quantum Programming: dann5 version 3

The dann5 project aims to simplify the programming of quantum computing models and to demonstrate ways of faster development of Python programs, which take advantage of distributed quantum computing resources, such as IBM Qiskit and D-Wave Leap. The interactive examples in Dann5 notebooks are designed to help traditional developers to learn how to formulate and solve problems on quantum computers by using quantum programing types and operations provided by dann5 libraries.

The d5 library in dann5 package provides means of defining problems in the form of one, or in a sequence of bitwise, logical, or mathematical statements, which constitute a human easily understandable quantum program. The dann5 package provides libraries which assist developers in converting a quantum program into a designated quantum computing language, such as QUBO/BQM, in case of D-Wave’s, or Qiskit circuits, in case of IBM’s, quantum solvers or simulators. Also, dann5 libraries provide simple ways of employing local quantum simulators, or simple ways of connecting to remote quantum solvers and simulators to compute and retrieve solutions for a given problem described by the quantum program. The dann5 quantum programs can be executed on IBM’s analog quantum gate computers, or on D-Wave’s quantum annealers or hybrid computers, or on Azure quantum simulators using Azure Quantum.

To write a quantum program, a Python or C++ developer uses dann5 predefined quantum types and operations, in a same way as using the language native types and corresponding operations. A developer can specify undefined and defined variables as a quantum-bit, q-boolean, q-binary, q-whole, q-integer, q-positive-rational and q-rational types. The specified quantum variables can be linked into quantum-statements such as q-expressions and q-assignments by applying type appropriate quantum-operations to describe a problem. The quantum-statements can be executed on different quantum computing resources or further organized into a q-block, q-routine, or a q-function forming a quantum program, which describes a more complex problem.

In a way dann5 is 3rd generation quantum programming language with built-in virtual machine that interprets a quantum program in a real time and executes a specific quantum assembler interpretation on a targeted quantum computing resource. In the case of IBM, dann5 virtual code (d5vc) is converted into a Qiskit interpretation forming a bespoke quantum-circuit. The bespoke Qiskit circuit can be executed on targets such as QasmSimulator or any IBM quantum computing backend. Also, dann5 virtual code can be translated into a QUBO interpretation and as such executed using D-Wave Ocean SDK on D-Wave’s quantum simulator, annealer and hybrid computing resources, such as Advantage2, or Hybrid Solver.

You can download dann5 libraries as a Python package, or cross platform (Windows, Mac and Linux) dann5 C++ source code projects .

Quantum programming with dann5

To start learning about quantum programming and quantum computing platforms, any developer can very quickly prepare a quantum programming environment by following these seven steps:

1. Download and install Python 3.10.11 (64-bit)
2. Create Python virtual environment for quantum programming
3. Install and test dann5 package
4. Install Use D-Wave Ocean SDK
5. Set up your D-Wave Leap account
6. Install IBM Qiskit
7. Add Azure Quantum
8. Download dann5 notebooks and start emerging into quantum programming

1. Download and install Python 3.10.11 (64-bit)

Windows installation

python.exe will be installed in %userprofile%/AppData/Local/Programs/Python/Python310 folder. Set up Python environment variables:

1. Open Administrative Tools (i.e. Control Panel)
2. Using Search Control Panel box (in top right corner) search for environment and press enter
3. Click on Edit the system environment variables link
4. System Properties box will open, click on Environment variables... button (at bottom-right corner)
5. Environment variables box will open
6. If PYTHONHOME already exist in User variables for ..., use *Edit..." button to change it, otherwise use New... button
7. Set PYTHONHOME to %userprofile%/AppData/Local/Programs/Python/Python310
8. If PYTHONPATH already exist in User variables for ..., use *Edit..." button to change it, otherwise use New... button
9. Set PYTHONPATH to %PYTHONHOME%
10. If PATH already exist in User variables for ..., use *Edit..." button to change it, otherwise use New... button
11. Make sure that PATH includes *%userprofile%\AppData\Local\Programs\Python\Python310\Scripts* and %userprofile%\AppData\Local\Programs\Python\Python310* paths

python --version

Python 3.10.11

MAC and Linux installation

python3 will be installed in /usr/bin folder, which is on the path.

python3 --version

Python 3.10.11

Now python is ready to set up a new virtual environment. To verify all is ready, open Command Prompt window and run

2. Create Python virtual environment for quantum programming (VE4QP)

Create a virtual environment for your Quantum work. To create d5 VE4QP):

1. Open Command Prompt window

2. Go to folder where d5 virtual environment folder should be created

   • e.g. in case of windows you can create Envs folder in %userprofile%/AppData/Local folder and run the following command

   cd %userprofile%/AppData/Local/Envs

   • in case of mac or linux Envs folder can be created in user's 'HOME' folder. Make shure the Envs is a working directory by moving into it:

   cd ~/Envs

3. Run the following command to create a virtual environment called d5

   • in case of Windows:

   python -m venv d5

   • or in case of mac or linux:

   python3 -m venv d5

4. Activate the VE4QP by running:

   • in case of Windows:

   %userprofile%/AppData/Local/Envs/d5/Scripts/activate

   • or in case of mac and linux

   source ~/Envs/d5/Scripts/activate

5. As a result, the prompt will change to begin with (d5)

6. Upgrade pip by running:

   python -m pip install --upgrade pip

7. To be able to execute or create jupyterlab and notebook install jupyter packages by running:

   pip install --upgrade jupyterlab

   pip install --upgrade notebook

Install Spyder

To be able to write or debug python code download and install Spyder 5.5.1

• Spyder comes with selected python 3.7.9 package. To use Python 3.10.11 in your virtual environment for quantum programming (VE4QP)

  1. Install spyder-kernels and matplotlib, using Command Prompt in the active VE4QP, e.g. d5

  pip install --upgrade matplotlib

  pip install spyder-kernels==5.5.*

  • Note: make sure spyder kernel version is correct!
  2. Open Spyder from Windows Start menu
  3. Change Spyder’s default Python interpreter by click the name of the current environment (i.e. custom(Python 3.7.5)) in the status bar,
  4. then click Change default environment in Preferences..., which will open the Preferences dialog in the Python interpreter section.
  5. select the option Use the following Python interpreter,
  6. use the text box or the Select file button to enter the path to the Python interpreter in your VE4QP, e.g.:

     %userprofile%/AppData/Local/Envs/d5/Scripts/python.exe

  • The name of the current environment in the status bar should change to custom(Python 3.10.11).
  • See the IPython Console for more information.

Now your d5 virtual environment is ready for installation of quantum programming packages!

3. Quantum programming with dann5.d5

To write a simple quantum program that you can run on a quantum simulator, quantum annealer or quantum computer you should install dann5 package of libraries, in your VE4QP, e.g. active d5 virtual environment:

pip install --upgrade dann5

This will install pybind11 and dann5 packages into %userprofile%/AppData/Local/Envs/d5/Lib/site-packages, in case of Windows, or ~/Envs/d5/Lib/site-packages, in case of mac or linux.

To test your local VE4QP, you can run the following code using python from a Command Prompt, or use spyder as an IDE.

• The following code finds all possible combinations of 3 numbers that will add to the number 10, where number p is unknown q-whole number with 3 q-bits in superposition state, while q and r are two unknown q-whole numbers with 2 q-bits each, and where python variable Sum references S q-whole number with deterministic value 10.

• sumAssignmnet is a python variable which references a quantum assignment of p, q and r addition expression to the S q-whole number.

import dann5.d5 as d5

from dann5.dwave import Solver

p = d5.Qwhole(3,"p")

q = d5.Qwhole(2, "q")

r = d5.Qwhole(2, "r")

Sum = d5.Qwhole("S", 10)

sumAssignment = Sum.assign(p + q + r)

print(sumAssignment)

S\4:10\ = ((p\3:U\ + q\2:U) + r\2:U)

• The sumAssignment.solve() method uses dann5.d5o quantum annealing simulator to identify all possible solutions for p, q and r (shown in code below).

• Before solve() methd is called, we need to call Solver.Active() to activate the default dann5 solver simulating solutions.

Solver.Active()

sumAssignment.solve()

print("d5o simulation solutions: \n{}".format(sumAssignment.solutions()))

The sumAssignment.solutions() method returns line by line all found solutions of expression S = 10 = p[3] + q[2] + r[2], where each variable is presented as

• variable_name / #_of_q-bits : varaible_value /

d5 simulation solutions: S\4:10; _+0\4:13; p\3:6; q\2:2; r\2:2
S\4:10; _+0\4:13; p\3:4; q\2:3; r\2:3
S\4:10; _+0\4:13; p\3:6; q\2:1; r\2:3
S\4:10; _+0\4:13; p\3:5; q\2:2; r\2:3
S\4:10; _+0\4:13; p\3:5; q\2:3; r\2:2
S\4:10; _+0\4:13; p\3:7; q\2:0; r\2:3
S\4:10; _+0\4:13; p\3:7; q\2:1; r\2:2
S\4:10; _+0\4:15; p\3:6; q\2:3; r\2:1
S\4:10; _+0\4:15; p\3:7; q\2:2; r\2:1
S\4:10; _+0\4:15; p\3:7; q\2:3; r\2:0\

The sumAssignment.solutions() method returns line by line all found solutions of expression S = 10 = p[3 qb] + q[2 qb] + r[2 qb], where each variable is presented as

variable_name \ #_of_q-bits : varaible_value \, e.g. p\3:6\; q\2:2\; r\2:2\.

Additionally, any variable named '_< sign >#' (where # is a number) is an auxiliary variable. For example, an addition auxiliary variable is _+0 with 4 qbits and value 13.

4. To Use D-Wave Install Ocean SDK

If you would like to develop a quantum solution to be executed on D-Wave quantum annealer, hybrid-computer or simulator, you have to create a developer account in D-Wave Leap cloud and install D-Wave Ocean SDK in local QVE.

1. To create D-Wave Leap developer account you need a github account. If you don't, create one.

2. Create a developer account on D-Wave Leap.

3. Log in using your D-Wave Leap developer account.

   • Explore D-Wave Leap landing page and locate API Token, which you will need to configure D-Wave Ocean in your local QVE.
   • You can develop/debug D-Wave specialized quantum solutions in Leap, by creating your Leap IDE, under Resources.

4. Install and by running following commands using Command Prompt in your local active QVE

   pip install --upgrade D-Wave-ocean-sdk

5. configure D-Wave Ocean in your local QVE by running:

   D-Wave config create

   1. when prompted Available profiles: defaults just press enter
   2. when prompted Profile (select existing or create new) [defaults]: just press enter
   3. when prompted to enter Authentication token [skip]: past the API Token that you have copied from your D-Wave Leap landing page and press enter
   • The result should be:

     Using the simplified configuration flow.

     Try 'D-Wave config create --full' for more options.

     Updating existing configuration file: %userprofile%\AppData\Local\D-Wavesystem\D-Wave\D-Wave.conf

     Available profiles: defaults

     Profile (select existing or create new) [defaults]:

     Updating existing profile: defaults

     Authentication token [skip]: DEV-#########################

     Configuration saved.

6. Test communications with the D-Wave quantum computer by running:

   D-Wave ping --client qpu

   • If you encounter SSLError, you need to download and past certificates recognized by D-Wave endpoint into cacert.pem file located in *Lib\site-packages\certifi* in your local QVE by following these instructions. Step-by-step instructions for Windows are one third down the page. Search for Windows specific instructions to locate them.

Now your local QVE is ready for development of quantum solutions, which you can confirm by submitting a random problem to a remote solver by running following command using Command Prompt in active QVE.

D-Wave sample --random-problem

Also, you can use installed python and spyder IDEs to develop python code and test it on D-Wave simulators, quantum solvers or hybrid sampler.

5. Attach your GitHub account to your D-Wave Leap account: Leap Link

For your D-Wave license to renew every month for free you will need to pass your GitHub account to the D-Wave account profile you have created.

Use the link above to sign in and then click on the profile name in the top right corner of the D-Wave leap home page. A dropdown will appear and click on the button labeled "Expand you access".

Once the button is clicked you will be moved to a page with the D-Wave account options, click on the "Explore Developer Access" button in the Developer section and you will be moved to the Developer access page.

On that page you should have the option to input your GitHub username and a repository link, insert any repository you have, save the information and you will get the automatic renewed license for your D-Wave Leap account.

6. Add IBM Qiskit to local quantum virtual environment

To be able to use IBM's analog quantum gates computer you will need to create IBM Quantum cloud account, install Qiskit python package and set up your API key.

1. You can sign in to IBM Quantum using your github account

2. Run the following command in your local virtual environment for quantum programming (VE4QP) to install qiskit package

   pip install qiskit==0.44.3

   pip install qiskit-aer==0.13.0

   pip install qiskit-ibm-provider==0.7.0

3. After installation check the version of installed 'qiskit-terra' package is 0.25.3 by running:

   pip list

4. Instal your IBM Quantum API key

   1. Copy API token from you IBM Quantum dashboard

   2. From Command Prompt with active VE4QP (e.g. d5) run

      python

   3. In python run

      >>> from qiskit_ibm_provider import IBMProvider

      >>> IBMProvider.save_account('#########')

      >>> exit()

      • NOTE: in the code above replace ######### with the API token that you have copied

Once all is done, you can run the follwing code to execute a block of quantum code timesBlock on IBM's qiskit simulator.

The problem statement: When two quantum whole numbers with 2 q-bits each, named x and y, are equal, a multiplication expression of x times y has to be eaqual to z q-whole number.

• Note: also timesBlock.toString(True) call provides a view into a decomposed dann5 virtual quantum machine code of timesBlock.

7. Add Azure Quantum to local quantum virtual environment

To be able to solve optimization problems from your local quantum virtual environment using Azure Quantum, you'll need to perform following 3 steps:

1. Create an Azure account with an active subscription account for free.
2. Create an Azure Quantum workspace with the Microsoft QIO provider enabled.
3. Install azure-quantum python package into your local quantum environment. From command prompt with active virtual environment, e.g. d5o, run the following command:

   pip install --upgrade azure-quantum

To test connection to Azure Quantum you can dann5.azure module to create QuantumRequest for mM multiplication. In this example the QuantumRequest will use Asure Quantum ParallelTempering solver to solve the multiplication. The solver will return only one of possible five solutions.

8. Download dann5 jupyter notebooks

1. Create a folder on your local machine
2. Open a remote dann5 notebooks folder in your browser
3. Right click on a jupiter notebook you would like to download and select 'Save Link As' from the popup menu
4. Save a notebook to the local folder
5. In a comand prompt/terminal window activate your VE4QP, e.g. d5
6. Go to the local notebook folder and start a jupiter notebook application
   • In Windows the command is: start jupyter notebook
   • In Mac or Linux the comand is: (jupyter notebook)&