qiskit-rng 0.1.0

Qiskit Random Number Generator.

Qiskit Random Number Generation

Qiskit is an open-source framework for working with noisy intermediate-scale quantum computers (NISQ) at the level of pulses, circuits, and algorithms.

This project contains support for Random Number Generation using Qiskit and IBM Quantum Experience backends. The resulting raw numbers can then be passed to Cambridge Quantum Computing (CQC) randomness extractors to get higher-quality random numbers.

Installation

You can install the project using pip:

pip install qiskit_rng

PIP will handle all python dependencies automatically, and you will always install the latest (and well-tested) version.

Usage

Setting up the IBM Quantum Provider

You will need setup your IBM Quantum Experience account and provider in order to access IBM Quantum backends. See qiskit-ibmq-provider for more details.

Generating random numbers using an IBM Quantum backend

To generate random numbers using an IBM Quantum backend:

from qiskit import IBMQ
from qiskit_rng import Generator

IBMQ.load_account()
rng_provider = IBMQ.get_provider(hub='MY_HUB', group='MY_GROUP', project='MY_PROJECT')
backend = rng_provider.backends.ibmq_ourence

generator = Generator(backend=backend)
output = generator.sample(num_raw_bits=int(1024), fast_path=True).block_until_ready()
print(output.mermin_correlator)

The output you get back contains useful information such as the Weak Source of Randomness (result.wsr) used to generate the circuits, the resulting bits (result.raw_bits), and the Mermin correlator value (result.mermin_correlator).

Using CQC extractors to get highly random output

If you have access to the CQC extractors, you can feed the outputs from the previous step to obtain higher quality random numbers:

print(rng_provider.random.services())  # Show a list of random services you have access to.
extractor = rng_provider.random.cqc_extractor
extractor_params = result.get_cqc_extractor_params()
random_bits = extractor.run(*extractor_params)

The code above uses the default parameter values, but the extractor is highly configurable. See documentation for some use case examples and parameter suggestions.

License

Apache License 2.0.