Skip to main content

A Python framework for the variational quantum classifier

Project description

# QClassify

## Description

QClassify is a Python framework for implementing variational quantum classifiers. The goal is to provide a generally customizable way of performing classification tasks using gate-model quantum devices. The quantum devices can be either simulated by a quantum simulator or a cloud-based quantum processor accessible via Rigetti Computing's [Quantum Cloud Services](

Variational quantum classification is a paradigm of supervised quantum machine learning that has been investigated actively in the quantum computing community (See for instance [Farhi and Neven](, [Schuld et al.](, [Mitarai et al.]( and [Havlicek et al.]( The general framework adopted in the design of QClassify follows from these contributions in the literature. The workflow can be summarized in Figure 1 below.

![Flow chart](
*Figure 1: Diagram illustrating the workflow of QClassify. Each rectangle represents a data object and each oval represents a method.*

## Main Components

The main data structure describing the quantum classifier setting is in ``. The implementation allows for modular design of the following components of a quantum classifier (Figure 1):

1. **Encoder**: transforms a classical data vector into a quantum state. See ``.

+ 1.1 **Classical preprocessor**: maps an input data vector to circuit parameters. See ``.

+ 1.2 **Quantum state preparation**: applies the parametrized circuit to an all-zero input state to generate a quantum state encoding the input data. See ``.

2. **Processor**: extracts classical information from the encoded quantum state. See ``.

+ 2.1 **Quantum state transformation**: applies a parametrized circuit to the encoded quantum state to transform it into a form more amenable for information extraction by measurement and classical postprocessing. See ``.

+ 2.2 **Information extraction**: extract classical information from the output quantum state. See ``.

- 2.2.1 **Measurement**: repeatedly run the quantum circuit, perform measurements and collect measurement statistics

- 2.2.2 **Classical postprocessing**: Glean information from the measurement statistics and produce the output label of the quantum classifier.

## Installation

To install QClassify using ``pip``:

pip install qclassify

Try executing ``import qclassify`` to test the installation in your terminal.

To instead install QClassify from source, clone this repository, ``cd`` into it, and run:

git clone
cd QClassify
python -m pip install -e .

Note that the pyQuil version used requires Python 3.6 or later. For installation on a user QMI, please click [here](

## Examples

We provide a Jupyter notebook to demonstrate the utility of QClassify.

Notebook | Feature(s)
[qclassify_demo.ipynb]( | Uses a simple two-qubit circuit to learn the XOR dataset.

## Disclaimer

We note that there is a lot of room for improvement and fixes. Please feel free to submit issues and/or pull requests!

## How to cite

When using QClassify for research projects, please cite:

> Sukin Sim, Yudong Cao, Jonathan Romero, Peter D. Johnson and Alán Aspuru-Guzik.
*A framework for algorithm deployment on cloud-based quantum computers*.
[arXiv:1810.10576]( 2018.

## Authors

[Yudong Cao](,
[Sukin (Hannah) Sim]( (Harvard)

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

qclassify-0.0.1.dev0.tar.gz (12.1 kB view hashes)

Uploaded source

Built Distribution

qclassify-0.0.1.dev0-py3-none-any.whl (16.4 kB view hashes)

Uploaded py3

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Huawei Huawei PSF Sponsor Microsoft Microsoft PSF Sponsor NVIDIA NVIDIA PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page