Skip to content

A hybrid (CQ) quantum neural network with Tensorflow and Qiskit/OpenQASM. Trained on a subset of the MNIST dataset.

Notifications You must be signed in to change notification settings

x86-512/BINvolution

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

21 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

QCNN

This is a project that is written in Qiskit to make a classical-hybrid quantum convolutional neural network trained on a subset of the MNIST dataset. This project was my attempt to write a Classical-Hybrid Quantum (CQ) Convolutional Neural Network.

The accuracy, training time, and running time are really bad (as expected) due to Barren Plateaus, unoptimized parameters, and suboptimal observables for the network. I strongly recommend NOT running this on an actual quantum computer because the Decoherence will be TERRIBLE due to the circuit depth. Also this will suck up all your minutes(if you are on IBM's free plan).

How it works:

Encoding Layer: Each pixel in the image is mapped to a qubit. The qubit's phase rotated by its pixel's corresponding value in the flattened tensor. I am aware that the expressibility of this is not great, but this was my attempt at recreating the ZFeatureMap circuit from Qiskit.

More documentation coming soon

About

A hybrid (CQ) quantum neural network with Tensorflow and Qiskit/OpenQASM. Trained on a subset of the MNIST dataset.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published