Skip to content

Official implementaion of WheelQNet, yet another toyish quantum binary classifier implemented in pyVQNet

License

Notifications You must be signed in to change notification settings

Kahsolt/WheelQNet

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

37 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

WheelQNet: Quantum Binary Classification via Rotation Averaging

Official implementaion of WheelQNet, yet another toyish quantum binary classifier implemented in pyVQNet

This repo contains code for the contest: 第一届量子信息技术与应用创新大赛 -- 本源量子VQNet量子机器学习大赛赛道
Contest page: https://contest.originqc.com.cn/contest/32/contest:introduction
Team Name: 做好坠机准备
Final Score: 84.6 (the 1st prize 😀)

wheelq

Model Param cnt. Train acc. Test acc.
HEA 32 78.608% 82.178%
CCQC 52 79.494% 81.188%
CCQC-q 52 80.253% 78.218%
WheelQ 32 77.342% 79.208%
kNN-q - 81.392% 87.129%

⚠ only CCQC has 1 classical parameter, and kNN-q is non-parametrical, the other models are all pure quantum parametricalized :) ℹ the proposed WheelQNet looks 花里胡哨 though, it just works!! 🎉 ℹ the proposed kNN-q looks good, but it may be our fortune 😂

Quickstart

⚪ install

  • conda create -n vq python==3.8
  • conda activare vq
  • pip install -r requirements.txt

⚪ run

  • python -m src.preprocess -f, make feature data
  • python -m src.eval -L log\<model>, get testset predictions
    • python -m src.eval for the default model (knnq)
  • run_vqnet.cmd, train on your own to reproduce the submission

⚪ development

  • pip install -r requirements_dev.txt
  • python preprocess.py -f
  • python run_sklearn.py for classical comparations
  • python run_vqnet.py -M <model> to train
    • see exmaples in run_vqnet*.cmd
  • python run_vqnet.py -L <logdir> to eval

refenrence

⚪ Q framework & method

⚪ problem & data

Citation

If you find this work useful, please give a star ⭐ and cite~ 😃

@misc{kahsolt2023,
  author = {Kahsolt},
  title  = {WheelQNet: Quantum Binary Classification via Rotation Averaging},
  howpublished = {\url{https://github.com/Kahsolt/WheelQNet}}
  month  = {December},
  year   = {2023}
}

by Armit 2023/10/27

About

Official implementaion of WheelQNet, yet another toyish quantum binary classifier implemented in pyVQNet

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published