Skip to content

PML-168 paper reproduction limitations of amplitude encoding on quantum classification#69

Open
LF-Vigneux wants to merge 39 commits intomerlinquantum:mainfrom
LF-Vigneux:PML-168-Paper-reproduction-Limitations-of-Amplitude-Encoding-on-Quantum-Classification
Open

PML-168 paper reproduction limitations of amplitude encoding on quantum classification#69
LF-Vigneux wants to merge 39 commits intomerlinquantum:mainfrom
LF-Vigneux:PML-168-Paper-reproduction-Limitations-of-Amplitude-Encoding-on-Quantum-Classification

Conversation

@LF-Vigneux
Copy link
Contributor

Reproduction of the paper https://arxiv.org/pdf/2503.01545

Copy link
Contributor

@CassNot CassNot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nice work !
2 things to change:

  • small one: remove the "TODO" from the readme
  • bigger one: change the defaults.json so it is applied to one of the first datasets and is faster

python implementation.py --paper AA_study --config configs/defaults.json --batch_size 50
```

## Project structure --> TODO
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think the TODO part is not needed here

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I am having a timeout on tests,

  • I think the number of epochs could be reduced to 1 (the goal is just to have a config for an experiment that runs in less than 120 seconds, the results do not need to be super great)
  • Otherwise, as most of the time required to run this experiment is used to download BAS (that is a bit dangerous because directly correlated with the quality of the internet connexion), maybe the defaults.json could be on one of the first datasets (trace distance + classification or only classification)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants