/DNN_dispersive_readout

Primary LanguageJupyter NotebookMIT LicenseMIT

Notes for running the scripts

About

This code is the base code of our work.

You, Z., Duan, J., Huang, W., Zhang, L., Liu, S., Zhong, Y. and Ian, H., 2023. Neural network based time-resolved state tomography of superconducting qubits. arXiv preprint arXiv:2312.07958.

Folder structure for data converting

All raw data, converted data, and train/test result are stored inside Data/data_from_some_experiment/.

For raw matlab mat. data, put under Data/data_from_some_experiment/ and use mat2txt.m to convert the mat data to txt., which will be separate in different folder.

The directory of exported data should be in their individual folder respective.

-Data
    - data_from_some_experiment
        - train
            - ground
            - excited
        - test
            - ground
            - excited
            - halfpix
            - halfpiy
            - halfpixy
            - halfpiyx

Then, use text_convert.py, which can automatically convert the data inside all these directory to the correct form. Look up the example in current path setting.

Python requirements

We suggest to use Anaconda to manage your virtual environment on your pc.

  • Python version
Python version >= 3.10
  • Requirement package

Use requirement.txt to install all require pipy package

pip install -r requirement.txt