https://github.com/kahsolt/iccad-2023-vqe

Quantum Computing for Drug Discovery Challenge at ICCAD 2023
https://github.com/kahsolt/iccad-2023-vqe

contest-solution quantum-computing vqe

Last synced: 5 days ago
JSON representation

Quantum Computing for Drug Discovery Challenge at ICCAD 2023

• Host: GitHub
• URL: https://github.com/kahsolt/iccad-2023-vqe
• Owner: Kahsolt
• License: mit
• Created: 2023-09-12T03:21:34.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-10-10T12:22:54.000Z (about 1 year ago)
• Last Synced: 2024-11-09T07:47:16.933Z (2 months ago)
• Topics: contest-solution, quantum-computing, vqe
• Language: Python
• Homepage:
• Size: 1.31 MB
• Stars: 1
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# ICCAD-2023-VQE

    Quantum Computing for Drug Discovery Challenge at ICCAD 2023

----

Contest page: [https://qccontest.github.io/QC-Contest/](https://qccontest.github.io/QC-Contest/)  

### How to run?

⚪ install

- better prepare a working Linux system, since [PySCF](https://pyscf.org/) **CANNOT** run on Windows!! :(

  - it's also ok to run on Windows, but must use a pre-calculated ham file like [OHhamiltonian.txt](QC-Contest-Demo/Hamiltonian/OHhamiltonian.txt)

  - and few features are unavailable, no big matter

- create venv (optional)

  - install [Miniconda](https://docs.conda.io/projects/miniconda/en/latest/) latest

  - `conda create -n qs`

  - `conda activate qs`

- install dependencies `pip install -r requirements.txt`

  - if you have CUDA on Linux, one more cmdline: `pip install qiskit-aer-gpu`

- test qiskit installation

  - `python _tutorial_qiskit_vqe.py`

- clone the code & data base `git clone https://github.com/qccontest/QC-Contest-Demo`

- test demo code runnable

  - `python _test_examplecode.py -T 2`; this might be very **slow**, just be patient :(

  - `python _test_NoiseModel_and_SystemModel.py`

⚪ run the results for submission

- run jupyter notebook `run_submission.ipynb`

⚪ run development

⚠ for Windows where PySCF is not available, you could **only** run with a pre-calculated ham file

- `python run.py`, the ham will be computed by Qiskit-PySCF (Linux only!)

- `python run.py -H ham_file` to specify an arbitary pre-calculated ham file

  - run the contest ham: `python run.py -H txt` or `python run.py -H QC-Contest-Demo\Hamiltonian\OHhamiltonian.txt`

  - run the ChemiQ pre-computed ham: `python run.py -H playground\run_chq.ham`

    - this ham seems to be inconsist with the `Qiskit-PySCF` analyzed one, with energy-diff approximately right the value of `nuclear_repulsion_energy`

    - have no idea about know whether it is a bug or feature :(

### references

- Qiskit

  - tutorial: [https://qiskit.org/documentation/tutorials.html](https://qiskit.org/documentation/tutorials.html)

  - doc: [https://qiskit.org/documentation/apidoc/index.html](https://qiskit.org/documentation/apidoc/index.html)

  - algorithms

    - tutorial: [https://qiskit.org/ecosystem/algorithms/tutorials/index.html](https://qiskit.org/ecosystem/algorithms/tutorials/index.html)

    - code: [https://github.com/Qiskit/qiskit-tutorials/tree/master/tutorials/algorithms](https://github.com/Qiskit/qiskit-tutorials/tree/master/tutorials/algorithms)

- Qiskit ecosystem

  - index: [https://qiskit.org/ecosystem/](https://qiskit.org/ecosystem/)

  - qiskit-aer: [https://github.com/Qiskit/qiskit-aer](https://github.com/Qiskit/qiskit-aer)

  - qiskit-nature-pyscf: [https://github.com/qiskit-community/qiskit-nature-pyscf](https://github.com/qiskit-community/qiskit-nature-pyscf)

- base code repo: [https://github.com/qccontest/QC-Contest-Demo](https://github.com/qccontest/QC-Contest-Demo)  

- error mitigation: [https://qiskit.org/ecosystem/ibm-runtime/tutorials/Error-Suppression-and-Error-Mitigation.html](https://qiskit.org/ecosystem/ibm-runtime/tutorials/Error-Suppression-and-Error-Mitigation.html)

----

by Armit

2023/09/05