Submission code for the ACM/IEEE Quantum Computing for Drug Discovery Challenge at ICCAD 2023

Team name: underwater

More detailed technical reflection is provided in tech_reflect.pdf.

Credit: some of codes are modified from tensorcircuit, qiskit and also the demo provided by commitee.

Requirements

Some basic packages, qiskit, and jupyter are required.

For running the optmized circuit,

pip install stim
pip install pylatexenc
pip install qiskit-aer
pip install qiskit-ignis
pip install qiskit-ibm-provider
pip install qiskit-ibm-experiment
pip install qiskit-nature
pip install qiskit-dynamics

For training,

pip install tensorcircuit-nightly
pip install tensorflow

Instructions

The code for running optmized circuits is in run_shvqe_qiskit_ncz0.ipynb.

If you want to train SHVQE for yourself, please run

python shvqe_clifford.py <n>

where $n$ is the depth of CZ gates in Schrodinger circuit.

Saved models

We saved the optimized circuit in saved_models.

• shvqe_clifford_ncz0_sch.qasm: optimized circuit.
• shvqe_clifford_ncz0_sch_transpiled.qasm: transplied optimized circuit.

Also, we save the Heisenberg circuit as shvqe_clifford_ncz0_hei.qasm, which is required by the notebook run_shvqe_qiskit_ncz0.ipynb. Note that it is not real circuit but virtual circuit that is used in preprossing the Hamiltonian.