Skip to content

Latest commit

 

History

History
30 lines (27 loc) · 3 KB

README.md

File metadata and controls

30 lines (27 loc) · 3 KB
Raw

OLSQ-DPQA Compiler

Optimal Layout Synthesizer of Quantum Circuits for Dynamically Field-Programmable Qubits Array. Open source under the BSD 3-Clause license.

Repo structure:

  • run.py is an example of using the compiler. Refer to python run.py -h for options.
  • solve.py contains the class DPQA where we encode the compilation problem to SMT, and use z3-solver to solve it.
  • graphs.json contains some random 3-regular graphs.
  • animation.py contains the class CodeGen that generates DPQA instructions (five types Init, Rydberg, Activate, Deactivate, and Move), and the class Animator that generates animations from DPQA instructions. Refer to python animation.py -h for options.
  • results/ is the default directory for the results.
    • results/smt/ contains the output of SMT variable assignments.
    • results/code/ contains the code files generated from SMT output.
    • results/animations/ contains a few example animations generated from code files.

How to use the compiler:

  • We used a Python 3 environment with z3-solver, networkx, and python-sat, and matplotlib. The Python scripts are run in the root directory of the repo.
  • Run python run.py <S> <I> where <S> is the size of the random 3-regular graph, <I> is the id of the graph. To try other graphs, please edit run.py as needed.
  • The specific runtimes can differ because of the hardware, environment, and updates of this repo. Please refer to branch(es) of this repo for specific versions corresponding to the paper(s), e.g., arxiv2306.03487
  • (Optional) To generate animation, run python animation.py <F> where <F> is the SMT output file, e.g., results/smt/rand3reg_90_4.json.

Explaination of run.py:

  • The main class is named DPQA which is in solve.py.
    • When we initialize it, there is a mandatory argument name which is used for saving output file (a JSON containing SMT variables).
    • Optionally, you can specify the directory for this file with argument dir.
    • There is another optional argument print_detail to specify the granularity of printout.
  • We need to specify the architecture with setArchitecture method of DPQA. It takes in a list of 4 numbers, which are the number of columns of interaction sites, the number of rows of interaction sites, the number of AOD columns, and the number of AOD rows.
  • We need to specify the two-qubit gates in a list with setProgram method of DPQA. For example, a circuit CZ(0,1), CZ(1,2), CZ(0,1) will be [[0,1], [1,2], [0,1]].
  • If all the gates are commutable with each other (e.g., the two-qubit gates in an iteration of QAOA), call the setCommutation method of DPQA. Otherwise, do not call it and the compiler will process the gates considering dependency.
  • We can set the ratio of switching from interative peeling to optimal (multi-stage) solving with setOptimalRatio method of DPQA. By default, the ratio is 0.
  • Finally, we can solve the formulated SMT problem with the solve method of DPQA.