8000 GitHub - chuanluocs/EffectiveQM
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

chuanluocs/EffectiveQM

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Towards Effective Local Search for Qubit Mapping Problem

EffectiveQM is an effective local search algorithm for solving qubit mapping problems. This repository contains the entire code implementation of EffectiveQM, the device coupling graph used in the experiment, the quantum circuits, full versions of the tables from the paper, and other data and tables related to the paper and EffectiveQM.

Instructions for Building EffectiveQM

sh build.sh	

By executing the script file, the user can build an executable of EffectiveQM in the repository root directory. Please note that this script should be run on 64-bit GNU/Linux OS. EffectiveQM is cross-platform, for other OS, please follow the Makefile in the src directory to build it by yourself.

Instructions for Running EffectiveQM

The command to run EffectiveQM once is as follows:

./EffectiveQM [RANDOM_SEED] [RESULT_FILE_PATH] [DEVICE_NAME] [DELTA] [THETA] [LAMBDA] [QASM_FILE_PATH]

DEVICE_NAME is the parameter used to specify the quantum device, and its enumerated values are listed in the following table:

value device in paper
TOKYO IBM Tokyo
Q16 Q16
GUADALUPE IBM Guadalupe
ROCHESTER IBM Rochester
torino IBM Torino
SYCAMORE Google Sycamore

Example Command for Running EffectiveQM

./EffectiveQM 1 log.txt TOKYO 5 5 500 ./benchmarks/4gt13_92.qasm

Running this command will call the EffectiveQM program, using 0 as random seed and set the hyperparameters $\delta=5$, $\theta=5$, $\lambda=500$, map ./benchmarks/4gt13_92.qasm quantum circuit on the IBM Tokyo device, and store the resulting in the ./log.txt file.

Implementation of EffectiveQM

The directory named src/ includes the implementation of EffectiveQM.

Testing Benchmarks for Evaluating EffectiveQM

The directory named benchmarks/ contains all testing benchmarks' qasm files.

Implementation of Main Competitors of EffectiveQM

As mentioned in the paper, the main competitors of EffectiveQM are ILS, SAHS, FiDLS, Qiskit, Tket, QMAP and TB-OLSQ2. Their codes can be obtained from the following link.

Among them, for the three stochastic algorithms ILS, SAHS and Qiskit, we have slightly modified the implementation of the random number generation part of them for generating reproducible experimental results.

Full Comparison for EffectiveQM and All Competitors

As we mentioned in Section V of the paper, we did not show the experimental results for EffectiveQM and SAHS due to space limits. To this end, we have placed the complete table containing the SAHS in the repository.

The directory is organized as follows:

  • full_table: Provides PDF file containing the original tables of the paper after added SAHS.

Other Experimental Results

The directory result/ contains the raw data after preliminary statistics. 73BE

The directory is organized as follows:

  • original_data: Provides the original experimental results of EffectiveQM, ILS, SAHS, FiDLS, Qiskit and Tket on all devices.
    • RevLib: the original experimental results on benchmark RevLib.
    • QV: the original experimental results of benchmark QV.
    • QUEKNO: the original experimental results of benchmark QUEKNO.
    • RW: the original experimental results of benchmark RW.
  • table: Provides collated experimental data against tables in the paper.
    • comp: Comparison data between EffectiveQM and all competitors.
    • abalation: Provides data for ablation analysis experiments.
      • abalation_for_Potential_Guided: The experimental data of the ablation analysis of Potential-guided scoring Function, that is, the comparative data of the Alt-2 version in the paper.
      • abalation_for_Mode_Aware: The experimental data of the ablation analysis of Mode Aware, that is, the comparative data of the Alt-1 version in the paper.
    • hyper-para: Experimental data for hyperparameter analysis experiments.
      • lambda: Analyzing the experimental data on the hyperparameters of $\lambda$.
      • delta: Analyzing the experimental data on the hyperparameters of $\delta$.
      • theta:Analyzing the experimental data on the hyperparameters of $\theta$.
    • discussion: Experimental data for discussion experiments in paper.

The running examples mentioned in Section IV-B

The directory example contains examples of mode-aware search strategies from Section IV-B of the manuscript.

The directory is organized as follows:

  • example.txt: Provide the complete examples mentioned in Section IV-B of the manuscript.
  • readme.md: Provide a detailed readme document to explain how to interpret the examples.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •  

Languages

0