Artifact of AccelerQ: Accelerating Quantum Eigensolvers With Machine Learning on Quantum Simulators

This artifact contains the code and data related to AccelerQ: Accelerating Quantum Eigensolvers With Machine Learning on Quantum Simulators paper. As AccelerQ takes as input quantum programs crafted by others with data coming from other datasets, we wish to note that these are not our data nor code. The majority of it is taken from QAGC with ML Implementation at GitHub. The rest is extracted using other papers that are already cited in the publication. Our Contribution. AccelerQ is a set of Python programs for Data mining - kcl_prepare_data.py, with two wrappers for each QE implementation: kcl_QCELS_stage_1.py and kcl_adapt_vqe_stage_1.py. Training ML models - kcl_train_xgb.py, with two wrappers for each QE implementation: kcl_QCELS_stage_2.py and kcl_adapt_vqe_stage_2.py. Deploying the models - kcl_opt_xgb.py, with two wrappers for each QE implementation: kcl_QCELS_stage_3.py and kcl_adapt_vqe_stage_3.py. We then collected the new hyperparameter values and added these to the QE implementations: first_answer_experiments-tests.py and QCELS_answer_experiments-tests.py for the evaluation of AccelerQ and first_answer_experiments.py and QCELS_answer_experiments.py to compare against the baseline and AccelerQ with ML models only (weaker version).